System and method for testing and analysis of visual acuity and its change

Abstract

To provide a system and method for testing and analysis of visual acuity and its changes using an acuity model.SOLUTION: The acuity model is generated based on one or more acuity chart design parameters and candidate acuity parameters calculated using acuity test data of a subject, the acuity model comprising a chart-specific psychometric function determined using a family of multiple-optotype psychometric functions, and the acuity model is configurable to estimate possibility of obtaining the acuity test data of the subject.SELECTED DRAWING: Figure 15

Description

[Technical field]

[0001] <Cross reference> This application claims the benefit of U.S. Provisional Application No. 62 / 524,414, filed June 23, 2017, which is incorporated herein by reference in its entirety. [Background technology]

[0002] Optotype charts are commonly used to measure visual acuity because they are standardized, easy to use, and widely available. Testing visual acuity typically involves asking the patient to identify optotypes--visual patterns of letters or objects--presented on a test slide (or a series of slides). The slide contains one or more optotypes arranged in one or more rows at a time. The patient's ability to identify the optotypes according to size is registered and scored to yield an estimate of visual acuity. The resulting estimate of visual acuity, e.g. the size of the smallest optotype that the patient can reliably see, may depend primarily on the test and scoring design, which is based on predetermined rules and heuristics specific to each chart. Summary of the Invention

[0003] Two approaches to visual acuity testing have come to dominate the current technology landscape: chart tests, in which the patient identifies multiple rows of targets presented at one time, or computerized tests, in which targets are presented one at a time. Computerized testing has gained popularity in recent years because it offers the potential advantages of increased test resolution through flexible and accurate sampling of targets on a digital display and the application of adaptive and individualized testing strategies that focus the test on the patient's specific vision.

[0004] One important shortcoming of existing visual acuity tests is the difference in chart design between the charts used in clinical trials and clinical practice. Thus, visual acuity test results collected in clinical practice can only be inaccurately related to those collected on the "gold standard" charts used in clinical trials. Due to the long testing times required for clinical trials, this chart design and testing standard remains impractical for clinical trials by eye care professionals or general practitioners. Even if the long testing times were allowed for clinical investigations, visual acuity test results obtained with charts designed for children (e.g., HOTV, Lea designs) show systematic bias compared to visual acuity test results obtained from charts designed for adults.

[0005] Accurate visual acuity testing poses certain challenges. For many charts currently used in clinical settings, visual acuity testing only provides qualitative rather than quantitative results (e.g., 20 / 20 or 20 / 40). Furthermore, with the variety of chart designs circulating in clinical trial settings, visual acuity tests obtained with different charts can be difficult to compare and align. Without precision and harmonization among charts used for clinical visual acuity testing, it remains difficult to track how visual acuity changes with disease or treatment.

[0006] The systems and methods disclosed herein for testing and analyzing visual acuity address existing problems in visual acuity testing. Advantages of the systems and methods disclosed herein include the development of personalized visual acuity charts based on subject-specific (interchangeable with chart-invariant) information related to visual acuity. In some embodiments, another advantage of the systems and methods disclosed herein includes the ability to analyze visual acuity data collected on a wide range of visual acuity charts, including but not limited to charts from the past. In some embodiments, yet another advantage of the systems and methods disclosed herein includes the generation of accurate visual acuity parameters that are independent of the visual acuity chart design used to collect the data.

[0007] In some embodiments, a scoring algorithm (also interchangeably referred to herein as algorithm S) is disclosed herein that allows for the analysis of visual acuity data collected at will on various vision chart designs. In some cases, an adaptive algorithm (also interchangeably referred to herein as algorithm A) is disclosed herein that improves the quality of visual acuity data by generating or selecting a personalized chart that precisely focuses on the subject's visual acuity. In some cases, a combination algorithm (combining algorithms A and S in an iterative manner) is disclosed herein: evaluation of previously collected visual acuity data (from a subject or population) is used to personalize a chart design for the collection of visual acuity chart data using algorithm A. In turn, that data can be scored using algorithm S, which provides an updated estimate of the patient's visual acuity. This iterative process can provide an estimate of visual acuity with high accuracy and reduced testing time. In some embodiments, a visual acuity model is disclosed herein for analyzing visual acuity that predicts visual acuity performance during testing with various vision chart designs. In some cases, such acuity model provides the basis for algorithms S and A. In some embodiments, the visual acuity models disclosed herein use subject-specific model parameters and / or chart design-specific model parameters to provide an effective description of visual acuity chart performance that can be generalized to predict visual acuity performance on a variety of visual acuity charts. These subject-specific model parameters (interchangeably referred to herein as chart-invariant parameters) can provide valuable visual acuity parameters that can be used to track visual health over time.

[0008] For example, in the visual acuity model disclosed herein, the features of chart-specific and chart-invariant metrics (also interchangeable as parameters) provide flexibility and versatility for visual acuity estimates. These two sets of visual acuity metrics can be important and complementary. The set of chart-specific metrics can describe visual acuity within the context of the particular visual acuity chart used during testing. The alternative set of chart-invariant metrics can describe visual acuity independent of the visual acuity chart design features. Thus, these generic parameters can have the flexibility to provide visual acuity metrics that are specific to the visual acuity chart used for testing or independent of the visual acuity chart. Chart-specific metrics can be important for research and development applications that use the same chart over time. The chart-invariant metric approach is important for evaluating and comparing data across different clinical practices and populations. As shown in FIG. 3A, a flow chart of an exemplary embodiment of generating a chart-invariant visual acuity model can include an operation of using chart-invariant visual acuity parameters (230), which can then be converted to chart-specific visual acuity parameters (330) depending on visual acuity chart design elements.

[0009] The visual acuity models disclosed herein can further provide the basis for novel presentations of visual acuity data. Visualization tools can help patients and clinicians understand the changes in visual acuity being measured.

[0010] Because the same charts may often be used to measure visual acuity in different eyes, at different distances, or in situations that require repeated measurements of visual acuity (e.g., tracking the progression of eye disease or treatment effects), contamination between conditions is difficult to prevent. Recommendations from the National Academy of Sciences National Research Council (NAS-NRC) Committee and the ANSI (American National Standards Institute) standards stipulate that visual acuity charts must contain multiple columns of optotypes that decrease in size in a logarithmic ratio. In addition to other design recommendations (5-10 optotypes per row, randomly sampled without replacement), standards are recommended for single-letter presentations used by computerized adaptive visual acuity methods.

[0011] Unlike conventional adaptive visual acuity testing methods that optimize the selection of only one size of one letter optotype, the systems and methods disclosed herein can advantageously enable a joint optimization of the visual acuity chart, simultaneously estimating chart-specific and chart-invariant visual acuity metrics. Additional advantages associated with the methods and systems disclosed herein include, but are not limited to, precise individualized focus on each individual subject while adhering to chart design criteria, rapid test times, accuracy comparable to time-consuming single-letter visual acuity testing, and visual acuity metrics that describe the subject's visual acuity independent of the size of the optotype, the test chart, or the test procedure. Both sets of visual acuity parameters, whether chart-specific or chart-invariant, are interchangeable and mutually translatable given the knowledge of the visual acuity chart design.

[0012] In some embodiments, the systems and methods disclosed herein include a chart-based adaptive visual acuity test that provides the standardization benefits provided by a chart-based test with the flexibility and accuracy benefits provided by a single-letter test. The systems and methods disclosed herein can apply a Bayesian adaptive algorithm to individualize and precisely focus on the visual acuity chart to optimally test each subject's visual acuity while following design constraints or recommendations from regulatory committees, or more generally, while maintaining any desired design constraints. In some embodiments, the development of a constrained optimization method for designing visual acuity tests, optionally using a Bayesian adaptive algorithm and a signal detection framework, provides one or more of the following: adherence to chart design criteria; precise individualization through adaptive testing that focuses on individuals; short chart testing time; accuracy of longer single-letter visual acuity tests; modification of behavioral assumptions using different optotype sets; and visual acuity estimates that are invariant to practical issues of visual acuity testing, such as the design of the visual acuity test or the optotypes used.

[0013] In one embodiment, a computer-implemented method for generating a visual acuity model for scoring a subject's visual acuity is disclosed herein, the method comprising: a) obtaining one or more visual acuity chart design parameters; b) collecting visual acuity test data of the subject; c) selecting a first set of chart-specific visual acuity parameters from one or more sets of candidate visual acuity parameters; d) generating a visual acuity model, the visual acuity model comprising: i) generating a psychometric function of a single visual target comprising the first set of chart-specific visual acuity parameters; ii) calculating a family of psychometric functions of multiple visual targets using the psychometric function of the single visual target; and iii) generating a chart-specific psychometric function of a single visual target comprising the second set of chart-specific visual acuity parameters using the family of psychometric functions of the multiple visual targets, wherein the visual acuity model is configurable to estimate a probability of obtaining the visual acuity test data of the subject based on one or more visual acuity chart design parameters. In some embodiments, the method further comprises, after (c) and before (i), generating a sensitivity-based psychometric function including a first set of table-invariant sensitivity parameters, where the first set of table-invariant sensitivity parameters include a sensitivity threshold and a sensitivity range, and the first set of table-invariant sensitivity parameters are generated based on a first set of chart-specific visual acuity parameters. In some embodiments, the sensitivity-based psychometric function is independent of one or more visual acuity chart design parameters. In some embodiments, the sensitivity-based psychometric function is generated based on one or more visual acuity chart design parameters and / or one or more additional parameters of the subject, where the one or more additional parameters are table-invariant. In some embodiments, the sensitivity-based psychometric function is generated based on signal detection theory. In some embodiments, the sensitivity-based psychometric function is configured to describe the visual acuity performance of the subject as a function d' of one or more optotype sizes and independent of one or more chart design parameters. In some embodiments, the method further comprises, before (i), converting the sensitivity-based psychometric function into a psychometric function of the single optotype.In some embodiments, the first set of chart-specific visual acuity parameters includes a visual acuity threshold and a visual acuity range. In some embodiments, the psychometric function of a single optotype is chart-specific. In some embodiments, the second set of chart-specific visual acuity parameters includes a visual acuity threshold and a visual acuity range. In some embodiments, the first or second set of chart-specific visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the first or second set of chart-specific visual acuity parameters includes one or more numerical values. In some embodiments, the first or second set of chart-specific visual acuity parameters includes at least one posterior probability density function for that parameter. In some embodiments, the subject's visual acuity test data includes data from at least two different test conditions. In some embodiments, the first or second set of chart-specific visual acuity parameters includes a posterior probability density function for visual acuity threshold from at least two different test conditions. In some embodiments, the one or more optotype design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof. In some embodiments, the psychometric function of the single optotype is based on at least one optotype size. In some embodiments, each function of the family of psychometric functions of the multiple optotypes is based on two or more different optotype sizes. In some embodiments, the psychometric function of the single optotype includes a visual acuity threshold and a visual acuity range of the subject. In some embodiments, one or more functions in the family of psychometric functions of the multiple optotypes are based on the number of optotypes per row in the optotype chart or sub-chart. In some embodiments, (iii) further includes using a weighted sum of the family of psychometric functions of the multiple optotypes. In some embodiments, the weighting of the family of psychometric functions is based on the subject's optotype chart data.In some embodiments, the weighting is determined manually by a user or automatically by a computer program. In some embodiments, (b) further comprises displaying to the subject one or more vision charts generated based on one or more vision chart design parameters. In some embodiments, the one or more vision charts include at least one optotype, and the at least one optotype is selected from a library of optotypes. In some embodiments, the one or more vision charts are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast in a standard ETDRS chart or Sloane chart. In some embodiments, the at least one optotype is a letter, a number, or a symbol. In some embodiments, the one or more sets of candidate vision parameters are generated based on one or more vision chart design parameters, vision test data of the subject, or both.

[0014] In another aspect, a computer-implemented method for scoring the visual acuity of a subject is disclosed herein, the method comprising: a) obtaining one or more visual acuity chart design parameters; b) collecting visual acuity test data of a subject; c) generating one or more sets of candidate visual acuity parameters based on the one or more visual acuity chart design parameters, the visual acuity test data of the subject, or both; d) generating a visual acuity model, comprising generating one or more chart-specific psychometric functions, each chart-specific psychometric function comprising a set of chart-specific visual acuity parameters of the one or more sets of chart-specific visual acuity parameters; e) using the visual acuity model and one or more sets of candidate visual acuity parameters to generate a probability of observing the visual acuity test data; f) ranking one or more sets of candidate visual acuity parameters of the subject based on the probability; and g) selecting one set from the one or more sets of candidate visual acuity parameters of the subject based on the ranking. In some embodiments, each set of candidate visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, each set of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, each set of candidate visual acuity parameters includes a posterior probability density function for one or more visual acuity parameters. In some embodiments, each set of candidate visual acuity parameters includes a posterior probability density function for each of the one or more visual acuity parameters. In some embodiments, (b) includes summarizing the visual acuity test data from the subject. In some embodiments, the method further includes, after (g), displaying test results of the visual acuity test to the subject or user, the test results including the selected set of candidate visual acuity parameters. In some embodiments, the visual acuity test data of the subject includes data from only the first test condition. In some embodiments, the visual acuity test data of the subject includes data from the first and second test conditions.In some embodiments, the selected set of candidate visual acuity parameters includes a first posterior probability density function for the visual acuity threshold or visual acuity range. In some embodiments, b) includes selecting data having a first test condition. In some embodiments, the method further includes, after g), selecting data having a second test condition from the subject's visual acuity test data; and repeating c) through g). In some embodiments, the selected set of candidate visual acuity parameters includes a second posterior probability density function for the visual acuity threshold or visual acuity range. In some embodiments, the method further includes obtaining a difference distribution using the first and second posterior probability density functions. In some embodiments, the method includes calculating a change in visual acuity threshold or visual acuity range between the first and second test conditions based on the difference distribution. In some embodiments, the visual acuity test results are table-invariant or chart-specific. In some embodiments, (e) is based on Bayes' theorem or maximum likelihood estimation. In some embodiments, the posterior probability density function is a Bayesian posterior probability density function. In some embodiments, the first and second posterior probability density functions are Bayesian posterior probability density functions. In some embodiments, (d) includes: generating a psychometric function of a single optotype including a first set of chart-specific visual acuity parameters; calculating a family of psychometric functions of multiple optotypes using the psychometric function of the single optotype; and generating a chart-specific psychometric function of a second set of chart-specific visual acuity parameters using the family of psychometric functions of multiple optotypes. In some embodiments, the method further includes generating a sensitivity-based psychometric function including a first set of table-invariant sensitivity parameters before generating the psychometric function of the single optotype. In some embodiments, the sensitivity-based psychometric function is independent of one or more visual acuity chart design parameters. In some embodiments, the first set of table-invariant sensitivity parameters includes a sensitivity threshold and a sensitivity range, where the first set of table-invariant sensitivity parameters is generated based on one or more sets of candidate visual acuity parameters.In some embodiments, the sensitivity-based psychometric function is generated based on one or more visual acuity chart design parameters and / or one or more additional parameters of the subject, the one or more additional parameters being table invariant. In some embodiments, the sensitivity-based psychometric function is generated based on signal detection theory. In some embodiments, the sensitivity-based psychometric function is configured to describe the visual acuity performance of the subject as a function d' of one or more visual target sizes and independent of one or more chart design parameters. In some embodiments, the method further comprises, prior to (i), converting the sensitivity-based psychometric function to a psychometric function of the single visual target. In some embodiments, the set of chart-specific visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, the psychometric function of the single visual target is chart-specific. In some embodiments, the set of chart-specific visual acuity parameters comprises a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the set of chart-specific visual acuity parameters comprises one or more numerical values. In some embodiments, the set of chart-specific visual acuity parameters includes at least one posterior probability density function for that parameter. In some embodiments, the subject's visual acuity test data includes data from at least two different test conditions. In some embodiments, the set of chart-specific visual acuity parameters includes a posterior probability density function for visual acuity thresholds of at least two different test conditions. In some embodiments, the one or more visual acuity chart design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof. In some embodiments, the psychometric function of a single optotype is based on at least one optotype size. In some embodiments, each function of the family of psychometric functions of multiple optotypes is based on two or more different optotype sizes.In some embodiments, the psychometric function of the single optotype includes the visual acuity threshold and visual acuity range of the subject. In some embodiments, one or more functions in the family of psychometric functions of the multiple optotypes are based on the number of optotypes per row in the optotype chart or sub-chart. In some embodiments, generating the chart-specific psychometric function further comprises using a weighted sum of the family of psychometric functions of the multiple optotypes. In some embodiments, the weighting of the family of psychometric functions is based on the subject's optotype data. In some embodiments, the weighting is determined manually by a user or automatically by a computer program. In some embodiments, (b) further comprises displaying to the subject one or more optotypes generated based on one or more optotype design parameters. In some embodiments, the one or more optotypes include at least one optotype, and the at least one optotype is selected from a library of optotypes. In some embodiments, the one or more optotypes are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast in a standard ETDRS chart or Sloane chart. In some embodiments, the at least one optotype is a letter, a number, or a symbol. In some embodiments, the method further comprises the steps of: h) generating a plurality of candidate eye charts based on one or more test chart design parameters; and i) selecting one or more of the plurality of candidate eye charts for visual acuity testing of the subject based on a ranking of the plurality of candidate eye charts, the ranking being based on a selected set of candidate eye charts for the subject. In some embodiments, the method further comprises, after (b); collecting data from the subject using the selected one or more candidate eye charts; and if a stopping criterion is not met, repeating steps c)-g) and h)-i) until a stopping criterion is met.

[0015] In yet another aspect, a computer-implemented method for adaptively assessing the visual acuity of a subject is disclosed herein, the method comprising: a) obtaining one or more visual acuity chart design parameters; b) collecting visual acuity test data of the subject; c) generating one or more sets of candidate visual acuity parameters based on the one or more visual acuity chart design parameters and the visual acuity test data of the subject, each of the one or more sets of candidate visual acuity parameters comprising a rank obtained by analyzing the visual acuity test data of the subject; d) generating a plurality of candidate visual acuity charts based on the one or more test chart design parameters; e) ranking the plurality of candidate visual acuity charts based on the rank of each of the one or more sets of candidate visual acuity parameters; and f) selecting one or more visual acuity charts from the plurality of candidate visual acuity charts based on the rank to be presented to the subject for the visual acuity test. In some embodiments, the method further comprises, after f), presenting one or more selected visual acuity charts to the subject. In some embodiments, the method further comprises, after f), updating the visual acuity test data with new data collected from the subject using the one or more selected visual acuity charts. In some embodiments, the method further comprises presenting the subject with one or more selected acuity charts; updating the visual acuity test data with new data collected from the subject using the one or more selected acuity charts; and repeating steps b) to f) until a stopping criterion is met. In some embodiments, each set of the one or more sets of candidate visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, each set of the one or more sets of candidate visual acuity parameters comprises a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, each set of the one or more sets of candidate visual acuity parameters comprises a posterior probability density function for the one or more visual acuity parameters. In some embodiments, each set of the one or more sets of candidate visual acuity parameters comprises a posterior probability density function for each of the one or more visual acuity parameters.In some embodiments, (b) includes summarizing the visual acuity testing data from the subject. In some embodiments, the visual acuity testing data of the subject includes data from only the first test condition. In some embodiments, the visual acuity testing data of the subject includes data from the first and second test conditions. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the visual acuity testing data of the subject includes data from at least two different test conditions. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a posterior probability density function for the visual acuity threshold or visual acuity range. In some embodiments, the one or more visual acuity chart design parameters include: a total number of optotypes, a number of optotypes per row, a method of optotype sampling, an optotype size, a number of rows, and a subject's response, or a combination thereof. In some embodiments, each of the candidate visual acuity charts includes at least one optotype, and the at least one optotype is selected from a library of optotypes. In some embodiments, one or more of the candidate eye charts are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast on a standard ETDRS or Sloane chart.The method of claim 90, wherein at least one of the indicia is a letter, a number, or a symbol.

[0016] In yet another aspect, a computer-implemented system for generating a visual acuity model for scoring a subject's visual acuity is disclosed herein, the system comprising: a digital processing device including an operating system and a memory configured to execute executable instructions; and a computer program, the computer program including instructions executable by the digital processing device to create a scoring application including a software module configured to: a) obtain one or more visual acuity chart design parameters; b) collect visual acuity test data of the subject; c) select a first set of chart-specific visual acuity parameters from one or more sets of candidate visual acuity parameters; d) generate a visual acuity model, the visual acuity model including: i. generate a psychometric function of a single optotype including the first set of chart-specific visual acuity parameters; ii. use the psychometric function of the single optotype to calculate a family of psychometric functions of multiple optotypes; and iii. use the family of psychometric functions of the multiple optotypes to generate a chart-specific psychometric function including a second set of chart-specific visual acuity parameters, wherein the visual acuity model is configurable to estimate a probability of obtaining the visual acuity test data of the subject based on one or more visual acuity chart design parameters. In some embodiments, the system includes, after (c) and before (i), generating a sensitivity-based psychometric function including a first set of chart-invariant sensitivity parameters, where the first set of chart-invariant sensitivity parameters include a sensitivity threshold and a sensitivity range, and where the first set of chart-invariant sensitivity parameters are generated based on a first set of chart-specific visual acuity parameters. In some embodiments, the sensitivity-based psychometric function is independent of one or more visual acuity chart design parameters. In some embodiments, the sensitivity-based psychometric function is generated based on one or more visual acuity chart design parameters and / or one or more additional parameters of the subject, where the one or more additional parameters are chart-invariant. In some embodiments, the sensitivity-based psychometric function is generated based on signal detection theory.In some embodiments, the sensitivity-based psychometric function is configured to describe a subject's visual acuity performance as a function d' of one or more object sizes and independent of one or more chart design parameters. In some embodiments, the system further comprises, prior to (i), converting the sensitivity-based psychometric function to a psychometric function of the single object. In some embodiments, the first set of chart-specific visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, the psychometric function of the single object is chart-specific. In some embodiments, the second set of chart-specific visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, the first or second set of chart-specific visual acuity parameters comprises a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the first or second set of chart-specific visual acuity parameters comprises one or more numerical values. In some embodiments, the first or second set of chart-specific visual acuity parameters comprises at least one posterior probability density function for the parameter. In some embodiments, the subject's visual acuity test data includes data from at least two different test conditions. In some embodiments, the first set or the second set of chart-specific visual acuity parameters includes posterior probability density functions for visual acuity thresholds from at least two different test conditions. In some embodiments, the one or more visual acuity chart design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof. In some embodiments, the psychometric function of the single optotype is based on at least one optotype size. In some embodiments, each function of the family of psychometric functions of multiple optotypes is based on two or more different optotype sizes. In some embodiments, the psychometric function of the single optotype includes a visual acuity threshold and a visual acuity range of the subject.In some embodiments, one or more functions in the family of psychometric functions of multiple optotypes are based on the number of optotypes per row in the optotype chart or sub-chart. In some embodiments, (iii) further comprises using a weighted sum of the family of psychometric functions of multiple optotypes. In some embodiments, the weighting of the family of psychometric functions is based on the subject's optotype data. In some embodiments, the weighting is determined manually by a user or automatically by a computer program. In some embodiments, (b) further comprises using one or more optotypes generated based on one or more optotype design parameters. In some embodiments, the one or more optotypes include at least one optotype, and the at least one optotype is selected from a library of optotypes. In some embodiments, the one or more optotypes are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast in a standard ETDRS chart or Sloane chart. In some embodiments, the at least one optotype is a letter, a number, or a symbol. In some embodiments, the one or more sets of candidate visual acuity parameters are generated based on one or more optotype design parameters, the subject's visual acuity test data, or both.

[0017] In yet another aspect, a computer system for generating a visual acuity model for scoring a subject's visual acuity is disclosed herein, the system comprising: a digital processing device including an operating system and a memory configured to execute executable instructions; and a computer program, the computer program comprising instructions executable by the digital processing device to create a scoring application including a software module configured to: a) obtain one or more visual acuity chart design parameters; b) collect visual acuity test data of the subject; c) generate one or more sets of candidate visual acuity parameters based on the one or more visual acuity chart design parameters, the visual acuity test data of the subject, or both; d) generate a visual acuity model including generating one or more chart-specific psychometric functions, each chart-specific psychometric function including a set of chart-specific visual acuity parameters of the one or more sets of chart-specific visual acuity parameters; e) generate a probability of observing the visual acuity test data using the visual acuity model and the one or more sets of candidate visual acuity parameters; f) rank the one or more sets of candidate visual acuity parameters of the subject based on the likelihood; and g) select a set from the one or more sets of candidate visual acuity parameters of the subject based on the rank. In some embodiments, each set of candidate visual acuity parameters includes a visual acuity threshold and a visual acuity range. In some embodiments, each set of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, each set of candidate visual acuity parameters includes a posterior probability density function for one or more visual acuity parameters. In some embodiments, each set of candidate visual acuity parameters includes a posterior probability density function for each of the one or more visual acuity parameters. In some embodiments, (b) summarizes the subject's visual acuity test data.In some embodiments, the system further includes, after (g), displaying a result of the visual acuity test to the subject or user, the result including the selected set of candidate visual acuity parameters. In some embodiments, the visual acuity test data of the subject includes data from only the first test condition. In some embodiments, the visual acuity test data of the subject includes data from the first and second test conditions. In some embodiments, the selected set of candidate visual acuity parameters includes a first posterior probability density function for a visual acuity threshold or visual acuity range. In some embodiments, b) includes selecting data having the first test condition. In some embodiments, the system further includes, after g), selecting data having the second test condition from the visual acuity test data of the subject; and repeating c) through g). In some embodiments, the selected set of candidate visual acuity parameters includes a second posterior probability density function for a visual acuity threshold or visual acuity range. In some embodiments, the system further includes using the first and second posterior probability density functions to obtain a difference distribution. In some embodiments, the system further includes calculating a change in visual acuity threshold or visual acuity range between the first and second test conditions based on the difference distribution. In some embodiments, the results of the visual acuity test are table-invariant or chart-specific. In some embodiments, (e) is based on Bayes' theorem or maximum likelihood estimation. In some embodiments, the posterior probability density function is a Bayesian posterior probability density function. In some embodiments, the first and second posterior probability density functions are Bayesian posterior probability density functions. In some embodiments, (d) includes generating a single-target psychometric function comprising a first set of chart-specific visual acuity parameters; calculating a family of multiple target psychometric functions using the single target psychometric function; and generating a chart-specific psychometric function comprising a second set of chart-specific visual acuity parameters using the family of multiple target psychometric functions.In some embodiments, the system further comprises generating a sensitivity-based psychometric function including a first set of chart-invariant sensitivity parameters prior to generating the psychometric function of the single optotype. In some embodiments, the sensitivity-based psychometric function is independent of one or more acuity chart design parameters. In some embodiments, the first set of chart-invariant sensitivity parameters includes a sensitivity threshold and a sensitivity range. In some embodiments, the sensitivity-based psychometric function is generated based on one or more acuity chart design parameters and / or one or more additional parameters based on the subject, the one or more additional parameters being table-invariant. In some embodiments, the sensitivity-based psychometric function is generated based on signal detection theory. In some embodiments, the sensitivity-based psychometric function is configured to describe the visual acuity performance of a subject as a function d' of one or more optotype sizes and independent of one or more chart design parameters. In some embodiments, the system further comprises converting the sensitivity-based psychometric function to the psychometric function of the single optotype. In some embodiments, the set of chart-specific visual acuity parameters includes a visual acuity threshold and a visual acuity range. In some embodiments, the psychometric function of the single optotype is chart-specific. In some embodiments, the set of chart-specific visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the first set of chart-specific visual acuity parameters includes one or more numerical values. In some embodiments, the set of chart-specific visual acuity parameters includes at least one posterior probability density function for that parameter. In some embodiments, the subject's visual acuity testing data includes data from at least two different test conditions. In some embodiments, the set of chart-specific visual acuity parameters includes posterior probability density functions for visual acuity thresholds of at least two different test conditions.In some embodiments, the one or more optotype design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof. In some embodiments, the psychometric function of the single optotype is based on at least one optotype size. In some embodiments, each function of the family of psychometric functions of the multiple optotypes is based on two or more different optotype sizes. In some embodiments, the psychometric function of the single optotype includes a visual acuity threshold and a visual acuity range of the subject. In some embodiments, one or more functions in the family of psychometric functions of the multiple optotypes are based on the number of optotypes per row in the optotype chart or sub-chart. In some embodiments, generating the chart-specific psychometric function further includes using a weighted sum of the family of psychometric functions of the multiple optotypes. In some embodiments, the weighted sum of the family of psychometric functions is based on the subject's optotype chart data. In some embodiments, the weighting is determined manually by a user or automatically by a computer program. In some embodiments, (b) further comprises displaying to the subject one or more optotypes generated based on the one or more optotype design parameters. In some embodiments, the one or more optotypes include at least one optotype selected from a library of optotypes. In some embodiments, the one or more optotypes are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast in a standard ETDRS or Sloane chart. In some embodiments, the at least one optotype is a letter, a number, or a symbol. In some embodiments, the system further comprises: h) generating a plurality of candidate optotypes based on the one or more test chart design parameters; and i) selecting one or more of the plurality of candidate optotypes for visual acuity testing of the subject based on a ranking of the plurality of candidate optotypes, the ranking being based on the selected set of candidate optotype parameters for the subject.In some embodiments, the system further includes, after (b); collecting data from the subject using the selected one or more candidate eye charts; and if the stopping criterion is not met, repeating steps c) through g) and h) through i) until the stopping criterion is met.

[0018] In yet another aspect, a computer-implemented system for adaptively assessing the visual acuity of a subject is disclosed herein, the method comprising: a digital processing device including an operating system and a memory configured to execute executable instructions; and a computer program, the computer program comprising: a) obtaining one or more visual acuity chart design parameters; b) collecting visual acuity test data of the subject; c) generating one or more sets of candidate visual acuity parameters based on the one or more visual acuity chart design parameters and the visual acuity test data of the subject, each of the one or more sets of candidate visual acuity parameters comprising a rank obtained by analyzing the visual acuity test data of the subject; d) generating a plurality of candidate visual acuity charts based on the one or more test chart design parameters; e) ranking the plurality of candidate visual acuity charts based on the rank of each of the one or more sets of candidate visual acuity parameters; and f) selecting one or more visual acuity charts from the plurality of candidate visual acuity charts based on the rank to be presented to the subject for visual acuity testing. In some embodiments, the system further comprises, after f), presenting one or more selected visual acuity charts to the subject. In some embodiments, the system further comprises, after f), updating the visual acuity test data with new data collected from the subject using the one or more selected visual acuity charts. In some embodiments, the system further includes presenting the subject with one or more selected acuity charts; updating the visual acuity test data with new data collected from the subject using the one or more selected acuity charts; and repeating steps b) to f) until a stopping criterion is met. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold and a visual acuity range. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a posterior probability density function for the one or more visual acuity parameters.In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a posterior probability density function for each of the one or more visual acuity parameters. In some embodiments, (b) summarizes the visual acuity test data of the subject. In some embodiments, the visual acuity test data of the subject includes data from only the first test condition. In some embodiments, the visual acuity test data of the subject includes data from the first and second test conditions. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the visual acuity test data of the subject includes data from at least two different test conditions. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a posterior probability density function for a visual acuity threshold or visual acuity range. In some embodiments, the one or more visual acuity chart design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof. In some embodiments, each of the candidate optotypes includes at least one optotype, the at least one optotype being selected from a library of optotypes. In some embodiments, one or more of the candidate optotypes is of contrast that is about 25%, 2.5%, or 1.5% of normal contrast on a standard ETDRS or Sloane chart. In some embodiments, the at least one optotype is a letter, a number, or a symbol.

[0019] Consistent with other disclosed embodiments, a non-transitory computer-readable storage medium can store program instructions that are executed by a processor to perform any of the methods described herein.

[0020] Further aspects and advantages of the present disclosure will become readily apparent to those skilled in the art from the following detailed description, in which only illustrative embodiments of the present disclosure have been shown and described. As will be realized, the present disclosure is capable of other and different embodiments, and its various details can be modified in various obvious respects, all without departing from the present disclosure. Accordingly, the drawings and description are to be regarded as illustrative in nature, and not as restrictive.

[0021] <Incorporation by reference> All publications, patents, and patent applications mentioned in this specification are incorporated herein by reference to the extent that each individual publication, patent, or patent application is specifically and individually indicated to be incorporated by reference. To the extent that a publication, patent, or patent application incorporated by reference conflicts with the disclosure contained herein, the present specification is intended to supersede and / or take precedence over such conflicting material. [Brief description of the drawings]

[0022] The novel features of the invention are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings (also referred to herein as "Figure" and "FIG.") of the following drawings.

[0023] [Figure 1] 1 shows a non-limiting example flow chart of a scoring algorithm S. [Diagram 2] A-B are non-limiting examples of the generation of a library of eye charts and sub-charts with various chart designs; in this case, the ETDRS chart (Early Treatment Diabetic Retinopathy Study, Figure 2A) and the Snellen eye chart (Figure 2B) are shown. [Figure 3A] 1 shows a non-limiting example of a flowchart for generating a chart-invariant visual acuity model. [Figure 3B]1 shows a non-limiting example of a flowchart for generating a chart-specific visual acuity model. [Figure 4] AB shows a non-limiting example of a single optotype psychometric function based on sensitivity (d'), which is a psychometric function that describes perceptual sensitivity independent of decision-level factors that are chart-invariant and influenced by optotype design parameters (d' units of signal detection theory). [Diagram 5] A non-limiting example set of chart-specific psychometric functions is shown that reflect the transformation of chart-invariant single-target psychometric functions (in d' units) to chart-specific single-target psychometric functions based on chart design factors, which can include the number of response options that determine guess rate (2-alternative forced choice (2AFC) to 10-alternative forced choice (10AFC)). [Figure 6A] Non-limiting examples of single target psychometric functions based on 10 guess options generated by the visual acuity model (Figure 6A) and various sets of multiple target psychometric functions (Figure 6B) for recognizing five targets presented on a slide as a function of target size are shown. [Figure 6B] Non-limiting examples of single target psychometric functions based on 10 guess options generated by the visual acuity model (Figure 6A) and various sets of multiple target psychometric functions (Figure 6B) for recognizing five targets presented on a slide as a function of target size are shown. [Figure 6C] 6C shows a non-limiting example of a chart-specific psychometric function generated from a weighted sum of a family of psychometric functions for multiple targets in FIG. 6B. [Figure 7] 1 shows non-limiting examples of chart-specific psychometric functions defined by chart-specific acuity thresholds and chart-specific acuity ranges for recognizing five targets presented on a slide as a function of target size. [Figure 8] AB show non-limiting examples of single target psychometric functions (FIG. 8A) and various sets of multiple target psychometric functions (FIG. 8B) presented as cumulative probability density functions. [Figure 9]1 shows a non-limiting example of a flowchart of an adaptation algorithm A disclosed herein. [Figure 10] A non-limiting example of generating a scalable library of candidate eye charts by determining varying sampling patterns of optotype sizes based on chart design principles is provided. [Figure 11] AF shows non-limiting examples of candidate eye charts that use various anchor points based on the full standard constrained eye chart. [Figure 12] AC shows non-limiting examples of specific regions or sub-charts of the complete standard constrained eye chart. [Figure 13] AC provides non-limiting examples of collecting responses from subjects using three different models, where subjects read down the first column (Figure 13A), read down the chart line by line from the top (Figure 13B), or read a single letter presented in the center of the screen (Figure 13C). [Figure 14] AB shows a non-limiting example comparing the results of a single row optimization computation with a composite multi-row subchart optimization. [Figure 15] A non-limiting example of selection of an individualized eye chart for a sequential test slide (n+1) based on the subject's response to the nth test slide is shown. [Figure 16] 1 shows a non-limiting example of a flow chart of a combination algorithm using algorithms A and S disclosed herein. [Figure 17] 1 shows another non-limiting example of a flow chart of a combination algorithm using algorithms A and S disclosed herein. [Figure 18] A non-limiting example of using Bayesian posterior probability to estimate a subject's visual acuity threshold, in this case using three slides, is shown. The inset shows the convergence of visual acuity estimates with increasing number of test slides. [Figure 19] 1 shows a non-limiting example of variability due to different numbers of test slides on a standard constrained adaptive visual acuity test with 1-5 slides and 4 different runs per test. [Figure 20] 13 shows a non-limiting example of validation of a combinatorial algorithm. [Figure 21] 13 shows a non-limiting example of validation of a combinatorial algorithm. [Figure 22] 1 shows a non-limiting example of displaying the results of a vision test. [Figure 23] 1 shows another non-limiting example of displaying the results of a vision test to a subject. [Figure 24] 1 illustrates a non-limiting example of a digital processing device as disclosed herein. [Diagram 25] A non-limiting example of a web / mobile application delivery system is shown, in this case a system that provides a browser-based and / or native mobile user interface. [Figure 26] A non-limiting schematic diagram of a cloud-based web / mobile application delivery system is shown, in this case including elastically load balanced, auto-scaling web server and application server resources, and a synchronously replicated database. [Figure 27] AB shows a non-limiting example of using Bayesian posterior probability density function (pdf) to estimate visual acuity change between various testing conditions. [Figure 28] AB shows a non-limiting example of the difference distribution of Bayesian posterior probability density functions for different test conditions, which provides an indication of visual acuity threshold shifts. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0024] The methods, systems, and devices disclosed herein relate to visual acuity assessment using a vision test.

[0025] The systems and methods disclosed herein may include a wide range of advantages over existing visual acuity tests, analysis methods, or systems. Unlike existing adaptive visual acuity testing methods that optimize the selection of only one character of one size, the current systems and methods may allow optimization of the entire visual acuity presented on each test slide. The systems and methods disclosed herein may allow simultaneous estimation of chart-specific and chart-invariant visual acuity parameters or metrics. Further advantages associated with the methods and systems disclosed herein may include precise focus on each subject while adhering to chart design criteria, fast test times, accuracy comparable to time-consuming single-character visual acuity tests, and visual acuity results independent of visual acuity target size, test chart, or test procedure. Both sets of chart-specific and chart-invariant visual acuity parameters are interchangeable and interchangeable herein.

[0026] In some embodiments, a computer-implemented method for generating a visual acuity model for scoring a subject's visual acuity is disclosed herein, the method comprising: a) obtaining one or more visual acuity chart design parameters; b) collecting visual acuity test data of the subject; c) selecting a first set of chart-specific visual acuity parameters from one or more sets of candidate visual acuity parameters; and d) generating a visual acuity model comprising: i) generating a psychometric function of a single visual target comprising the first set of chart-specific visual acuity parameters; ii) calculating a family of psychometric functions of a plurality of visual targets using the psychometric function of the single visual target; and iii) generating a chart-specific psychometric function of a second set of chart-specific visual acuity parameters using the family of psychometric functions of the plurality of visual targets, wherein the visual acuity model is configurable to estimate a probability of obtaining the visual acuity test data of the subject based on the one or more visual acuity chart design parameters. In some embodiments, the method includes, after (c) and before (i), generating a psychometric function including a first set of chart-invariant sensitivity parameters based on the sensitivity, where the first set of chart-invariant sensitivity parameters includes a sensitivity threshold and a sensitivity range, and where the first set of chart-invariant sensitivity parameters is generated based on a first set of chart-specific visual acuity parameters. In some embodiments, the sensitivity-based psychometric function is independent of one or more visual acuity chart design parameters. In some embodiments, the sensitivity-based psychometric function is generated based on one or more visual acuity chart design parameters and / or one or more additional parameters of the subject, where the one or more additional parameters are chart-invariant. In some embodiments, the sensitivity-based psychometric function is generated based on signal detection theory. In some embodiments, the sensitivity-based psychometric function is configured to describe the visual acuity performance of the subject as a function d' of one or more target sizes and independent of one or more chart design parameters.In some embodiments, the method further comprises, prior to (i), converting the sensitivity-based psychometric function to a single-target psychometric function. In some embodiments, the first set of chart-specific visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, the single-target psychometric function is chart-specific. In some embodiments, the second set of chart-specific visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, the first or second set of chart-specific visual acuity parameters comprises a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the first or second set of chart-specific visual acuity parameters comprises one or more numerical values. In some embodiments, the first or second set of chart-specific visual acuity parameters comprises at least one posterior probability density function of the parameter. In some embodiments, the subject's visual acuity test data comprises data from at least two different test conditions. In some embodiments, the first or second set of chart-specific visual acuity parameters includes a posterior probability density function of visual acuity thresholds from at least two different test conditions. In some embodiments, the one or more visual acuity chart design parameters include: total number of visual acuities, number of visual acuities per row, method of visual acuity chart sampling, visual acuity size, number of rows, and subject response, or a combination thereof. In some embodiments, the psychometric function of the single visual acuity is based on at least one visual acuity size. In some embodiments, each function of the family of psychometric functions of the multiple visual acuities is based on two or more different visual acuity sizes. In some embodiments, the psychometric function of the single visual acuity includes a visual acuity threshold and a visual acuity range of the subject. In some embodiments, one or more functions in the family of psychometric functions of the multiple visual acuities are based on the number of visual acuities per row in the visual acuity chart or sub-chart. In some embodiments, (iii) further includes using a weighted sum of the family of psychometric functions of the multiple visual acuities.In some embodiments, the weighting of the family of psychometric functions is based on the subject's eye chart data. In some embodiments, the weighting is determined manually by a user or automatically by a computer program. In some embodiments, (b) includes displaying to the subject one or more eye charts generated based on one or more eye chart design parameters. In some embodiments, each of the one or more eye charts includes at least one optotype. In some embodiments, the one or more eye charts are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast in a standard ETDRS chart or Sloane chart. In some embodiments, the at least one optotype is a letter, a number, or a symbol. In some embodiments, the one or more sets of candidate eye chart parameters are generated based on one or more eye chart design parameters, the subject's eye test data, or both.

[0027] In some embodiments, a computer-implemented method for scoring the visual acuity of a subject is disclosed herein, the method comprising: a) obtaining one or more visual acuity chart design parameters; b) collecting visual acuity test data of the subject; c) generating one or more sets of candidate visual acuity parameters based on the one or more visual acuity chart design parameters, the visual acuity test data of the subject, or both; d) generating one or more chart-specific psychometric functions, each chart-specific psychometric function comprising one set of chart-specific visual acuity parameters of the one or more sets of chart-specific visual acuity parameters; e) generating a probability of observing visual acuity test data using the visual acuity model and the one or more sets of candidate visual acuity parameters; f) ranking the one or more sets of candidate visual acuity parameters of the subject based on the probability; g) selecting one set from the one or more sets of candidate visual acuity parameters for the subject based on the ranking. In some embodiments, each set of candidate visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, each set of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, each set of candidate visual acuity parameters includes a posterior probability density function of one or more visual acuity parameters. In some embodiments, each set of candidate visual acuity parameters includes a posterior probability density function of each of the one or more visual acuity parameters. In some embodiments, (b) includes summarizing the visual acuity test data of the subject. In some embodiments, the method includes, after (g), displaying results of the visual acuity test to the subject or user, the test results further including the selected set of candidate visual acuity parameters. In some embodiments, the visual acuity test data of the subject includes data from only the first test condition. In some embodiments, the visual acuity test data of the subject includes data from the first and second test conditions.In some embodiments, the selected set of candidate visual acuity parameters includes a first posterior probability density function of a visual acuity threshold or visual acuity range. In some embodiments, b) includes selecting data having a first test condition. In some embodiments, the method includes, after g), selecting data having a second test condition from the visual acuity test data of the subject, and repeating c)-g). In some embodiments, the selected set of candidate visual acuity parameters includes a second posterior probability density function of a visual acuity threshold or visual acuity range. In some embodiments, the method further includes obtaining a difference distribution using the first and second posterior probability density functions. In some embodiments, the method includes calculating a change in visual acuity threshold or visual acuity range between the first and second test conditions based on the difference distribution. In some embodiments, the visual acuity test results are chart invariant or chart specific. In some embodiments, (e) is based on Bayes' theorem or maximum likelihood estimation. In some embodiments, the posterior probability density function is a Bayesian posterior probability density function. In some embodiments, the first or second posterior probability density function is a Bayesian posterior probability density function. In some embodiments, (d) includes: generating a psychometric function of a single optotype including a first set of chart-specific visual acuity parameters; calculating a family of psychometric functions of multiple optotypes using the psychometric function of the single optotype; and generating a chart-specific psychometric function including a second set of chart-specific visual acuity parameters using the family of psychometric functions of multiple optotypes. In some embodiments, the method further includes generating a sensitivity-based psychometric function including a first set of chart-invariant sensitivity parameters before generating the psychometric function of the single optotype. In some embodiments, the sensitivity-based psychometric function is independent of one or more visual acuity chart design parameters. In some embodiments, the first set of chart-invariant sensitivity parameters includes a sensitivity threshold and a sensitivity range, where the first set of chart-invariant sensitivity parameters is generated based on one or more sets of candidate visual acuity parameters.In some embodiments, the sensitivity-based psychometric function is generated based on one or more visual acuity chart design parameters and / or one or more additional parameters of the subject, the one or more additional parameters being chart invariant. In some embodiments, the sensitivity-based psychometric function is generated based on signal detection theory. In some embodiments, the sensitivity-based psychometric function is configured to describe the visual acuity performance of the subject as a function d' of one or more visual target sizes and independent of one or more chart design parameters. In some embodiments, the method further comprises converting the sensitivity-based psychometric function to a psychometric function of a single visual target. In some embodiments, the set of chart-specific visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, the psychometric function of a single visual target is chart-specific. In some embodiments, the set of chart-specific visual acuity parameters comprises a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the set of chart-specific visual acuity parameters comprises one or more numerical values. In some embodiments, the set of chart-specific visual acuity parameters includes at least one posterior probability density function of the parameter. In some embodiments, the subject's visual acuity test data includes data from at least two different test conditions. In some embodiments, the set of chart-specific visual acuity parameters includes at least one posterior probability density function of visual acuity thresholds of at least two different test conditions. In some embodiments, the one or more visual acuity chart design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof. In some embodiments, the psychometric function of the single optotype is based on at least one optotype size. In some embodiments, each function of the family of psychometric functions of the multiple optotypes is based on two or more different optotype sizes. In some embodiments, the psychometric function of the single optotype includes the subject's visual acuity threshold and visual acuity range.In some embodiments, one or more functions in the family of psychometric functions of the multiple optotypes are based on the number of optotypes per row in the optotype chart or sub-chart. In some embodiments, generating the chart-specific psychometric function further comprises using a weighted sum of the family of psychometric functions of the multiple optotypes. In some embodiments, the weighting of the family of psychometric functions is based on the subject's optotype data. In some embodiments, the weighting is determined manually by a user or automatically by a computer program. In some embodiments, (b) comprises displaying to the subject one or more optotypes generated based on one or more optotype design parameters. The method of claim 71, wherein each of the one or more optotypes comprises at least one optotype. In some embodiments, the one or more optotypes are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast in a standard ETDRS chart or Sloane chart. In some embodiments, the at least one optotype is a letter, a number, or a symbol. In some embodiments, the method further comprises: h) generating a plurality of candidate eye charts based on one or more chart design parameters; and i) selecting one or more of the plurality of candidate eye charts for visual acuity testing of the subject based on a ranking of the plurality of candidate eye charts, wherein the ranking is based on a selected set of candidate eye charts for the subject. In some embodiments, the method further comprises, after (b), collecting data from the subject using the selected one or more candidate eye charts; and if a stopping criterion is not met, repeating steps c)-g) and h)-i); until a stopping criterion is met.

[0028] In some embodiments, a computer-implemented method for adaptively assessing the visual acuity of a subject is disclosed herein, the method comprising: a) obtaining one or more visual acuity chart design parameters; b) collecting visual acuity test data of the subject; c) generating one or more sets of candidate visual acuity parameters based on the one or more visual acuity chart design parameters and the visual acuity test data of the subject, each of the one or more sets of candidate visual acuity parameters having a rank obtained by analyzing the visual acuity test data of the subject; d) generating a plurality of candidate visual acuity charts based on the one or more visual acuity chart design parameters; e) ranking the plurality of candidate visual acuity charts based on the rank of each of the one or more sets of candidate visual acuity parameters; and f) selecting one or more visual acuity charts from the plurality of candidate visual acuity charts based on their ranks to be presented to the subject for visual acuity testing. In some embodiments, the method further comprises, after f), presenting the one or more selected visual acuity charts to the subject. In some embodiments, the method further comprises, after f), updating the visual acuity test data with new data collected from the subject using the one or more selected visual acuity charts. In some embodiments, the method further comprises presenting the subject with one or more selected acuity charts; updating the visual acuity test data with new data collected from the subject using the one or more selected acuity charts; and repeating steps b)-f); until a stopping criterion is met. In some embodiments, each set of the one or more sets of candidate visual acuity parameters comprises a visual acuity threshold and a visual acuity range. In some embodiments, each set of the one or more sets of candidate visual acuity parameters comprises a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, each set of the one or more candidate visual acuity parameters comprises a posterior probability density function of the one or more visual acuity parameters. In some embodiments, each set of the one or more candidate visual acuity parameters comprises a posterior probability density function of each of the one or more visual acuity parameters. In some embodiments, (b) comprises summarizing the visual acuity test data of the subject.In some embodiments, the subject's visual acuity testing data includes data from only the first test condition. In some embodiments, the subject's visual acuity testing data includes data from the first and second test conditions. In some embodiments, each of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the subject's visual acuity testing data includes data from at least two different test conditions. In some embodiments, each of the one or more sets of candidate visual acuity parameters includes a posterior probability density function of a visual acuity threshold or an acuity acuity range of acuity acuity. In some embodiments, the one or more visual acuity chart design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof. In some embodiments, each of the candidate visual acuity charts includes at least one optotype. In some embodiments, the candidate eye chart is of contrast that is about 25%, 2.5%, or 1.5% of normal contrast on a standard ETDRS or Sloane chart. The method of claim 90, wherein at least one of the optotypes is a letter, a number, or a symbol.

[0029] In some embodiments, a computer-implemented system for generating a visual acuity model for scoring a subject's visual acuity is disclosed herein, the computer-implemented system comprising: a digital processing device including an operating system and a memory configured to execute executable instructions; and a computer program, the computer program including instructions executable by the digital processing device for generating a scoring application including a software module configured to perform the following: a) obtaining one or more visual acuity chart design parameters; b) collecting visual acuity test data of the subject; c) selecting a first set of chart-specific visual acuity parameters from one or more sets of candidate visual acuity parameters; d) generating a visual acuity model, the visual acuity model including: i. a psychometric function of a single visual acuity chart generating the first set of chart-specific visual acuity parameters; ii. using the psychometric function of the single visual acuity chart to calculate a family of psychometric functions of multiple visual acuity charts; and iii. using the family of psychometric functions of the multiple visual acuity charts including a second set of chart-specific visual acuity parameters to generate a chart-specific psychometric function, wherein the visual acuity model is configurable to estimate the likelihood of obtaining the visual acuity test data of the subject based on one or more visual acuity chart design parameters. In some embodiments, the system includes, after (c) and before (i), generating a sensitivity-based psychometric function including a first set of chart-invariant sensitivity parameters, where the first set of chart-invariant sensitivity parameters includes a sensitivity threshold and a sensitivity range, and where the first set of chart-invariant sensitivity parameters is generated based on a first set of chart-specific visual acuity parameters. In some embodiments, the sensitivity-based psychometric function is independent of one or more visual acuity chart design parameters. In some embodiments, the sensitivity-based psychometric function is generated based on one or more visual acuity chart design parameters and / or one or more additional parameters of a subject, where the one or more additional parameters are chart-invariant. In some embodiments, the sensitivity-based psychometric function is generated based on signal detection theory.In some embodiments, the sensitivity-based psychometric function is configured to describe a subject's visual acuity performance as a function d' of one or more optotype sizes and independent of one or more chart design parameters. In some embodiments, the system further comprises, prior to (i), converting the sensitivity-based psychometric function to a psychometric function of the single optotype. In some embodiments, the first set of chart-specific visual acuity parameters comprises an acuity threshold and an acuity range. In some embodiments, the psychometric function of the single optotype is chart-specific. In some embodiments, the second set of chart-specific visual acuity parameters comprises an acuity threshold and an acuity range. In some embodiments, the first or second set of chart-specific visual acuity parameters comprises an acuity threshold, an acuity range, an acuity slope, a change in acuity threshold between two test conditions, a change in acuity range between two test conditions, a change in acuity slope between two test conditions, or a combination thereof. In some embodiments, the first or second set of chart-specific visual acuity parameters comprises one or more numerical values. In some embodiments, the first or second set of chart-specific visual acuity parameters comprises at least one posterior probability density function for the parameter. In some embodiments, the subject's visual acuity test data includes data from at least two different test conditions. In some embodiments, the first set or the second set of chart-specific visual acuity parameters includes posterior probability density functions for visual acuity thresholds from at least two different test conditions. In some embodiments, the one or more visual acuity chart design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof. In some embodiments, the psychometric function of the single optotype is based on at least one optotype size. In some embodiments, each function of the family of psychometric functions of multiple optotypes is based on two or more different optotype sizes. In some embodiments, the psychometric function of the single optotype includes a visual acuity threshold and a visual acuity range of the subject.In some embodiments, one or more functions in the family of psychometric functions of the multiple optotypes are based on the number of optotypes per row in the optotype chart or sub-chart. In some embodiments, (iii) further comprises using a weighted sum of the family of psychometric functions of the multiple optotypes. In some embodiments, the weighting of the family of psychometric functions is based on the subject's optotype data. In some embodiments, the weighting is determined manually by a user or automatically by a computer program. In some embodiments, (b) further comprises using one or more optotypes generated based on one or more optotype design parameters. In some embodiments, each of the one or more optotypes includes at least one optotype. In some embodiments, the one or more optotypes are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast in a standard ETDRS chart or Sloane chart. In some embodiments, the at least one optotype is a letter, a number, or a sign. In some embodiments, the one or more sets of candidate visual acuity parameters are generated based on one or more optotype design parameters, the subject's visual acuity test data, or both.

[0030] In some embodiments, a computer system for generating a visual acuity model for scoring a subject's visual acuity is disclosed herein, the computer system comprising: a digital processing device including an operating system and a memory configured to execute executable instructions; and a computer program, the computer program comprising instructions executable by the digital processing device for generating a scoring application including a software module configured to: a) obtain one or more visual acuity chart design parameters; b) collect visual acuity test data of the subject; c) generate one or more sets of candidate visual acuity parameters based on the one or more visual acuity chart design parameters, the visual acuity test data of the subject, or both; d) generate a visual acuity model including generating one or more chart-specific psychometric functions, each chart-specific psychometric function including one or more sets of chart-specific visual acuity parameters; e) generate a probability of observing the visual acuity test data using the visual acuity model and the one or more sets of candidate visual acuity parameters; f) rank the one or more sets of candidate visual acuity parameters of the subject based on the probability; and g) select a set from the one or more sets of candidate visual acuity parameters of the subject based on the ranking. In some embodiments, each set of candidate visual acuity parameters includes a visual acuity threshold and a visual acuity range. In some embodiments, each set of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, each set of candidate visual acuity parameters includes a posterior probability density function for one or more visual acuity parameters. In some embodiments, each set of candidate visual acuity parameters includes a posterior probability density function for each of the one or more visual acuity parameters. In some embodiments, (b) summarizes the subject's visual acuity test data.In some embodiments, the system further includes, after (g), displaying a result of the visual acuity test to the subject or user, the result including the selected set of candidate visual acuity parameters. In some embodiments, the visual acuity test data of the subject includes data from only the first test condition. In some embodiments, the visual acuity test data of the subject includes data from the first and second test conditions. In some embodiments, the selected set of candidate visual acuity parameters includes a first posterior probability density function for a visual acuity threshold or visual acuity range. In some embodiments, b) includes selecting data having the first test condition. In some embodiments, the system further includes, after g), selecting data having the second test condition from the visual acuity test data of the subject; and repeating c) through g). In some embodiments, the selected set of candidate visual acuity parameters includes a second posterior probability density function for a visual acuity threshold or visual acuity range. In some embodiments, the system further includes using the first and second posterior probability density functions to obtain a difference distribution. In some embodiments, the system further includes calculating a change in visual acuity threshold or visual acuity range between the first and second test conditions based on the difference distribution. In some embodiments, the visual acuity test results are chart invariant or chart specific. In some embodiments, (e) is based on Bayes' theorem or maximum likelihood estimation. In some embodiments, the posterior probability density function is a Bayesian posterior probability density function. In some embodiments, the first or second posterior probability density function is a Bayesian posterior probability density function. In some embodiments, (d) includes: generating a single-target psychometric function comprising a first set of chart-specific visual acuity parameters; calculating a family of multiple target psychometric functions using the single-target psychometric function; and generating a chart-specific psychometric function comprising a second set of chart-specific visual acuity parameters using the family of multiple target psychometric functions.In some embodiments, the system further comprises generating a sensitivity-based psychometric function including a first set of chart-invariant sensitivity parameters prior to generating the psychometric function of the single optotype. In some embodiments, the sensitivity-based psychometric function is independent of one or more acuity chart design parameters. In some embodiments, the first set of chart-invariant sensitivity parameters includes a sensitivity threshold and a sensitivity range. In some embodiments, the sensitivity-based psychometric function is generated based on one or more acuity chart design parameters and / or one or more additional parameters based on the subject, the one or more additional parameters being chart-invariant. In some embodiments, the sensitivity-based psychometric function is generated based on signal detection theory. In some embodiments, the sensitivity-based psychometric function is configured to describe the visual acuity performance of a subject as a function d' of one or more optotype sizes and independent of one or more chart design parameters. In some embodiments, the system further comprises converting the sensitivity-based psychometric function to the psychometric function of the single optotype. In some embodiments, the set of chart-specific visual acuity parameters includes a visual acuity threshold and a visual acuity range. In some embodiments, the psychometric function of the single optotype is chart-specific. In some embodiments, the set of chart-specific visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the set of chart-specific visual acuity parameters includes one or more numerical values. In some embodiments, the set of chart-specific visual acuity parameters includes at least one posterior probability density function for that parameter. In some embodiments, the subject's visual acuity testing data includes data from at least two different test conditions. In some embodiments, the set of chart-specific visual acuity parameters includes a posterior probability density function for visual acuity threshold in at least two different test conditions.In some embodiments, the one or more eye chart design parameters include a total number of eye charts, a number of eye charts per row, a method of eye chart sampling, an eye chart size, a number of rows, and a subject's response, or a combination thereof. In some embodiments, the psychometric function of the single eye chart is based on at least one eye chart size. In some embodiments, each function of the family of psychometric functions of the multiple eye charts is based on two or more different eye chart sizes. In some embodiments, the psychometric function of the single eye chart includes a visual acuity threshold and a visual acuity range of the subject. In some embodiments, one or more functions in the family of psychometric functions of the multiple eye charts are based on the number of eye charts per row in the eye chart or sub-chart. In some embodiments, generating the chart-specific psychometric function further includes using a weighted sum of the family of psychometric functions of the multiple eye charts. In some embodiments, the weighted sum of the family of psychometric functions is based on the subject's eye chart data. In some embodiments, the weighting is determined manually by a user or automatically by a computer program. In some embodiments, (b) further includes displaying the one or more eye charts generated based on the one or more eye chart design parameters. In some embodiments, each of the one or more optotypes includes at least one optotype. In some embodiments, the one or more optotypes are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast in a standard ETDRS or Sloane chart. In some embodiments, the at least one optotype is a letter, a number, or a symbol. In some embodiments, the system further includes: h) generating a plurality of candidate optotypes based on one or more chart design parameters; and i) selecting one or more of the plurality of candidate optotypes for visual acuity testing of the subject based on a ranking of the plurality of candidate optotypes, where the ranking is based on a selected set of candidate visual acuity parameters for the subject.In some embodiments, the system further includes, after (b), collecting data for the subject using the selected one or more candidate eye charts, and if the stopping criterion is not met, repeating steps c) through g) and h) through i) until the stopping criterion is met.

[0031] In some embodiments, a computer-implemented system for adaptively evaluating the visual acuity of a subject is disclosed herein, the computer-implemented system comprises: a digital processing device, comprising an operating system and a memory, configured to execute executable instructions; and a computer program, the computer program comprises instructions executable by the digital processing device to generate a scoring application, the scoring application comprises a software module configured to: a) obtain one or more visual acuity chart design parameters; b) collect visual acuity test data of the subject; c) generate visual acuity test data based on one or more visual acuity chart design parameters of the subject and one or more sets of candidate visual acuity parameters, each of the one or more sets of candidate visual acuity parameters comprises a rank obtained by analyzing the visual acuity test data of the subject; d) generate a plurality of candidate visual acuity charts based on the one or more visual acuity chart design parameters; e) rank the plurality of candidate visual acuity charts based on the rank of each of the one or more sets of candidate visual acuity parameters; and f) select one or more visual acuity charts from the plurality of candidate visual acuity charts based on their ranks to be presented to the subject for visual acuity testing. In some embodiments, the system further comprises, after f), presenting the one or more selected visual acuity charts to the subject. In some embodiments, the system further includes, after f), updating the vision test data with new data collected from the subject using the one or more selected vision charts. In some embodiments, the system further includes presenting the subject with the one or more selected vision charts; updating the vision test data with new data collected from the subject using the one or more selected vision charts; and repeating steps b)-f); until a stopping criterion is met. In some embodiments, each set of the one or more sets of candidate vision parameters includes a vision threshold and a vision range.In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a posterior probability density function for the one or more visual acuity parameters. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a posterior probability density function for each of the one or more visual acuity parameters. In some embodiments, (b) summarizes the visual acuity testing data of the subject. In some embodiments, the visual acuity testing data of the subject includes data from only a first test condition. In some embodiments, the visual acuity testing data of the subject includes data from a first and a second test condition. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof. In some embodiments, the visual acuity testing data of the subject includes data from at least two different test conditions. In some embodiments, each set of the one or more sets of candidate visual acuity parameters includes a posterior probability density function for visual acuity threshold or visual acuity range. In some embodiments, the one or more visual acuity chart design parameters include: total number of visual acuities, number of visual acuities per row, method of visual acuity sampling, visual acuity size, number of rows, and subject response, or a combination thereof. In some embodiments, each of the candidate visual acuity charts includes at least one visual acuity chart. In some embodiments, one or more of the candidate visual acuity charts are of contrast that is about 25%, 2.5%, or 1.5% of normal contrast on a standard ETDRS chart or Sloane chart. In some embodiments, the at least one visual acuity chart is a letter, a number, or a symbol.

[0032] <Optotype> In some embodiments, the optotype is a standardized code for testing vision or visual acuity, hi some embodiments, the optotype chart or vision test is according to one or more standards selected from the list of the American National Standards Institute, the National Academy of Sciences, the National Research Council, the American Academy of Ophthalmology, the National Eye Institute, and the U.S. Food and Drug Administration.

[0033] In some embodiments, the optotype is a letter, figure, number, picture, or geometric symbol of any shape. In some embodiments, the optotype is a predetermined size. In some embodiments, the minimum size of the optotype is to generate an acuity of -0.3 logMAR. In some embodiments, the minimum size of the optotype is to generate an acuity of -0.25, -0.35, -0.4, -0.45, or -0.5 logMAR. In some embodiments, the maximum size of the optotype is to generate an acuity of 1.2, 1.25, 1.3, 1.35, 1.4, 1.45, 1.5, or 1.55 logMAR.

[0034] <Eyelash chart and sub-charts> In some embodiments, an optotype chart (or interchangeably referred to herein as a "chart", "optotype chart", or "vision chart") is used to measure the spatial resolution of a subject's visual system. In some embodiments, a vision sub-chart is a portion or part of a complete vision chart. In some embodiments, the vision chart is an optotype chart, a visual chart, a visual chart, or the like. In some embodiments, a vision chart or sub-chart includes one or more optotypes. In some embodiments, vision charts are classified into various types depending on at least one chart parameter or visual test parameter. These parameters may include one or more selected from type of optotype, number of optotypes, number of rows, number of columns, optotype size, optotype size difference between adjacent rows or columns, number of charts in a visual test, optotype sampling method, and the like. In some embodiments, the vision chart is an ETDRS (Early Treatment of Diabetic Retinopathy Study) logMAR (X and Y axes are logarithms of the minimum angle of resolution) chart, a Snellen chart, a tumbling E chart, a Landolt c chart, a Lea test, or a Jaeger chart.

[0035] In some embodiments, the visual chart or sub-chart includes one or more optotypes. In some embodiments, at least one optotype forms the visual chart or sub-chart. In some embodiments, at least one row of optotypes, each row containing at least one optotype, forms the visual chart or sub-chart. In some embodiments, at least one column of optotypes, each column containing at least one optotype, forms the visual chart or sub-chart. In some embodiments, the various optotypes in each column are of the same optotype size. In some embodiments, the optotype size decreases monotonically from the first column to the last column of the visual chart or sub-chart. In some embodiments, the number of optotypes in each column of the visual chart or sub-chart is the same. In some embodiments, each row has five optotypes. In some embodiments, the number of optotypes in each column of the visual chart or sub-chart ranges from 1 to 5 optotypes. In some embodiments, the chart or sub-chart has three columns of optotypes. In some embodiments, the chart or sub-chart has one to five columns of optotypes. In some embodiments, the complete visual chart has 14 rows of optotypes.

[0036] In some embodiments, the optotype size for each row is specified as the logarithm of the decimal visual acuity to base 10. As an example, a decimal visual acuity of 1 is 0.00 in LogMAR visual acuity, and a decimal visual acuity of 0.1 is 1.0 in LogMAR visual acuity.

[0037] <Visual acuity chart design> In some embodiments, the methods, systems, or devices disclosed herein include an eye chart and sub-charts, hi some embodiments, the eye chart or sub-charts are constrained by one or more specific chart design features and / or optotype design features.

[0038] In some embodiments, design features of the optotypes are taken into account in the systems and methods disclosed herein in addition to design features of the eye chart (or sub-chart) used to present the optotypes. In further embodiments, parameters (interchangeable as features) of the optotypes and / or the charts on which they are presented include, but are not limited to: i. The number of possible target sizes (e.g., 11 target sizes for the Snellen chart and 14 target sizes for the ETDRS chart); ii. Possible sizes of targets (e.g., Figures 11A-11F) ii. Identity of target: 1. Characters 2. Numbers 3. Pattern 4. Landolt C 5.Tumbling E; iii. The size of the pool for sampling the targets 1. Sloan, Bailey-Lovie have 10 2.4 for Landolt C, Tumbling E, HOTV, Lea, Hands 3. Traditional Snellen chart has 9 4. The British Snellen standard BS 4274-1:2003 has 12; 5. The English alphabet has 26 iv. The method of sampling the pool of targets (e.g., with or without replacement); v. The number of lines presented on a chart, subchart, or set of subcharts; vi. the number of targets presented per line; vii. Definition of optotype metrics (e.g., optotype threshold, optotype range, and / or optotype gradient);

[0039] Existing techniques in clinical care focus on the nominal goal of visual acuity testing: determining the rows of optotypes where the patient cannot reliably identify the optotype at some pre-specified criterion: e.g., optotype sizes where the patient cannot identify at least two of five optotypes. Current tests ignore some design factors that contribute to visual performance across various chart designs. For example, a small pool of optotypes makes it easier to guess accurately in the absence of visual information. Specifically, when testing visual acuity with the Landolt C optotype (distinguishing one of four potential orientations), it is easier to guess accurately (25% probability) compared to visual acuity charts that have lower guess rates (e.g., 10% for the Sloan set). Currently, visual acuity results are not adapted or corrected for guess rates in various charts. There may be a confusion of performance contributions arising from chart-specific design factors and patient-specific factors.

[0040] In some embodiments, a comprehensive consideration of one or more of the eye chart design features, including chart-specific parameters and subject-specific parameters (corresponding to the chart-invariant parameters herein), is disclosed herein and forms the basis of the visual acuity model or predictive model herein. By using the eye chart design to precisely define the contribution to performance and by considering the various contributions to performance of each chart, the model may allow the measurement of the visual acuity performance of the subject independent of the chart used to collect the visual acuity data.

[0041] In some embodiments, the visual acuity test conforms to design principles based on committee standards and / or historical and cultural preferences. In a particular embodiment, referring to FIG. 2A, the visual acuity chart design recommended by the ETDRS committee includes 14 rows of symbols, with 5 symbols per row, each with a logarithmic decrease in size down each column. A progression of 0.10 decimal logarithmic units results in approximately a 25% decrease in size for each successive column of symbols. Each column includes 5 letters sampled from a Sloan set of 10 letters without replacement. In this embodiment, a visual acuity chart that meets the ETDRS criteria is shown in FIG. 2A.

[0042] In some embodiments, the Snellen eye chart (FIG. 2B), which presents more letters at smaller target sizes or fewer letters at larger target sizes, remains the dominant chart used in clinical practice. In some embodiments, ETDRS, Snellen and similar standard-constrained chart designs are associated with a static range and resolution for sampling targets. Due to the non-adaptive deterministic testing practice, all subjects are presented with the same range of targets. Furthermore, the invariant sampling resolution of the test results in insensitivity to measurements of visual acuity values ​​that correspond to intermediate target sizes compared to those presented at the 14 established target sizes.

[0043] In some embodiments, the eye chart may conform to design principles determined by standards committees and / or historical preferences. For example, the ETDRS chart is the eye chart recommended for clinical trials and clinical research (FIG. 2A). The design of the ETDRS chart is 14 rows of symbols, each row presenting 5 symbols sampled without replacement from the 10-letter Sloan set, with a constant logarithmic progression of symbol size between rows. A progression of 0.10 decimal logarithmic units corresponds to an approximately 25% increase in size from each row to the row above.

[0044] The design of the Snellen eye chart (Figure 2B) contains 12 rows with different numbers of optotypes per row and different size progressions between the rows. Despite design features that are considered inferior to those of the ETDRS chart, the Snellen eye chart remains the dominant eye chart design used in clinical practice by ophthalmologists and general practitioners.

[0045] In some embodiments, a method for generating an expanded library of optotype charts is used to improve the sampling resolution for optotype chart sizes in a visual acuity test while closely adhering to visual acuity design standards and principles. In a particular embodiment, referring to FIG. 10, the top pattern (1001) represents the sampling scheme represented by the ETDRS standard chart, which presents 14 standard sizes of optotypes from -0.3 to 1.0 logMAR, with a logarithmic unit sampling resolution of 0.10 between successive sizes. The anchor point (1002) of 0.35 represents the average optotype size of the ETDRS standard chart. In some embodiments, modification of the anchor points (1002) of the standard constrained chart is used to generate a larger complementary set of optotype charts that improves the aggregate resolution for sampling optotype sizes while maintaining adherence to the optotype chart standards. In this particular embodiment, a set of 10 optotype charts is generated (1003) generated by modifying the anchor points of the standard chart within the range of 0.3 to 0.4 logMAR, with an anchor point resolution of 0.01 logMAR. In some embodiments, each test slide uses the same sampling resolution as the ETDRS standard. However, when the set of design-constrained charts is considered together, these charts provide a high resolution of the target size represented by the bottom pattern, which is 10 times the ETDRS standard resolution. Taken together, in this embodiment, the set of 10 charts is developed to assess vision across a wide range of 141 different target sizes at 0.01 log unit resolution.

[0046] In some embodiments, for standard ETDRS charts that follow the design principle of log-linear, line-by-line progression of optotype sizes, the "anchor point" is the eye chart size that corresponds to the average (or median) logMAR size of the 14 optotype sizes that make up the complete chart. In the case of the ETDRS chart, the anchor point corresponds to an optotype size between the 20 / 40 line and the 20 / 50 line (in Snellen notation).

[0047] In some embodiments, referring to Figures 11A-11F, an exemplary subset of design-constrained eye charts are shown. In these embodiments, the eye charts are generated using ETDRS design principles (Figures 11A-11C) with three different anchor points (i.e., 0.30, 0.35, and 0.40 logMAR) and Snellen design principles (Figures 11D-11F) with three different anchor points (i.e., 0.30, 0.35, and 0.40 logMAR). In these embodiments, the eye charts can optionally have a fixed or variable number of optotypes per column.

[0048] In some specific embodiments, referring to Figures 12A-12C, the test slide is focused on a specific area of ​​the complete standard constrained adaptive test chart. The complete design-constrained eye chart is subsampled by presenting only a limited set of the optotypes of the complete eye chart, which is a subsample of a much larger library of design-constrained charts. Presenting a set of subcharts, rather than presenting a full row of one ETDRS chart as in Figure 2A, focuses the visual test on a subsample of the space of the complete chart (e.g., three rows of optotypes). The target letters of the focused visual test are identified by presenting only the subsampled area of ​​the complete chart (Figure 12A), presenting the subsampled area with high contrast compared to the low contrast of the rest of the chart (Figure 12B), or by showing only the subsampled area without blur (Figure 12C) and blurring the rest of the chart.

[0049] In current technology, the analysis of visual acuity tests is based on heuristics (line assignment or letter-by-letter) that provide results that are inaccurate and difficult to reconcile across different charts.

[0050] In some embodiments, the systems or methods disclosed herein provide a general statistical and computational framework for the accurate testing of visual acuity and changes therein, as measured using ETDRS, Snellen, or charts of any design.

[0051] In some embodiments, for visual acuity testing, a response from the subject follows presentation of the test slide to the subject. In some embodiments, the response from the subject may include identification of one or more objects and / or identification of one or more features of the object. In some embodiments, referring to FIGS. 13A-13C, the subject responds by reading down the first column (FIG. 13A) until instructed to stop. In some embodiments, the subject responds by reading across at least one row (FIG. 13B) until instructed to stop. In some embodiments, the subject reads a single letter that is presented (FIG. 13C). In some embodiments, the object is either solid black or gray. In some embodiments, gray objects show a lower preference than solid black objects.

[0052] In some embodiments, a library of optotype charts is generated that exhibits high precision for sampling optotype sizes to improve the accuracy of vision testing. Individually, these charts have similar sampling precision for sampling optotype sizes as existing optotype charts, and their appearance conforms to design principles recommended by standards committees for optotype charts (e.g., ANSI, ISO, NSC / NAS). Collectively, however, this expanding library may exhibit finer resolution for sampling optotype chart sizes than existing optotype charts (see Figures 2A and 15). This library, or similar libraries, may include a wide range of design features for optotype slides, including, but not limited to: (1) slides with many rows that match the full ETDRS chart and the Snellen chart; (2) slides that sample the full ETDRS chart and present only 1-5 rows at a time; and (3) slides with a single optotype. Similarly, the library may further include a high precision family of charts motivated by the Snellen design.

[0053] In some embodiments, the optotypes are randomly sampled from a library of optotypes. In some embodiments, the optotypes are adaptively sampled from a library of optotypes. In some embodiments, the adaptive sampling of the optotypes is based on a priori information of the subject's visual acuity. In some embodiments, the adaptive sampling is based on luminosity parameters (i.e., visual acuity threshold and / or visual acuity range and / or visual acuity slope) estimated from previous tests. In other embodiments, the adaptive sampling is based on the results of previous tests, the patient's medical records, and the subject's responses in previous tests or test tests. In some embodiments, the adaptive sampling is to optimize the optotype size to estimate visual acuity parameters quickly and accurately. For example, if in a first test chart, subject A correctly identifies all three columns with five optotypes in each column, and subject B correctly identifies five, four, and four optotypes in each column, respectively, then in a second test chart, subject A is presented with an optotype that is smaller in average size than the optotypes presented to subject B.

[0054] In some embodiments, the eye chart and / or sub-chart provides 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or even 20 times higher resolution than the standard resolution of existing eye charts, e.g., the ETDRS standard resolution.

[0055] In some embodiments, the Snellen chart remains the dominant visual acuity test in clinical practice, while the ETDRS chart serves as the standard for ocular spatial clinical research and testing. The ETDRS chart can present 14 rows of optotypes (see FIG. 2A): a minimum optotype size of logMAR=-0.3, a maximum optotype size of 1.0, and intermediate sizes arranged in a constant log-linear progression (e.g., log units of .10). These design principles may be determined by a standards committee. One drawback of these existing chart designs is the lack of flexibility of static, predefined, non-adaptive testing. Testing all patients in the same way (using a limited number of optotype sizes available on a single chart) can make it difficult to accurately assess a wide range of vision in a population (from healthy vision to visual impairment). To improve the flexibility and accuracy of vision testing, and to better adapt the test to various levels of healthy vision and visual impairment, current technology is moving towards computerized vision testing with digital displays. Currently, adaptive testing algorithms focus on changing optotype size for one letter based on correct or inaccurate responses. These existing algorithms use simple heuristics that are limited to single-character response outcomes rather than more complex multi-character response outcomes.

[0056] The design features of various visual acuity tests are determined by external or internal guidelines or design principles. External guidelines reflect recommendations issued by ANSI or ISO committees, while internal guidelines may reflect corporate or cultural preferences. External guidelines can be viewed as mandatory, while internal guidelines can be viewed as preferences. Both types of guidelines also limit the appearance and implementation of visual acuity tests. Currently, single-letter visual acuity tests do not conform to the design principles and guidelines recommended by ANSI and ISO committees. As an example, consider a standard ETDRS chart with an anchor point at 0.35 logMAR, with a logarithmic unit resolution of 0.01, and a set of visual acuity charts generated by 10 other anchor points ranging from 0.3 to 0.4 logMAR inclusive. (Higher resolution for sampling of anchor points is possible, but the benefits of higher resolution are ultimately limited by the print resolution of a paper chart or the pixel resolution of a digital display.) The generation of these 10 charts (in addition to the ETDRS standard), each consisting of 14 target sizes, provides a precise coverage of 141 different targets ranging from -0.4 to 1.1 logMAR. Due to the chart design criterion of 0.10 log unit progression between rows of the same chart, the complete set of target sizes represented by these complementary charts can be represented as a subset of target sizes linearly inserted into logarithmic space.

[0057] In some embodiments, the development of an eye chart that maintains the design constraints of the ETDRS chart but uses different anchor points generates a library of new charts with a much larger set of eye size compared to the standard ETDRS chart (see FIG. 2A). The new charts defined to meet the ETDRS chart design criteria appear to be largely the same as the original chart, but provide improved range and resolution for sampling different eye sizes. In some embodiments, the different anchor points result in slightly different maximum and minimum sizes (FIGS. 11A-11C), but the standard constrained size progression between adjacent columns is maintained. Similarly, a library of new charts can be generated with a much larger set of eye size compared to the standard Snellen eye chart.

[0058] The use of a set of anchor points (e.g., anchor points that are not limited by an existing optotype chart) can generate an alternative set of standard constrained charts that provide a more flexible sampling resolution of optotype sizes. Considering sub-charts with fewer rows can further expand the library of full 14-row optotype charts. These libraries of optotype charts can be generated by subsampling the chart (e.g., only columns 1-5 can be presented at a time). Spatial subsampling of the full chart focuses on a subset of the chart, e.g., rows 1-5 of optotypes rather than the 14 rows of the full chart. In some embodiments, the top and / or bottom of the chart represent relatively useless areas that can be avoided during testing. Testing on these rows may result in accurate reporting of all optotypes or none at all. A better sampling of the full chart would include optotype sizes with a variety of possible intermediate results (e.g., 2, 3, or 4 of the 5 optotypes are correct). The complementary chart sets can represent stimulus sets that can be sampled once or repeatedly by deterministic, randomized, and / or optimized sampling. A sampling of the standard ETDRS eye chart is shown in Figures 11A-11C. The size progression between the rows is constant in logarithmic space. The anchor points of these sizes can be selected as the average in logarithmic space. In this embodiment, seven sizes are greater than the selected anchor point, and seven sizes are less than the selected anchor point.

[0059] In some embodiments, the libraries of charts disclosed herein exhibit design features that provide a greatly expanded set of optotype combinations within the limited set of 14 sizes used by the ETDRS standard charts. Different libraries may be generated specific to each chart and its design principles (e.g., ETDRS, Snellen, HOTV, Lea's code, etc.).

[0060] In the prior art, despite the emerging use of ETDRS in clinical research and trials, the Snellen chart has remained the dominant visual acuity test in clinical practice for 150 years. Due to practical needs, methods for calculating visual acuity metrics rely on mental scoring heuristics that are explicitly linked to letters and design principles that are obvious to the test administrator. The two dominant approaches to assess visual acuity are line-by-line assignment or letter-by-letter. For example, in the Snellen test, the line-assignment method calculates a final visual acuity estimate as the optotype size corresponding to the last line in which the patient correctly reports three of five optotypes. To compare with line-assignment, the ETDRS chart uses letter-by-letter scoring, where each correctly reported letter is given 0.02 logMAR units.

[0061] Unlike the prior art, which only considers a single psychometric function that defines the probability of correctly identifying a single optotype as a function of its size (sometimes also known as a "vision curve"), the methods and systems herein involve analyzing responses to assess visual acuity by considering a family of psychometric functions that define the probability of observing a total constant combination of correct responses for multiple optotypes as a function of the optotype (see Figures 4-9). In the example of an ETDRS chart that presents five optotypes per row, there are six complementary psychometric functions that describe the probability of observing correct reports of 0, 1, 2, 3, 4, or 5 optotypes out of the total of five presented optotypes (Figures 6B, 8B). Measuring six psychometric functions using the prior art only increases the problem of measuring a single psychometric function. Practical limitations on clinical experiment time similarly make it impossible to use the prior art to assess visual acuity metrics using these chart-specific empirically observed six psychometric functions.

[0062] <Visual acuity parameters> Visual acuity may be visual acuity or clarity or resolution. In some embodiments, visual acuity is represented by at least one psychometric function. In some embodiments, visual acuity is represented by at least one psychometric function, either implicit or explicit. In some embodiments, the psychometric function is chart-specific or chart-invariant. In some embodiments, visual acuity is represented by at least one sensitivity (d') psychometric function. In signal detection theory, the sensitivity (d') parameter can represent behavioral performance independent of task-based factors that affect response bias or decision criteria. In some embodiments, the sensitivity parameter in a visual acuity test determines visual function independent of the specific design features used in the visual acuity test. In some embodiments, the psychometric function is uniquely defined by at least two visual acuity parameters. In some embodiments, the two visual acuity parameters form a two-dimensional parameter space. In some embodiments, visual acuity is represented by at least one single-target psychometric function and / or multiple-target psychometric function.

[0063] In some embodiments, visual acuity is determined or measured using at least one visual acuity parameter. The visual acuity parameters disclosed herein are interchangeable with visual acuity model parameters or visual acuity metrics. In some embodiments, the visual acuity parameter is a visual acuity threshold. In some embodiments, the visual acuity parameter is a visual acuity range. In some embodiments, the visual acuity parameter is a visual acuity slope. In some embodiments, visual acuity is determined or measured using two visual acuity parameters. In some embodiments, the two visual acuity parameters are a visual acuity threshold and a visual acuity range. In some embodiments, the two visual acuity parameters are a visual acuity threshold and a visual acuity slope.

[0064] In some embodiments, the visual acuity parameters herein include probability density functions, cumulative probability density functions, or probability distribution functions of visual acuity thresholds, visual acuity ranges, visual acuity slopes, sensitivity thresholds, sensitivity ranges, sensitivity slopes, or combinations thereof. In some embodiments, the visual acuity parameters herein include probability density functions, cumulative probability density functions, or probability distribution functions of changes in visual acuity thresholds, visual acuity ranges, visual acuity slopes, sensitivity thresholds, sensitivity ranges, sensitivity slopes, or combinations thereof between two different test conditions. In some embodiments, the probability distribution functions and probability density functions are interchangeable herein.

[0065] For example, the density threshold can include a probability density function (pdf) over a range of target sizes, the pdf including the probability of different target sizes just detectable to the subject. In some embodiments, the target threshold includes a range of different target sizes, the range including target sizes just detectable to the subject. In some embodiments, the visual acuity threshold includes a target size value and size variation, the target size being the smallest size detectable to the subject.

[0066] In some embodiments, the visual acuity parameters herein include sensitivity (d') thresholds configured to provide visual performance criteria, similar to visual acuity thresholds. The visual acuity thresholds determined by chart-invariant psychometric functions may be thresholds based on sensitivity (d'). In some embodiments, the visual acuity thresholds can be defined in chart-invariant domains represented by chart-invariant psychometric functions, e.g., psychometric functions based on sensitivity (d'). Alternatively, the visual acuity thresholds can be defined by chart-specific psychometric functions, e.g., empirical psychometric functions that depend on factors of the visual acuity chart design. There may be a directional transformation between the visual acuity thresholds defined by chart-specific and chart-invariant psychometric functions.

[0067] With reference to Figures 4A-4B and / or 7A-7B, in certain embodiments, the thresholds (401), (701) are shown using different coordinates, and the sensitivity threshold (401) and the acuity threshold (701) may be in different units. As shown in Figure 7A, in some embodiments, the acuity threshold (701) represents the target size that corresponds to a pre-specified standard of visual function: for example, the size that corresponds to the probability of correctly identifying a target of a certain size, or the size that corresponds to the correct reporting of three targets out of five targets in a row of the ETDRS chart. In the same embodiment, a second important metric for describing the performance of the visual acuity chart is the sensitivity slope (402) (Figure 4B) or acuity slope (702) (Figure 7B), which describes how sharply the chart-specific psychometric function changes around the thresholds (401) and (701). In some embodiments, the slope represents how quickly visual acuity changes with changes in target size. In some embodiments, the acuity slope (702) or sensitivity slope (402) is inversely related to the acuity range, which represents the width of the dynamic range of vision function. In this case, the acuity range (703) or sensitivity range (403) determines the logarithmic difference between two optotype sizes that correspond to the expected performance level of correctly recognized 1.5 and 4.5 letters. A difference in optotype size of 0.40 logMAR (visual acuity range in FIG. 4B) corresponds to a visual acuity range equivalent to four rows of the ETDRS chart. In this case, subjects reading the ETDRS chart move from the best average performance (five correct answers out of five on the higher rows) to the worst average performance (half correct answers out of five on the lower rows) across the range of four rows of the ETDRS chart. The algorithm S (scoring algorithm) can generate visual acuity thresholds and visual acuity ranges as key assessments of vision. In some embodiments, the visual acuity thresholds can provide the primary metric or parameter, but recognizing changes in visual acuity range / visual acuity slope can provide a secondary assessment or parameter useful in characterizing vision loss in eye disease. The power of the visual acuity model may include reducing the rich and complex data collected from visual acuity tests into a simpler prediction of vision using two parameters.

[0068] In some embodiments, the visual acuity parameters herein include visual acuity threshold, visual acuity range, visual acuity slope, sensitivity threshold, sensitivity range, sensitivity slope, or combinations thereof, at one or more test conditions. In some embodiments, the visual acuity parameters herein include the change in visual acuity threshold, visual acuity range, visual acuity slope, sensitivity threshold, sensitivity range, sensitivity slope, or combinations thereof between two test conditions. In some embodiments, the visual acuity parameters herein are numerical values ​​in units reflecting the size of the optotype, e.g., 20 / 20, 20 / 40, 3 / 5 (three out of five), 1 / 5 (one out of five), or 0.3 logMAR. In some embodiments, the visual acuity parameters herein correspond to predetermined levels of visual performance defined in chart-invariant conditions (sensitivity) or chart-specific conditions, respectively, thus generating sensitivity-based visual acuity parameters (e.g., sensitivity threshold and sensitivity range) and visual acuity parameters (e.g., visual acuity threshold and visual acuity range). In some embodiments, referring to FIGS. 4A-4B, the chart-invariant visual acuity threshold can be the target size (e.g., 0.3 logMAR corresponds to a pre-specified sensitivity level of d'=1 or d'=2). In the same embodiment, the chart-invariant visual acuity range can be the difference between two target sizes corresponding to low sensitivity (e.g., -0.3 logMAR with d'=0.5) and higher sensitivity (e.g., 1 logMAR with d'=4). Referring to FIGS. 7A-7B, which show chart-specific psychometric functions, the chart-specific threshold can be the target size, e.g., 0.3 logMAR corresponds to a pre-specified number of predicted correctly reported letters out of 5 letters (e.g., 3 out of 5 or 3 / 5, or 2, 4, or 3.5 out of 5). In the same embodiment, the visual acuity range can be the difference between two optotypes corresponding to a small number of correctly reported letters (e.g., -0.3 logMAR for 1 out of 5) and a large number of correctly reported letters (e.g., 0.9 logMAR for 4 out of 5).

[0069] In some embodiments, visual acuity change, e.g., change between two different test conditions, is measured using the systems and methods herein, e.g., by analysis of a Bayesian posterior probability density function (pdf). A Bayesian pdf for visual acuity difference between two test conditions can be generated from the individual pdfs of the test conditions for contrast sensitivity (Hou et al. 2016, which is incorporated herein by reference in its entirety):

[0070]

number

[0071] In certain embodiments, referring to Figures 27A-27B, Bayesian posterior probability density functions of visual acuity thresholds under various test conditions 1, 2, 3, and 4 are generated. In this embodiment, the probability of visual acuity change between two test conditions can be calculated by the relative overlap of the Bayesian posterior probability density functions measured under various test conditions. As shown in Figure 27A, low variability, e.g., width of the probability density function, in the visual acuity threshold estimates of the Bayesian posterior probability density functions provides confidence regarding visual acuity change between test 1 and test 2 reflected in the minimal overlap (2710a) of the posterior probability density functions of visual acuity thresholds collected under various conditions. Referring to Figure 27B, in some embodiments, the average difference in visual acuity between test conditions (2810) is the same, but there is a greater variation, e.g., width of the probability density function, in the visual acuity threshold estimates, represented by an increased variance and an increased overlap (2710b) of the Bayesian posterior probability density functions (pdf).

[0072] Calculation of visual acuity change or visual acuity difference by Bayesian signal detection analysis as described herein may include generation of a difference distribution between visual acuity values ​​obtained at different conditions (2810), such as before or after a medical intervention (e.g., Bayesian posterior probability density function for test 1 through Bayesian pdf for test 4 shown in (2811) of Figures 28A-28B). Following collection of the posterior probability density functions for each test condition, generation of the difference distribution (2811) provides an equivalent index of visual acuity threshold change as shown in Figure 28B.

[0073] In some embodiments, generating a difference distribution of the Bayesian posterior probability density function of two test conditions, e.g., Test 1 and Test 2, provides an estimate that the change between conditions is greater than a threshold reference level of visual acuity change. For example, the threshold reference level may be one or more rows of a given visual acuity chart, or two or more rows of a given visual acuity chart, or three or more rows of a given visual acuity chart.

[0074] In some embodiments, instead of testing high-contrast visual acuity with a black target on a white background, low-contrast visual acuity can also be tested with a gray-scale target that shows a smaller difference in brightness from a gray-scale background. For example, using the design of the Sloane low-contrast chart. In some embodiments, the visual acuity chart used for testing visual acuity with low contrast looks the same as ETDRS, but various charts can present visual acuity with contrasts of about 25.0%, 2.5%, and 1.5% calculated against the standard background of the visual acuity chart. In some embodiments, the low-contrast visual acuity chart has a contrast of about 60%, 50%, 40%, 30%, 25%, 24%, 20%, or even lower than the contrast of the standard ETDRS chart and / or Sloane visual acuity chart by 1 or more. In some embodiments, the low contrast chart has a contrast that is about one or more orders of magnitude lower than or equal to 25%, 20%, 18%, 15%, 12%, 10%, 8%, 5%, 2.5%, 2%, 1.5%, or even lower than the contrast of a standard ETDRS chart and / or Sloane chart.

[0075] In some embodiments, where a more complex visual acuity model is used, the difference distribution can alternatively be estimated directly, for example using two visual acuity thresholds and two visual acuity ranges.

[0076] In some embodiments, the algorithms S and A can be used to directly measure the change in the visual acuity model that reflects the visual acuity change or visual acuity difference. In some embodiments, the two-dimensional visual acuity model determined by using the two visual acuity parameters herein as two dimensions can be elaborated into a model with higher dimensions (e.g., three, four, five, or even higher dimensions). For example, the two-dimensional visual acuity model can be used to extend into a four-dimensional model in two ways: (a) measuring visual acuity threshold and visual acuity range in two test conditions; then, algorithm A can select the optimal stimulus and optimal condition to test on a trial-to-trial basis; (b) measuring visual acuity threshold and visual acuity range in one condition, and the difference between the threshold and range (i.e., δ threshold, δ range) as a second set of parameters. For example, the visual acuity model can assume that there is no difference in the expected visual acuity range between the two conditions. In other words, the visual acuity difference between the two test conditions can be characterized as a lateral shift of the chart-specific visual acuity function without a change in the visual acuity range parameter.

[0077] In some embodiments, the results from the visual acuity model analysis are evaluated to determine the probability of a change in visual acuity threshold, visual acuity slope, and / or visual acuity range.

[0078] In some embodiments, the visual target model herein is configured to directly infer visual target changes. Such visual acuity changes can be changes between two test conditions or changes between more than two test conditions. In some embodiments, the different test conditions can include any differences in the subjects. It can be, for example, before or after a medical intervention. Another example is testing a diseased eye and a normal eye, or testing a left eye and testing a right eye. In some embodiments, the visual acuity model herein can estimate visual acuity thresholds, visual acuity ranges, and changes in visual acuity thresholds and visual acuity ranges. In some embodiments, the visual acuity parameters include additional parameters that determine the visual acuity thresholds and visual acuity ranges from a first test condition and a second test condition different from the first test, as well as the difference between those parameters (e.g., δ visual acuity threshold between the first and second test conditions). In some embodiments, the first condition is the left eye and the second condition is the right eye. In some embodiments, the first condition is before treatment and the second condition is after treatment. In some embodiments, the first condition is measured at a lower luminance and the second condition is measured at a higher luminance. In some embodiments, the first condition is measured at a lower contrast and the second condition is measured at a high contrast.

[0079] Summarizing eye chart data. In some embodiments, the systems and methods herein include eye chart data obtained by collecting subjects' responses to presented eye charts or sub-charts. Eye chart data is interchangeable herein as vision test data and / or responses. In some embodiments, the subjects' responses are summarized, for example, in a data chart. In the case of a single eye, the response (r) can be the accuracy of a single response for that eye. In the case of a row of multiple eyes, the composite response (r) summarizes how many eyes (m) are correctly recognized out of a total of n eyes presented per row. Since 0 to n eyes (out of n) per row can be accurately reported, there can be n+1 potential outcomes for the composite response.

[0080] In some cases, design features of the visual acuity chart can be determined and chart design parameters can be summarized, as shown in operations (100) and (400) of Figure 1. In some embodiments, data charts herein (e.g., Table 1) co-locate parameters that define the chart design and parameters of subject response that are important to the application of the visual acuity model.

[0081] Table 1 shows an exemplary summary of eye chart data collected from subjects.

[0082] [Table 1]

[0083] In Table 1, each column can define a single optotype condition consisting of a single optotype or optotype size of a single row of optotypes, as well as the corresponding single or compound response to the presented optotypes. In this particular example shown in Table 1, the optotype data collected from the subjects is summarized. In order to apply algorithm S to score the optotype data, this summary can assist in identifying design features that contribute to the visual function observed during the optotype test. In this summary table, each row entry indicates reading a single optotype or a single row of multiple optotypes from the optotype chart or sub-chart. Each row of multiple optotypes can be considered as a compound stimulus, which, in addition to the important feature of optotype size, is described by: the size of the subset of sampled optotypes in each row, the size of the superset of optotypes from which they are sampled, and whether the sampling from the superset is done with or without replacement. For each column of chart design parameters in the summary table, the response summary represents the number of optotypes correctly recognized, whether it is a single optotype or a row of multiple optotypes. In some embodiments, the optotype parameters shown in Table 1 are required to generate the chart-specific psychometric functions because they can have a significant effect on the chart-specific psychometric functions.

[0084] In some embodiments, the vision test data may include test data from 1, 2, 3, 4, 5, or even more test conditions.

[0085] Candidate Parameters for Visual Acuity Model In some embodiments, one or more sets of candidate parameters can be generated using the systems and methods herein. A set of candidate visual acuity model parameters may include a combination of two or more visual acuity parameters, such as visual acuity threshold and visual acuity range. Various combinations of visual acuity parameters can create multiple sets of candidate parameters. Such sets of candidate parameters can be configured to describe summarized visual acuity chart data, as shown in Table 2 below. In operation (300) of FIG. 1, such candidate parameters, such as those shown in Table 2, can be at least partially determined by visual acuity chart data or visual acuity test data collected from the operation to be tested (400). Alternatively or in combination, such candidate parameters can be at least partially determined by visual acuity chart design parameters determined in operation (100). In some embodiments, operation (300) may not depend on the design parameters of operation (100). In some embodiments, operation (300) occurs before operation (200). Each set of candidate parameters of operation (300) is also used in operation (200), e.g., FIG. 3A and / or FIG. 3B in determining the general visual acuity model. In some embodiments, operation (300) is performed in parallel or sequentially with operation (400) and / or operation (100).

[0086] Table 2 shows candidate sets of visual acuity model parameters. These sets of visual acuity threshold and visual acuity range parameters include visual acuity model parameters that can potentially describe the visual acuity chart data collected from the subject. Each entry column reflects a combination (out of a total of q) of a candidate visual acuity threshold parameter of q1 and a single candidate visual acuity range / slope of q2. In this embodiment, the table has a number of combinations of q1 x q2.

[0087] [Table 2]

[0088] In some embodiments, the observed eye chart data is analyzed to infer the visual acuity model parameters that best predict the subject's visual acuity. In some examples, the set of visual acuity model candidate parameters is a list of parameter combinations of visual acuity threshold and visual acuity range. Statistical inference can describe the probability of each parameter combination describing an observer. This is equivalent to evaluating the probability of class membership. In other cases, the set of candidate parameters includes a two-dimensional space of visual acuity parameters - visual acuity threshold and visual acuity range - that provides a comprehensive combination of parameters to describe the subject's visual acuity performance. In this approach, marginal probabilities can be calculated over the two dimensions of visual acuity parameters.

[0089] The set of candidate visual acuity parameters can be implemented into a visual acuity model for visual acuity analysis. For example, the set of candidate visual acuity parameters can be a one-dimensional array of pairs of visual acuity threshold ranges. As another example, the set of candidate visual acuity parameters can be two-dimensional and can be used to generate a visual acuity model that describes a comprehensive combination of visual acuity thresholds and visual acuity ranges. As yet another example, the set of candidate visual acuity parameters can be four-dimensional: visual acuity thresholds and visual acuity ranges for baseline conditions, (e.g., left eye or treatment conditions), and changes in visual acuity thresholds and visual acuity ranges for those conditions.

[0090] In some embodiments, each set of visual acuity parameters may include 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or any other integer number of visual acuity parameters.

[0091] In some embodiments, there may be a set of 1, 2, 3, 4, 5...1000...10000, or any other integer number of candidate visual acuity parameters.

[0092] In some embodiments, each set of candidate visual acuity parameters includes an equal number of parameters. In some embodiments, each set of candidate visual acuity parameters includes one or more of the following: acuity threshold, acuity range, acuity slope, a change in acuity threshold between two test conditions, acuity range, or acuity slope, at one or more test conditions.

[0093] In some embodiments, each set of candidate visual acuity parameters includes one or more of the following: a sensitivity threshold, a sensitivity range, a sensitivity slope, a change in sensitivity threshold between two test conditions, a sensitivity range, or a sensitivity slope. In some embodiments, the candidate visual acuity parameters herein include a visual acuity threshold, a visual acuity range, a visual acuity slope, a sensitivity threshold, a sensitivity range, a sensitivity slope, or a combination thereof, in one or more test conditions. In some embodiments, the candidate visual acuity parameters herein include a change in visual acuity threshold, a visual acuity range, a visual acuity slope, a sensitivity threshold, a sensitivity range, a sensitivity slope, or a combination thereof, between two test conditions. In some embodiments, the candidate visual acuity parameters herein are numerical values ​​in units reflecting the size of the target, e.g., 20 / 20, 20 / 40, 3 / 5 (three out of five), 1 / 5 (one out of five), or 0.3 logMAR. In some embodiments, each of the candidate visual acuity parameters herein corresponds to a predetermined level of visual function defined in chart-invariant terms (sensitivity) or chart-specific terms, thus generating sensitivity-based visual acuity parameters (e.g., sensitivity threshold and sensitivity range) and visual acuity parameters (e.g., visual acuity threshold and visual acuity range). In some embodiments, the candidate visual acuity parameters defined in chart-invariant terms (sensitivity) or chart-specific terms can be converted to each other. In other words, each set of chart-invariant sensitivity parameters can be converted to a set of chart-specific visual acuity parameters using the same optotype information (e.g., optotype size (e.g., 0.3 logMAR) and sensitivity or visual acuity condition (e.g., d'=2, or 3 out of 5) related to visual acuity performance). In some embodiments, the candidate visual acuity parameters are interchangeable with the candidate visual acuity model parameters or candidate parameters.

[0094] <Visual acuity model> In some embodiments, a method, system, or device disclosed herein includes one or more visual acuity models (interchangeably referred to herein as "statistical models" or "predictive models"). Such terms are equivalent and interchangeable herein. In some embodiments, a visual acuity model is used to predict a subject's visual acuity chart data (e.g., the probability of observing a particular response from a subject), said visual acuity chart data indicative of the subject's visual acuity performance. In some embodiments, such predictions using a visual acuity model are based on the design features of the visual acuity chart used to collect the visual acuity chart data from the subject. In some embodiments, such predictions by a visual acuity model are invariant to the design features of a particular visual acuity chart.

[0095] In certain embodiments, referring to FIG. 1, the visual acuity model can be determined using operations (200), (300), and / or (350). In some embodiments, the visual acuity model herein includes model parameters. Such model parameters can include: (a) visual acuity chart design parameters that are fixed and at least partially determined by the visual acuity chart and at least partially determined by operation (100); (b) a set of candidate visual acuity parameters describing the patient's vision that can be at least partially determined in operation (300). In operation (300), a set of candidate visual acuity parameters used to add to the visual acuity model is estimated. In operation (200), each set of estimated sets of parameters in operation (300) is used to generate chart-specific features in operation (200) and FIG. 3A and / or FIG. 3B, possibly in combination with the visual acuity chart design parameters. The chart-specific features are then used in operation (350) to add to the visual acuity model used in operation (500). In some embodiments, the function of the visual acuity model is to output a prediction of the visual acuity chart data or test data (e.g., response probability) using the visual acuity chart design parameters (in operation (100)), the candidate visual acuity parameters (in operation (300)), and the collected visual acuity chart data (in operation (400)). In some embodiments, operation (350) can use at least a portion of the multiple sets of candidate parameters and their corresponding chart-specific psychometric functions to generate a prediction of the visual acuity chart data or test data (e.g., response probability). In some embodiments, operation (350) can use all of the multiple sets of candidate parameters and their corresponding chart-specific psychometric functions to generate a prediction of the visual acuity chart data or test data (e.g., response probability).

[0096] The visual acuity model may include one or more visual acuity parameters. Non-limiting examples of visual acuity parameters include: visual acuity threshold, visual acuity slope, visual acuity range, sensitivity threshold, sensitivity range, change in visual acuity threshold between two test conditions, and change in visual acuity range between two test conditions. Given a visual acuity model, the visual acuity model parameters can be used to predict visual performance observed during testing with any visual acuity chart design. These model parameters can provide valuable visual acuity metrics to characterize a subject's vision.

[0097] In some embodiments, the visual acuity model includes a key component - a chart-specific psychometric function - that is configured to enable the expected number of correctly recognized optotypes in a row of an eye chart to be determined as a function of the optotype size in that row.

[0098] 7A-7B show an exemplary embodiment of a chart-specific psychometric function (706), which represents the expected number of correctly recognized symbols for each row of symbols as a function of the number of symbols presented in that row. Algorithm S can apply the visual acuity model to the eye chart data to evaluate the characteristics of the chart-specific psychometric function.

[0099] The systems and methods disclosed herein can estimate chart-specific psychometric functions (606), (706), chart-invariant single-target psychometric functions (404) or chart-specific single-target psychometric functions (604), (804), and / or multiple target psychometric functions (605), (805), and / or visual acuity parameters thereof, such as chart-specific visual acuity threshold (701), visual acuity range (703), visual acuity slope (702), or chart-invariant sensitivity threshold (401), sensitivity range (403), sensitivity slope (402), to assess visual acuity.

[0100] For example, the chart-specific psychometric functions (604) and (804) may be highly dependent on the particular design features of the eye chart used to collect the chart visual acuity data. The chart-specific psychometric functions may also predict eye chart data collected with different eye charts by incorporating features of the eye chart design. These features include, but are not limited to: The number of rows presented in a chart or subchart; number of presented targets per row; the number of potentially sampled targets per row; the number of potential sampled target sizes per row; Whether the pool of targets is sampled with or without replacement; or a combination thereof.

[0101] As a non-limiting example, for a standard logMAR eye chart presenting five targets on each row, a chart-specific psychometric function describes the number of targets expected to be correctly recognized on each row as a function of target size.

[0102] As another example, for a logMAR eye chart, two important visual acuity metrics that summarize the chart-specific psychometric function are: the chart-specific visual acuity threshold, i.e., the target size (in logMAR) at which the subject correctly recognizes the reference number of targets per row, and the chart-specific visual acuity range, i.e., a parameter that is correlated with the width of the dynamic range of the psychometric function and anti-correlated with the slope of the function. In some embodiments, the visual acuity range describes the difference in logMAR between two target sizes: one above the visual acuity threshold performance level and the other below. In some embodiments, the range parameter describes how rapidly the eye chart-specific psychometric function changes around the visual acuity threshold. A small range indicates that the observer's vision rapidly transitions from inaccurate to accurate visual markings as the visual acuity size increases from small to large. In some embodiments, the visual acuity range is a useful metric because one sign of eye disease is increased visual fluctuation, which corresponds to an increase or change in visual acuity range.

[0103] For example, for the logMAR eye chart, the visual acuity threshold is the target size that corresponds to correct identification of 3 out of 5 targets per row, and the visual acuity range corresponds to the distance between the rows of the chart with the expected number of 1.5 and 4.5 targets correctly identified per row.

[0104] In some embodiments, other characteristics of the psychometric function are determined by the design features of the eye chart. These characteristics may include a Lower Asymptote, which is determined by estimating a percentage over the total number of possible objects, and an Upper Asymptote, which is determined by the number of objects present in each row.

[0105] In some embodiments, a chart-specific single target psychometric function is generated by operation (220) of FIG. 3B and operation (230) of FIG. 3A to construct a chart-specific psychometric function (706) (606) that describes the expected number of correctly recognized targets per row as a function of target size. The chart-specific single target psychometric function can be generated based on the chart-invariant single target psychometric function (404) that is defined in terms of sensitivity parameters that are independent of the task-based characteristics (FIGS. 4A-4B) of the eye chart, e.g., eye chart design parameters or features. In some embodiments, the single target psychometric function indicates the probability of correctly recognizing a single target as a function of eye chart size, and is determined by two visual acuity model parameters, e.g., visual acuity threshold and visual acuity range, possibly using a cumulative distribution function, such as a Gaussian cumulative probability distribution function. The sensitivity-based chart-invariant single-target psychometric functions are then transformed into single-target chart-specific psychometric functions and target set characteristics based on the principles of signal detection theory (Green & Swets, 1961; McMillan & Creelman, 2004; Lesmes et al., 2015, which are incorporated herein by reference in their entirety) (Figure 5). This transformed single-target chart-specific psychometric function depends on the characteristics of the eye chart design, such as the number of potential targets. The same sensitivity threshold (target corresponding to d'=1) can represent different levels of accurate performance for signal targets based on the number of target options. As noted in signal detection theory, it is easier to guess accurately when there are few guess options. Therefore, for the same target size corresponding to a given sensitivity threshold (d' = 1), the number of guess choices in the target set can result in different levels of percent correct performance for discriminating a single target: for a single target, at a sensitivity threshold of d' = 1, it ranges from 70-80% for two choices and 30-40% for ten choices (Figure 5).

[0106] In some embodiments, a combination of operations, e.g., (300), (200), and (350), generates a visual acuity model applicable to score the eye chart data collected and summarized in operation (400). The visual acuity model may include a number of predicted probabilities, e.g., in a chart related to the observed eye chart data, where the predicted probabilities may be predicted probabilities of different responses recorded to optotypes presented to a patient for a given set of candidate visual acuity model parameters describing the patient's vision.

[0107] In one embodiment, a two-parameter description of visual acuity may use a sigmoid function to define visual performance as a function of target size (see, for example, FIGS. 4A and 6A). One parameter, visual acuity threshold, may describe the lateral location of the sigmoid function on the horizontal axis of target size. The other parameter, visual acuity range, may describe how rapidly the sigmoid function changes across the horizontal axis of target size near the threshold. A small range indicates that the visual acuity function changes rapidly from a lower visual performance indicated by the lower asymptotic line of the function to an upper visual performance indicated by the upper asymptotic line of the function. Both visual acuity threshold and visual acuity range parameters may be defined in units of logMAR target size.

[0108] With reference to FIG. 7A, the visual acuity threshold and visual acuity range are defined in some embodiments specific to the ETDRS visual acuity chart design. The ETDRS chart can present five letters per row for a range of visual acuity sizes. In FIGS. 7A and 7B, the visual acuity function thus demonstrates that the upper asymptotic row of five letters reported correctly out of five. In FIG. 7A, the visual acuity threshold (701) is defined as the visual acuity size at which the patient is predicted to report three of the five presented visual acuities correctly. In FIG. 7B, the visual acuity range (703) defines the horizontal range in which visual behavior on the ETDRS chart changes from a correct report of 1.5 out of five letters to a correct report of 4.5 out of five letters. In these embodiments, since the visual acuity threshold and visual acuity range are defined by three performance levels defined by the characteristics of the ETDRS chart, namely 1.5, 3.0, and 4.5 letters reported correctly out of five, the visual acuity threshold and range can be considered as visual acuity model parameters specific to the chart.

[0109] In some embodiments, as presented in Figures 4A and 4B, the visual acuity threshold (401) and visual acuity range (403) are defined in sensitivity units (d') that are independent of the visual acuity chart design. The visual acuity threshold can be defined as the target size (in logMAR) that corresponds to visual performance with d' = 1. The visual acuity range can be defined as the difference (in logMAR) between the target size that corresponds to the inferior level of visual performance (d' = 0.25) and the superior level of visual performance (d' = 4). With the definition of these visual acuity model parameters in units of sensitivity, the sensitivity threshold and range can be considered as chart-invariant visual acuity model parameters. To understand how operation (350) generates a predictive model for visual acuity chart data from chart-invariant parameters, it is useful to consider a common pair of visual acuity model parameters: one is the visual acuity threshold and the other is the visual acuity range (e.g., a pair of parameters including the first column of chart 2). From this selected pair of visual acuity model parameters, the probability of observing different responses during visual acuity chart testing can be obtained as a function of the target size. In addition to these visual acuity model parameters, consider one factor in eye chart design: the range of target sizes that will be tested on the eye chart. For example, the ETDRS chart uses target sizes ranging from -0.3 to 1.0 logMAR, with a logMAR resolution of 0.10.

[0110] FIG. 4A shows an exemplary embodiment of a chart-invariant psychometric function that may result from a continuous sampling of target sizes similar to the ETDRS test. Signal detection theory explains how task-independent sensitivity parameters can be transformed into task-based behavioral measurements (e.g., the probability of correctly identifying a target). Using signal detection theory (Green & Swets, 1961; McMillan & Creelman, 2004; Jacobs & Fine, 2002; Lesmes et al, 2015, all of which are incorporated herein by reference), the probability of correctly identifying a single target is derived from a d' value based on a Gaussian distribution, which is assumed for the internal representation of the signal and noise for discrimination tasks with different numbers of guessing options. For the equation that describes this transformation, see Equation 7 in Jacobs & Fine, 2002, or Equation 1 in Hacker & Ratcliff, 1979, as Appendix A. Alternatively, McMillan & Creelman (2004) present Table A5.7 as a look-up table that can convert between corresponding levels of Sensitivity (d') and Probability Correct for m-AFC alternative tasks ranging from m = 2 to m = 1000.

[0111] As an example, FIG. 5 demonstrates that for a typical stimulus corresponding to a sensitivity level of d'=1, when there are two guess options (2AFC) (one option is correct and the other is incorrect on any one trial, independent of any visual proficiency of the observer), the probability of correct identification is 70%-80%. As the number of guess options increases, the probability of an incorrect response may become more frequent. Hence, the probability of an accurate response decreases as the number of guess options increases (see the relative layering of functions (504) in FIG. 5). To compare with 2AFC, when 10 guess options are presented for a stimulus corresponding to a sensitivity level of d'=1, the probability of reporting the stimulus correctly is about 30%.

[0112] Based on such a transformation (330) from sensitivity (d') to probability correct (FIG. 5), a pair of visual acuity model parameters generating the sigmoid function of FIG. 4A can generate a monotonically increasing function that describes the probability of correctly reporting a target as a function of target size. The transformation to use from FIG. 5 may depend on the number of other targets (guessing options) in the sampling pool of targets. In the case of the ETDRS using the Sloan set of 10 targets, the 10AFC function may be used to generate a psychometric function (604) for a single target presentation (FIG. 6A).

[0113] Just as the number of optotypes in the sampling pool can be an optotype design factor used for the conversion from a chart-invariant psychometric function to a psychometric function for a single optotype (operation (330)), another optotype design factor, the number of optotypes presented on each row, can be used to generate a family of psychometric functions for multiple optotypes via serial multiplication. A weighted sum of this family of psychometric functions in turn provides a chart-specific psychometric function (FIG. 6C).

[0114] In some embodiments, given a single pair of visual acuity model parameters, e.g., visual acuity threshold and visual acuity range, six psychometric functions may be generated that describe the probability of correctly identifying 0, 1, 2, 3, 4, or 5 targets out of 5 as a function of target size. In operation (500), these probabilities may be used to evaluate which visual acuity model parameters best describe the test patient, given the eye chart data collected during testing.

[0115] In some embodiments, operation (300) may include multiple sets of candidate parameters that may be used for the visual acuity model. The purpose of the scoring algorithm S is to evaluate these sets of candidate visual acuity model parameters and evaluate which candidate parameters best explain the visual acuity chart data summarized in operation (400).

[0116] In some embodiments, in operation (300), the set of candidate parameters can be defined in one or more of the following ways: a one-dimensional array / list of vision model pairs, e.g., vision threshold and vision range; a one-dimensional array of vision model pairs including different values ​​of the vision threshold and a fixed value of the vision range; a two-dimensional space defined by an exhaustive combination of values ​​of two parameters, e.g., a vision threshold parameter and a vision range parameter; a high-dimensional space (more than two dimensions) of vision threshold and vision range parameters that account for the vision model parameters and their variations across different conditions.

[0117] For each of these sets of candidate visual acuity model parameters, generation of visual acuity model probabilities from visual acuity thresholds and ranges can be applied. In operation (500), these probabilities obtained with each set of these candidate visual acuity parameters can be used to find visual acuity model parameters that explain the visual acuity chart data.

[0118] For example, the number of possible targets can vary for different chart designs: Sloan, 10 Bailey-Lovie targets; Landolt C, Tumbling E, HOTV, Lea, Hands 4 optotypes; 9 optotypes for conventional Snellen; 12 targets to British Snellen standard BS 4274-1:2003; 26 symbols for the English alphabet

[0119] Later, in some embodiments, the psychometric function of a single target, P(s), is derived from a family of psychometric functions of multiple targets, P(s), specific to the design of the eye chart. m (s), are used to calculate the family of psychometric functions for such a multi-objective set, P m(s), m=1, 2, 3..., show a systematic pattern of performance for correctly recognizing multiple optotypes presented in multiple rows of an eye chart, thereby determining the combination of probabilities of correctly recognizing multiple optotypes presented in a row in an eye chart as a function of optotype size. In some embodiments, the family of n+1 psychometric functions includes one function defined for each of n+1 composite response results capable of correctly recognizing m out of n optotypes per row as a function of optotype size, where m ranges from 0 to n. In some embodiments, the number of psychometric functions in the family of psychometric functions of multiple optotypes depends on the number of optotypes presented in each row. For example, six multiple optotype psychometric functions for ETDRS chart designs with five optotypes per row. As another example, four multiple optotype psychometric functions include a family for chart designs with three optotypes presented in each row, and two multiple optotype psychometric functions include a family for e-ETDRS tests and other computerized chart designs with single optotype presentation.

[0120] In some embodiments, a family of psychometric functions for multiple targets determines the probability for a composite response to multiple targets as a function of target size. Such probabilities can be calculated for a complete sequence of targets by multiplying the probabilities for correctly recognizing a single target, P(s), and for incorrectly recognizing a single target, 1-P(s), as a function of target size, s. Specifically, in certain embodiments, for a row of n targets, the probability of correctly recognizing m of the n targets is calculated as a function of target size, s, according to the following formula: P m (s)=P(s) m ×(1-P(s)) n-m where m ranges from 0 to n and P(s) is a single target psychometric function that describes the probability of correctly recognizing a single target as a function of target size. In some embodiments, a family of psychometric functions for multiple targets is a set of n psychometric functions, i.e., P(s)=0, where m ranges from 0 to n, for n rows of targets. m Figure 6A shows a non-limiting example of a single-object psychometric function (604) determined by an object set with ten guessing options, and Figure 6B shows a non-limiting example of a family of six multi-object psychometric functions (605) determined by a chart design presenting five objects per row.

[0121] 7A-7B, in certain embodiments, scoring of eye chart data using algorithm S may require a visual acuity model predicting a composite response to multiple targets as a function of target size. The probability of different composite responses to multiple targets can be predicted by a pair of visual acuity model parameters (e.g., threshold and range / slope) that determine a single target psychometric function (704), and then a family of target psychometric functions (705) for multiple targets can be determined via serial multiplication of the single target psychometric function, P(s), and its complement, 1-P(s). For example, a multiple target psychometric function for five targets, Pr('m=5'), can be calculated as (Pr('m=1')^ 5m is calculated as (1-Pr('m=1')). In this case, for example, 'm=5' represents correctly identifying exactly five of the five targets. In this embodiment, the psychometric function of the single target is obtained by presenting a single target in each test performed on the subject (FIG. 7A), and the psychometric function of the multiple targets is generated by presenting multiple targets to the subject in each test performed. In this case, the eye chart is designed to present five targets per row, and the psychometric function of the multiple targets defines the probability of correctly recognizing a partial number of the multiple targets presented in a row of the eye chart as a function of the target size. In this particular case, only two of these psychometric functions of the multiple targets are monotonic. The probability of correctly recognizing a target as zero, P(m=0), decreases as the target size increases, and the probability of correctly recognizing five targets increases as the target size increases. The performance pattern for intermediate composite responses is marked by staggered peaks, indicating the various regimes of target sizes likely to demonstrate an increase in the number of correctly recognized targets. In some embodiments, a chart-specific psychometric function, describing the expected number of correctly recognized targets, is generated by a weighted sum of the psychometric functions of these multiple targets (FIGS. 6A-6B).

[0122] In some embodiments, a computer model is used to predict a subject's performance in a visual acuity test based on the shade sensitivity parameters. With reference to Figures 6A-6B, in this particular embodiment, to more clearly illustrate the concept of multiple target psychometric functions, the single target psychometric function (604) as shown in Figure 6A and the multiple target psychometric functions, e.g. (605), are presented as cumulative probability distribution functions (Figures 6A-6B). In this case, 'm=5' represents correctly identifying exactly 5 out of 5 targets. In this case, the cumulative multiple target psychometric function describes the probability of correctly recognizing at least a given number of targets as a function of target size. In some embodiments, the relative stratification of these functions suggests a systematic pattern of performance demonstrating an increase in the probability of correctly recognizing an increasing number of targets with increasing target size.

[0123] In certain embodiments, scoring of a subject's eye chart data using algorithm S requires a visual acuity model that predicts a composite response to multiple targets as a function of target size. The probability of different composite responses to targets can be predicted by a pair of visual acuity model parameters (e.g., threshold and range) that detail a single target psychometric function, and then a family of multiple target psychometric functions can be generated via serial multiplication of the single target psychometric function, P(s), and its complement, 1-P(s). The system and method advantageously use multiple target psychometric functions, and their application to visual acuity testing is novel.

[0124] In the example of an eye chart design with five targets per row, each of the multiple target psychometric functions may determine the probability of correctly recognizing as many of the multiple targets (five or less in number) presented on each row as a function of target size. As can be seen in FIG. 6B, among these psychometric functions for composite responses, only two multiple target psychometric functions are monotonic. The probability of correctly recognizing zero targets, P(m=0), may decrease monotonically as target size increases, and the probability of correctly recognizing all targets, P(m=5), may increase as target size increases. The performance patterns for other intermediate composite responses are marked by non-monotonicity, and the progression of alternating peaks may provide a quantitative approach to the intuition that increasing target size increases the probability of correctly recognizing an increasingly greater number of targets. The alternating peaks of the multiple target functions support the idea that, considering the increase in target size, different and successive regimes of target size represent an increase in the probability for correctly recognizing targets of increasing numbers. The model provided herein can be a rich and complex characterization of eye chart performance, involving psychometric functions of multiple targets. Existing methods do not allow this family of psychometric functions to be evaluated in a feasible process for eye chart testing. Chart-specific psychometric functions, which describe the expected number of correctly recognized targets, can be generated by a weighted sum of the psychometric functions of these multiple targets.

[0125] 8A-8B, exemplary embodiments of single target psychometric functions (e.g., (801) in FIG. 8A) and multiple target psychometric functions (e.g., (802) in FIG. 8B) are shown as cumulative probability functions. In this particular embodiment, the cumulative multiple target psychometric function describes the probability of correctly recognizing at least a given number of targets as a function of target size. The relative stratification of these functions suggests a systematic pattern of performance demonstrating an increase in the probability of correctly recognizing an increasing number of targets with increasing target size.

[0126] In some embodiments, the family of psychometric functions of multiple targets relies on the simplifying assumption of statistical independence of responses to multiple targets. This assumption relies heavily on sampling of targets with replacement and can be valid for a variety of eye charts. Non-limiting examples of these eye charts include those that use at least Landolt C, Tumbling E, HOTV, and Lea targets.

[0127] In some embodiments, for eye chart designs that sample the optotypes without replacement, such as ETDRS, an additional factor is required for the calculation of the family of psychometric functions for multiple optotypes. This additional factor can correct for the gradual decrease in available optotypes that occurs when the eye chart is sampled without replacement, specifically as the sequence of responses progresses to the end of the top row of the chart.

[0128] In some embodiments, a family of multiple target psychometric functions can be used to generate a chart-specific psychometric function that determines the expected number of correctly recognized targets per row on the eye chart as a function of target size. The chart-specific psychometric function optionally represents a weighted sum of the family of multiple target psychometric functions. As shown in FIG. 6C, the chart-specific psychometric function (606), E(s), can be specifically calculated as a weighted sum, weighting each of the multiple target psychometric functions (605) by the number of correctly recognized targets:

[0129]

number

[0130] Similar to the single target psychometric functions (604)(804) and / or multiple target psychometric functions (605)(805), such chart-specific psychometric functions (606)(706) may be determined by the following two chart-specific visual acuity model parameters (701)(703), which may provide important visual acuity metrics to characterize vision: a. a chart-specific threshold value that defines the expected number of correctly recognized targets at a target size that meets a predetermined criterion of visual performance; and b. Chart-specific ranges that define the difference (in logMAR) between target sizes that define two levels of visual acuity performance, one above and one below threshold performance

[0131] As a non-limiting example, for a standard ETDRS chart, the chart-specific psychometric functions (606)(706) are defined by the following two eye chart-specific parameters: a. a visual acuity threshold that defines the target size that provides an expectation of three correctly recognized targets out of five; and b. A visual acuity range that defines the number of rows of optotypes (logMAR of 0.10 per row) that separate sizes that provide an expectation of 1.5 and 4.5 correctly recognized optotypes out of 5.

[0132] In the case of the Snellen chart, chart-specific psychometric functions can be defined with the same parameters as the ETDRS chart, but two issues (visual acuity thresholds and ranges defined in sensitivity units) require consideration that the Snellen chart does not have the same number of symbols per row, and only one row on the Snellen chart has the same five symbols as the ETDRS chart. Therefore, conversions between chart-invariant sensitivity-based functions and chart-specific psychometric functions must be performed independently for each row of the Snellen chart that has more or fewer symbols than the five used by the ETDRS chart.

[0133] In the visual acuity model, visual acuity metrics, i.e. visual acuity thresholds and ranges, may be derived from the visual acuity model parameters at the level of the single-target chart-specific psychometric function as shown in Figures 6A-6B, at the level of the single-target chart-invariant psychometric function as shown in Figures 4A-4B and / or 5A, or at the level of the chart-specific psychometric function, E(s), as shown in Figures 6C and / or 7A-7B. These functions can provide visual acuity metrics as a function of target size, which are commonly associated with visual acuity testing. General metrics can also be defined at the level of multiple target functions, but are usually more difficult to resolve.

[0134] In some embodiments, an alternative critical formulation of the visual acuity model is defined by chart-invariant model parameters that define an alternative formulation of the psychometric function of a single target, and can: a. Complementing earlier formulations of visual acuity models that provide specific metrics for visual charts; b. Following the same calculation steps for chart-specific psychometric functions from single to multiple targets as for chart-specific metrics, but including an additional preparatory step preceding the single target psychometric functions.

[0135] In some embodiments, additional preparatory steps include application of signal detection theory to determine acuity threshold and range parameters on a sensitivity scale (d'), and corresponding transformation of the single-target sensitivity psychometric functions to single-target psychometric functions based on task-based chart features, including the number of guessing options.

[0136] Conversion of sensitivity (d') based psychometric functions to chart-specific single target psychometric functions may be based on signal detection theory incorporating design features of the eye chart design. Considering the same level of sensitivity threshold (target corresponding to d'=1), the exact recognition performance level for a single target may vary from 30-40% accuracy for a discrimination task with 10 choices to 70-80% accuracy for a discrimination task with two choices.

[0137] In some embodiments, the chart invariant model helps to propose and address that certain reductions in visual performance may be caused by design features of the vision test that are not strictly related to visual factors.

[0138] In some embodiments, the chart-invariant model provides a probabilistic behavioral explanation in which the visual acuity performance observed in a visual acuity test reflects the dual and independent contributions of sensory and decision-making processes. In some embodiments, the chart-invariant model proposes that a chart-specific or empirical threshold on the psychometric function depends on the number of guess options. For example, observing a performance of 60% correct recognition when a subject is presented with a visual acuity chart that samples a visual acuity chart set with 10 options is not equivalent to 60% correct recognition when sampling a set of 4 options.

[0139] In a particular embodiment, the chart invariant model proposes that the definition of the threshold for correctly recognizing 3 out of 5 optotypes depends on the total number of optotypes sampled. For statistical reasons, apart from the sensitivity capabilities of the detection system, it may be easier to correctly recognize 3 out of 5 letters when there are 4 options for each optotype instead of 10.

[0140] In certain embodiments, the chart invariant model makes an important distinction between factors that depend on the visual acuity test and factors that depend on the observer / subject and the observer / subject's performance in the visual acuity test: sensitivity is related to visual performance, other detection level factors are related to visual acuity test factors and inferential behavior.

[0141] In some embodiments, more options for target recognition provide more opportunities for inaccurate estimation of the target and degrade recognition performance.

[0142] In some embodiments, the chart-specific or conventional visual acuity thresholds are contaminated by the guessing behavior. In some embodiments, the change in recognition performance reflects factors of the visual acuity chart design. For example, in visual acuity chart designs with fewer options, correct guessing is more likely. Increasing the number of options may decrease recognition performance based on factors not related to vision. Considering the guessing behavior on recognition performance, signal detection theory proposes an equation to convert between chart-specific behavior (defined by % accuracy) and chart-invariant behavior (defined in units of sensitivity: d').

[0143] In some embodiments, the chart-invariant model defines the visual acuity metric using a chart-invariant psychometric function that describes the sensitivity (d') as a function of the target size. This function is not dependent on the particular eye chart design used to collect data from the test subjects, and such a model can be defined by the difference between the chart-invariant sensitivity threshold, i.e., the target size (in logMAR) corresponding to d'=2.0; and the chart-invariant sensitivity range, i.e., the target size (in logMAR) corresponding to d' values ​​of 1.0 and 3.0.

[0144] The usefulness of a sensitivity-based approach can be supported by the observation that visual acuity measured using the HOTV (with only four options) is superior to visual acuity measured using the Sloan Set.

[0145] Algorithm S can simultaneously evaluate chart-specific and chart-invariant visual acuity metrics. The advantage of visual acuity metrics, whether chart-specific or chart-invariant, is that both sets of visual acuity parameters are compatible and interconvertible.

[0146] The visual acuity model herein can provide novel predictions of eye chart performance and represents a data generation mechanism for model inference of visual acuity metrics. The model can use only the parameters of visual acuity threshold and visual acuity range to predict a rich set of relationships between eye chart size and performance observed during testing with an eye chart.

[0147] The visual acuity model can analyze eye chart performance by evaluating how the probability for correctly recognizing multiple targets varies as a function of target size.

[0148] In some embodiments, the visual acuity model provides a fine-scale analysis of the performance of the visual acuity chart, which provides a better explanation for the stochastic behavior observed around the visual acuity threshold. The visual acuity model may be more robust to random factors that may affect the end criteria that are actually currently being tested for visual acuity. The visual acuity model may be more robust to the stochastic behavior that occurs around the threshold. For example, due to random effects, when a subject is presented with a visual acuity threshold (501), e.g., a visual acuity threshold specific to the chart, the subject is approximately equally likely to observe two, three, or four correctly recognized visual acuities (e.g., FIG. 5). The visual acuity model may be versatile and flexible. For visual acuity chart designs with varying numbers of visual acuities per row, as the Snellen chart changes, the predictive model generates psychometric functions of different visual acuities for different rows. The visual acuity model may finally be applied to the visual acuity chart data to infer visual acuity model parameters that describe the subject's vision. The visual acuity model may be defined to remove bias from the Snellen visual acuity chart. Despite its use in many clinical settings, Snellen has been replaced in ophthalmic testing because it cannot be scored in the same way as the gold standard. The visual acuity model can allow for confusion between Snellen, ETDRS, HOTV, and any visual acuity chart design.

[0149] The visual acuity model can be succinctly defined with only two visual acuity parameters, but is flexible enough to generate chart-specific visual acuity metrics that describe performance based on a particular visual acuity chart, and chart-invariant visual acuity metrics that describe performance independently of the visual acuity chart used to collect the data.

[0150] In some embodiments, for the algorithm S to score the eye chart data, the entry of each row of the eye chart data table indicates the reading of a single optotype or a single row of optotypes from the eye chart or sub-chart. Each row of optotypes may be considered as a complex stimulus, determined by at least one of the following variables: the number of optotypes sampled per row, the number of potential optotypes sampled from the optotype set (i.e., the number of optotypes in the library), and whether the sampling from the optotype set is with replacement or not. The response entry in the summary table is the number of optotypes correctly recognized, regardless of a single optotype or a single row of optotypes. In this case, the response is a ratio calculated by the number of optotypes correctly recognized divided by the total number of optotypes in the same row / column.

[0151] In some embodiments, at least one visual acuity model is used to calculate the visual acuity parameter. In some embodiments, Bayesian adaptive estimation is used to calculate the visual acuity parameter. In certain cases, a two-dimensional Bayesian probability density function over two sensitivity parameters is used to calculate the visual acuity parameter. In some embodiments, Bayes' rule is used to calculate the visual acuity parameter. In some embodiments, at least one probability function is used to calculate the visual acuity parameter. In further embodiments, the prior probability of obtaining a response r is determined based on the visual acuity parameter, d. 1,x and d 2,y , p(r z |d 1,x ,d 2,y), is used to calculate the visual acuity parameter, where x and y are integers from 1 to the maximum number of parameters d1, m and parameters d2, n, respectively, and z is an integer from 1 to the maximum number of responses i. In some embodiments, the different visual acuity parameters p(d 1,x ,d 2,y ) and / or different responses p(r z ) in some embodiments. 1,x ,d 2,y |r z ), charts visual acuity, where p(d 1,x ,d 2,y |r z )=[p(d 1,x ,d 2,y )p(rz|d 1,x ,d 2,y ) / Σp(d 1,x ,d 2,y )p(r z |d 1,x ,d 2,y ), where Σp(d 1,x ,d 2,y )p(r z |d 1,x ,d 2,y ) charts the sum, where x runs on integers from 1 to m, y runs on integers from 1 to n, and z runs on integers from 1 to i.

[0152] To refresh or generate test pieces for the next test run, in some embodiments, the pdf, p(d 1,x ,d 2,y |r z ) is calculated over the space of sensitivity parameters. In some embodiments, the entropy H(rz)=Σp(d 1,x ,d 2,y |r z )log(p(d 1,x ,d 2,y |r z )), where Σ charts the sum, where x runs on integers from 1 to m, y runs on integers from 1 to n, and z runs on integers from 1 to i.

[0153] In some embodiments, at least one principle related to the vision chart design is used as an input to the statistical model. In some embodiments, the at least one principle is selected from the following: the number of rows presented on each test chart or sub-chart, the number of optotypes presented on each row of the test chart or sub-chart, the total number of optotypes in the library of optotypes, and whether sampling from the optotype library is done with or without replacement.

[0154] The visual acuity models herein may incorporate at least a portion of the complete definition details of the visual acuity chart design, i.e., five symbols presented per row. In some embodiments, the primary visual acuity metric provided by the visual acuity models herein is a chart-specific visual acuity threshold, e.g., (701), at which a subject at that symbol size (or equivalently, row on the chart) correctly recognizes three of the five symbols per row (or some other specific performance criterion that defines the threshold).

[0155] Predicting eye chart data considering candidate parameters In some embodiments, the visual acuity model is applied to generate a series of tables containing predictions of the probability of observing the collected visual acuity chart data given the candidate visual acuity model parameters, as shown in operation (500) of FIG.

[0156] For each eye chart condition or column in Table 1, a table of candidate parameter combinations is shown in Table 4, where the first entry is the candidate parameter and subsequent entries are the predicted probability of the observed composite response. For each eye chart condition depicted in the table shown in Table 1, the probability of observing that response is calculated for the complete candidate parameter set. In some embodiments, for each column in Table 2, a table for the complete candidate parameter set such as Table 4 can be generated.

[0157] Table 3 presents a set of candidate visual acuity parameters collected from test subjects and their predictions of eye chart data for a single test condition (i.e., 1 "data entry" of response r1 in Table 2).

[0158] [Table 3]

[0159] Each chart represents a set of candidate visual acuity parameters and a probability estimate of the observed response for the condition of that chart or sub-chart. In this case, q0 is equal to q1×q2. Optionally, there are multiple candidate visual acuity model entries for a single data entry. In some embodiments, each complete visual acuity chart data set includes multiple data tables. In some embodiments, the analysis of the complete visual acuity chart data set includes generating multiple data tables, one for each chart condition, each table described by a row of the data table corresponding to the operations in the flowchart in FIG. 17. In this case, the probability of the observed response is calculated. In other words, the probability of generating a single response, r1, is detailed in S1, taking into account the visual acuity parameters, i.e., α1 and β1, and the stimulus parameters.

[0160] Evaluating and Ranking Candidate Parameters for a Visual Acuity Model In some embodiments, after creating a complete set of tables, such as Table 3, a new schematic table is shown in operation (600) of FIG.

[0161] In the series of tables of charts of candidate parameters and predicted probabilities of observed responses, each column can summarize the presentation and response of a single target or a single row of targets. Each table entry position across the series of tables can correspond to a pair of candidate visual acuity model parameters. The same entry position in each table can correspond to the same combination of candidate parameters across the series of tables. Each table in the series of tables corresponds to the predicted probability of observing a single or combined response to the target size condition described on the table column for the complete set of candidate parameters.

[0162] In some embodiments, the eye chart data represented in the complete set of tables is summarized by combining the set of tables into a schematic table chart. In some embodiments, serial multiplication of the predicted probability values ​​at the corresponding entry positions over the complete set of tables results in a schematic probability of observing the complete set of eye chart data for such a particular candidate parameter. As shown in Table 4, a new schematic table is calculated by multiplying everything corresponding to all columns summarized in the schematic table by serial multiplication, the corresponding entries are multiplied, and a new table is created. The set of tables can be combined into a schematic metric. In some embodiments, serial multiplication provides a maximum likelihood estimate for the candidate visual acuity model parameters, given the observation of the complete set of eye chart data. In some embodiments, the resulting schematic table represents an evaluation of the candidate parameter estimates based on the observed chart data. In some cases, the resulting schematic table represents a likelihood calculation of the candidate parameters as a function of the data. In some cases, the resulting schematic table represents a likelihood calculation over the candidate visual acuity parameters as a function of the visual acuity data, which can be combined with a calculation of a prior probability function over the candidate visual acuity parameters, as in Bayes' rule. In some embodiments, the prior probability function can be determined by information about the subject, such as the subject's age, the status of an ocular disease, the status of a neuropathology affecting vision, or a previous vision assessment.

[0163] Table 4 shows the aggregate probability of multiple responses from test subjects based on different candidate visual acuity model parameters and visual acuity models. This table can be used to first predict visual acuity chart data and then retrospectively record data by ranking visual acuity model parameters, taking into account the composite response to prediction quality and / or goodness of fit. This table can also be applied prospectively to dynamically improve the collection of visual acuity chart data.

[0164] A set of candidate visual acuity parameters and their predictions for the eye chart data collected from the test subjects are presented for a single test condition (data entry 1), as shown in Table 4. Analysis of the complete eye chart data set involves generating a series of these tables, one table for each chart condition described by the columns of data summarized in Table 1. Each table can represent a set of candidate visual acuity parameters and a probability estimate of the observed response for that chart or sub-chart condition.

[0165] Optionally, the probability of responding to the test with a response of r1 to rk is calculated and presented in a ranking table, taking into account the visual acuity threshold and visual acuity range / gradient. In this case, the ranking table optionally ranks the aggregate probability of response, e.g., p((r1, r2, ..., rk);α1, β1, taking into account the visual acuity parameters. In this embodiment, the visual acuity parameter with the highest ranking is the visual acuity estimated based on a given response in the test.

[0166] [Table 4]

[0167] Evaluation of visual acuity model parameters describing test subjects Table 4 shows all the candidate sets of visual acuity parameters in Table 3 and the probability of obtaining a composite response for the visual acuity chart data collected in Table 2 calculated across those candidate visual acuity parameters.

[0168] Considering a table of likelihood values, i.e., probabilities of response aggregation, a successive ranking of the likelihood values ​​(e.g., from highest to lowest) may provide the equivalent of a maximum likelihood estimation analysis, as shown in Table 4. As shown in operation (700) in FIG. 1, the highest ranked value and the corresponding candidate visual acuity model parameters may be selected in the sorting table to obtain the highest likelihood estimate of the visual acuity model parameters. Alternatively, if prior information, e.g., demographic information, or results of previous visual tests, is used to weight the serial multiplications, the sorted ranking may correspond to a weighted likelihood or Bayesian parameter estimate.

[0169] In some embodiments, the prior information may include, but is not limited to, the subject's genetic risk, the results of previous vision tests, the results of past medical diagnoses, or any other physiological or medical information. When the table is standardized to sum to one, the table of prior information reflects Bayesian inference. With standardization, the table represents a Bayesian probability distribution that can be defined by its mean, median, or mode. In addition to such statistics and metrics of central tendency (e.g., mean, median, mode), there is a statistical metric of the spread or dispersion or variability of the Bayesian posterior probability (is this correct?) function that is defined across the candidate parameters of the visual acuity chart model. In some embodiments, the evaluation and ranking of the visual acuity model parameters includes the standard deviation or percentile of the visual acuity model parameters. In some embodiments, the evaluation and ranking of the visual acuity model parameters includes statistical inference (maximum likelihood estimation, Bayesian, or any other applicable method), which takes into account the observed visual acuity data and model-generated predicted probabilities for visual acuity performance, resulting in visual acuity metrics via visual acuity parameter estimation. In some embodiments, the evaluation and ranking of visual acuity model parameters involves statistical inference of the visual acuity threshold and range parameters that best explain the eye chart data collected from the test subjects.

[0170] To yield visual acuity parameters that are either chart-specific or chart-invariant, the systems and methods herein may include one or more sub-steps for analysis and inference: 1. Create a table, e.g., Table 2, describing candidates for visual acuity threshold and / or range parameters. 2. For combinations of these candidate parameters, and given prescribed features of the eye chart design, a predictive model can be constructed for scoring eye chart behavior. 3. Given the response of a single target or multiple targets determined by an eye chart design (e.g., eye test data or eye chart data), statistical inference can be achieved by maximum likelihood estimation or Bayesian analysis, providing estimates of visual acuity parameters. 4. Consider the number of rows of targets presented during the eye chart test and the target sizes and responses observed on that test. 5. Start by considering the parameter space, including the eye chart design parameters for a range of single target sizes. 6. Based on signal detection theory, determine the parameters in the sensitivity (d')-based function 7. Use the scoring model from step 2 to generate predicted probabilities for parameter estimates made by either maximum likelihood estimation or Bayesian estimation. 8. Give a model that predicts the probability of observing a pattern in the eye chart data 9. The general model for analytics involves three sets of values ​​that can be organized in tables: 10. Three tables are important for building the analytical framework: a) Data collected from subjects, e.g., Table 1 b) Candidate visual acuity model parameters, e.g., Table 2 c) A corresponding table of probabilities determined by the candidate visual acuity parameters: for the table of probabilities, the stimulus-response combinations collected during the visual acuity test will correctly recognize the size of the targets presented in the rows of the chart, and the number of targets, e.g., Table 4 11. Generate candidate psychometric functions using predicted probabilities of response outcomes as a function of target size 12. For each target size presented during the test, a predicted probability for observing that response is calculated for at least a portion of the candidate visual acuity parameters. 13. For each target size presented on each row of the chart, a different chart presents the predicted response probability. 14. The method and system for scoring may yield two visual acuity metrics: threshold and range of chart-invariant psychometric functions. 15. Bayesian inference can provide estimates of visual acuity metrics along with confidence in those estimates, or maximum likelihood estimation can provide estimates of visual acuity metrics.

[0171] Determining a set of candidate charts for potential collection of eye chart data from test subjects In some embodiments, a set of candidate eye charts may be evaluated based on their ability to provide potential conditions for collecting eye chart data, as shown in FIG. 10. In some embodiments, this analysis evaluates the probability of observing expected eye chart data and how the resulting analysis determines the test conditions for collecting eye chart data. In some embodiments, due to the generality of the visual acuity model, the set of candidate eye charts may include any eye chart design. In other cases, to adhere to the recommendations of the standards committee, the most common approach to define a set of candidate charts may be to define eye charts that follow the same design principles as existing charts, such as logMAR charts (ETDRS, Bailey-Lovie), Snellen charts, Sloan low contrast charts, and / or sub-charts of logMAR or Snellen (single row or single target).

[0172] In some embodiments, principles of eye chart design include characteristics of the objects presented on the eye chart: i. The size of the target set being sampled; ii. the number of targets presented per row; iii. the number of optotype rows presented per chart; iv. Optotype sampling methods; or a combination thereof.

[0173] The set of candidate charts follows the same general design principles as the prior art for standard eye charts, but the candidate charts can be constructed with finer resolution and wider ranges for sampling eye size (FIGS. 11A-11F, and / or FIG. 12). Each individual chart may demonstrate the same precision of sampling eye size as the prior / current art, but taken together, the set of candidate charts may demonstrate a precision of sampling eye size that is far greater than any existing technology in the art. Such an expanded library of eye charts can improve precision and expand flexibility for sampling eye size.

[0174] In order to improve the sampling resolution for optotype size in visual acuity testing, a method for generating an expanded library of optotype charts while closely following the standards and principles of visual acuity design is disclosed herein, as shown in Figures 11A-11F. Figures 11A-11C show a typical sampling scheme represented by the ETDRS standard chart, which presents 14 standard sizes of optotypes from -0.3 to 1.0 logMAR, with a sampling resolution of 10 logarithmic units between successive sizes. In these embodiments, the anchor point of 0.35 logMAR represents the average optotype size of the ETDRS standard chart. By changing the anchor point of the standard constrained chart to generate a larger complement of optotype charts, the resolution for sampling optotype sizes is greatly improved across a collection of charts, while maintaining the adherence of each individual chart to the optotype chart standard.

[0175] In some embodiments, a set of 10 vision charts is generated by modifying the anchor points of a standard chart in the range of 0.3-0.4 logMAR, with an anchor point resolution of 0.01 logMAR. Each chart uses the same sampling resolution as the standard. However, when the entire set of charts with design limitations is considered, these charts provide finer resolution of the target size. Considered relatively, this set of 10 charts shows a higher sampling resolution for target size (0.01 log units) and across a wide range of 141 different target sizes.

[0176] As an example, the ETDRS chart design includes 14 rows of symbols, with 5 symbols per row, with a logarithmic decrease in size as each row descends. A logarithmic progression of 0.10 decimal units results in approximately a 25% decrease in size with each successive row of symbols. These design elements can be maintained in a library of charts with different symbol sizes that conform to the constraints of the design elements (FIGS. 11A-11C). An example subset of a library of design-constrained eye charts was generated using the ETDRS design principles with three different anchor points, 0.30, 0.35, and 0.40 logMAR (FIGS. 11A-11C), and the Snellen design principles with three different anchor points (FIGS. 11D-11F).

[0177] Shortcomings of these chart designs may be related to the invariant range and resolution for sampling of target sizes. Due to the non-adaptive and deterministic test routine, all patients are presented with the same range of targets. Furthermore, the invariant sampling resolution of the test results in insensitivity in measuring visual acuity values ​​corresponding to target sizes that are intermediate compared to those presented for the 14 established target sizes.

[0178] As shown in Figures 12A-12C, the visual acuity test can be performed with optional sub-charts, which represent tests focused on specific regions of the chart. Sub-charts are sub-sampled from the full design-constrained visual acuity chart by presenting only a limited set of symbols from the full chart, which is then sub-sampled from a larger library of design-constrained charts. Rather than presenting the full 14 rows of the ETDRS chart, the presentation of a set of sub-charts can focus the visual acuity test on a spatial sub-sample of the full chart (e.g., 3 rows of symbols). The target letters of the focused visual acuity test can be distinguished by presenting only the sub-sampled regions of the full chart (Figure 12A), by presenting the sub-sampled regions with high contrast compared to the low contrast of the rest of the chart (Figure 12B), or by presenting the sub-sampled regions without blurring and without blurring the rest of the chart (Figure 12C).

[0179] Evaluating and ranking candidate visual acuity charts taking into account ranking of candidate parameters for visual acuity model In some embodiments, given such candidate parameters, a table of predicted probabilities for potential outcomes for each row of the vision chart can be constructed. The ranking of the vision model parameters is inherent to the probability distribution defined over the model parameters.

[0180] In order for the adaptive algorithm to personalize and focus testing on individual patients, it may be necessary to calculate a ranking for a set of candidate eye charts. It may also be necessary to derive and calculate a utility function used for ranking. Eye charts can be evaluated and ranked according to different criteria. In some embodiments, eye chart designs are evaluated for their utility in testing a given subject. The utility function may include one or more of the following: active learning, variance minimization, entropy minimization, maximum information gain, information gain of chart-specific parameters, information gain of chart-invariant parameters, or combinations thereof.

[0181] Existing techniques (including the QUEST method) only calculate design optimization for a single optotype. One advantage of the system and method disclosed herein is the ranking of optotype chart designs for a single optotype, a single row of optotypes, or a complete chart of optotypes. The goal of this ranking process is to personalize testing to a particular subject by selecting an optimal or near-optimal optotype chart or sub-chart (e.g., a highly ranked chart or sub-chart). In some embodiments, composite optimization refers to the optimization of more than one optotype, and a composite or aggregate set of optotypes is presented for a chart or sub-chart. A composite utility function may be calculated over the entire optotype chart design. Composite stimulus optimization allows for optimal design selection of an optotype, rather than a single character. Optimization is calculated for each component part of the composite stimulus. The optimal design may be selected based on a chart or sub-chart. In this case, the utility function is calculated for each component part of the composite and then summed.

[0182] Algorithm A may provide a prospective predictive analysis of the optimal eye chart to be used to collect eye chart data from a test subject. For comparison, algorithm S may provide a retrospective analysis of eye chart data collected from a subject. The principle underlying algorithm A is optimal design selection: analyze a set of candidate charts to find the best chart for collecting eye chart data for a particular subject.

[0183] To rank the eye charts, a utility function may be calculated for a set of candidate charts. Algorithm A may select an eye chart that focuses on a particular test subject. This is done by optionally evaluating, ranking, and selecting the best candidate eye charts for collecting eye chart data. The best design selection may be calculated on chart-specific or chart-invariant parameters. In some embodiments, the adaptive eye chart optionally includes a combination of algorithms S and A. The advantage of algorithm A over the post-testing of algorithm S is that it "closes the loop" between the scoring and testing phases. Algorithm A can improve accuracy by iterative testing, scoring, and retesting of the eye chart metric.

[0184] For visual acuity testing with an adaptive visual acuity chart, the optimal design selection achieved by algorithm A may be calculated by a computationally intensive process that foresees predicting the outcome of the next test in the test and evaluates the best visual acuity chart design for the next test in the test. The scoring algorithm may be applied after every test in the test. In addition, the visual acuity model and scoring may be applied to evaluate the likely outcome of the next test. In this case, the visual acuity test may include multiple tests, where one or more responses from the test subject are collected in each test.

[0185] In some embodiments, the ranking of the candidate charts can be used to select the individualized sub-eye chart (e.g., with the highest rank). In this embodiment, the candidate chart can include three rows with a standardized progression of 0.10 log units between the rows. For example, to mimic the detection of changes in visual acuity between two health states (0.10 log units / 5 letter difference), a relative pattern of optimization can be presented for the two health states. The relative shifts in utility functions reveal the accuracy of the combination algorithm for individualizing visual acuity testing and for detecting subtle changes in vision.

[0186] In some embodiments, using a selected utility function, e.g. (1407a), (1407b), that varies with the change in the target size, the candidate charts can be evaluated based on their corresponding utility values, and a ranking can be done for each of the candidate charts. Highly ranked charts can provide high utility values ​​(e.g., arbitrary units) and may be used for testing, while low ranked charts may be less useful and not used for testing. The advantage of using a composite utility function (1407a) vs. a single utility function (1407b) to select highly ranked sub-charts is demonstrated in FIG. 14A. As shown in FIG. 14A-14B, the composite utility function (1407a) may share a similar curve shape as the single utility function (1407b). However, for the same target size, e.g., in the range of 0.2-0.4, the utility value is higher. In addition, a peak can be obtained with a small target size compared to the single utility function. Current technology, including the QUEST method, bases the utility function and its optimization on a single target size. When the optimization of a three-row subchart is considered, due to the standard constrained progression in target size between rows (represented by the space between diagonals), the utility function peaks (1408a) (1408b) for single vs. composite optimization may be evident at different target sizes. As shown in FIG. 14B, in some embodiments, with a selected utility function that varies with the change in target size (selecting the utility function to determine whether the two conditions indicate a change in visual acuity threshold of 0.15 units vs. 0.25 units), the utility function shows different peaks (1408a) (1408b) when the stimulus is a subchart containing a row of a single target or a composite stimulus of multiple target sizes.

[0187] With reference to FIG. 15, in a particular embodiment, an example of a two-slide sequence from an adaptive standard constrained visual acuity test is presented. A typical slide n and a longer sequence n+1 are presented, with branches representing different potential outcomes for reading slide n. The branches are not exhaustive and serve to show a small subset of potential outcomes for the presentation of slide n (1501). On slide n (1501), the test focuses on the most informative middle area of ​​the ETDRS chart. Following the presentation of slide n (1501), which includes three rows of ETDRS sub-charts, the subject responds and responses r (1503) are collected. In this case, the composite three-column response describes how many symbols (out of a maximum of five) are correctly signaled for each of the three columns. If a maximum of five characters per column are correctly signaled for each column (lowest branch), then the test on the next slide (1502) can be adapted to present a sub-chart with smaller symbols. In this embodiment, when the subject's response falls short of reporting all three columns correctly at will, the next presentation of the slide tests a different area of ​​the complete eye chart. In this embodiment, the presented sub-chart of eye charts conforms to the design principles of eye charts: multiple rows of letters with a size progression of 0.10 logarithmic units. On each subsequent slide, the adaptive testing algorithm focuses the test on the eye chart size that is most useful for testing each individual's visual acuity. Accurate and fine-grained sampling of eye chart sizes is achieved while maintaining a 0.10 logarithmic unit progression between each row. On each slide, the sub-chart of eye charts conforms in some embodiments to the design principles of eye charts: multiple rows of letters with a size progression of 0.10 logarithmic units. On each subsequent slide, the adaptive testing algorithm focuses the test on the eye chart size that is most useful for testing each individual's visual acuity. Accurate and fine-grained sampling of eye chart sizes is achieved while maintaining a 0.10 logarithmic unit progression between each row.

[0188] Referring to FIG. 18, in a particular embodiment, the convergence of the adaptive visual acuity test is reflected in the Bayesian posterior probability estimates obtained from the execution of multiple mimic tests using three slide tests. In this embodiment, the widest range of posterior probabilities reflects the initial visual acuity threshold estimate obtained from the patient reading the first column of the full 14-row ETDRS chart shown on slide 0. On the next chart presented on slide 1, the patient is presented with an even smaller 5-row chart that adheres to the ETDRS principles and is patient-specifically optimized. The Bayesian posterior function resulting after presentation of slide 1 reflects the Bayesian update of the parameters of the visual acuity model. The reduction in variance of the Bayesian posterior function reflects the convergence of the visual acuity sensitivity assessment. The inset presents the rapid reduction in the test variance as the number of test slides increases.

[0189] With reference to Figure 19, in a particular embodiment, experimental results are shown for an adaptive chart-based visual acuity test. In this embodiment, subjects with normal vision are tested using a standard constrained adaptive test that includes five test slides in a sequence, each test slide includes five rows, and each row includes five characters sampled without replacement from the Sloan set. In this embodiment, four test runs are completed, and the variability of the test results is calculated for multiple runs. Convergence is reflected in a decrease in the variability of visual acuity threshold estimates as the number of test slides increases.

[0190] Referring to FIG. 20, in a particular embodiment, subjects with normal vision are tested using a standard constrained adaptive test with five test slides in a sequence, each test slide containing five rows, each row containing five characters sampled without replacement from the Sloan set. To validate the functions evaluated by the adaptive test, a predetermined stimulus sampling (constant excitation method) of the target size was used, and the probability of correctly reporting various numbers of targets (0 to 5) was calculated and plotted. Only two psychometric functions show monotonicity. The probability of correctly reporting 0 (0 / 5) of 5 targets P decreases monotonically with increasing target size. As the target increases in size, the probability of correctly reporting none of the 5 converges to 0%. As a supplement, the probability of correctly reporting 5 of 5 targets starts at 0% when the target size is small, and increases to 100% as the target size increases in larger and larger amounts. These are the intermediate multi-object psychometric functions P(1 / 5)(), P(2 / 5), P(3 / 5), and P(4 / 5), which show the non-monotonic behavior of interest. As predicted by the visual acuity model, the functions may show ordered and staggered peaks with increasing number of correctly recognized objects. Such a probability function for correctly reporting 0-5 out of 5 objects is the multi-object psychometric function (2005). The chart-specific function, E(s) in this embodiment, is denoted as (2006).

[0191] To complement FIG. 20 and to demonstrate the validity of the visual acuity model, FIG. 21 presents cumulative probability functions that can be obtained from the psychometric functions (2005) of multiple targets in FIG. 20. The validity of the visual acuity model as a proper data generating model can be supported by the relative location and laminarity of these cumulative functions (2109). These data provide evidence that the psychometric functions (605)(805)(2005) of multiple targets can provide richer information about performance on the visual acuity chart than the psychometric functions and / or visual acuity parameters of a single target.

[0192] In some embodiments, prior information about the patient's vision is used to calculate a personalized eye chart design that ranks high according to the utility function. In some embodiments, prior information about the test subject may be collected, such as results of previous vision tests, results of other eye tests, diagnoses of coexisting diseases, demographic information such as age or race, or similar information. Based on the prior information, adaptive algorithm A may select the optimal chart design for that individual for the next test or for the entire test.

[0193] The advantage of such adaptive vision testing is that it is a comprehensive framework for evaluating and optimizing the information transmitted by the vision test: such methods may compute optimization (via the expectation of information gain) for m-ary responses rather than simply binary responses.

[0194] In some embodiments, an initial ranking of candidate vision model parameters may be provided based on a diagnosis of ocular disease.

[0195] In some embodiments, the visual acuity test begins with a broader, simpler visual acuity task. For example, the patient may be instructed to read only the first column of a chart of complete indicators, as shown in Figures 12A-12C. The last row, where the first indicator can be identified, provides a reference point to begin the complete testing procedure. Despite its usefulness (see below), this method is conceptually different from taking an initial sample from a library of visual acuity charts.

[0196] For visual acuity testing, the display of a chart or subchart of indicators is followed by a response collected by the technician or a response or self-registered by the patient. In different response modes, patients can (1) read the first column, (2) read across the row until instructed to stop, or (3) read a single indicator that is presented to them. (During the actual test, the gray indicators look the same as the other indicators. In this schematic, the different shades of gray indicate that they are not as prioritized for patient report as the black-only indicators.)

[0197] The visual acuity models herein can be adapted to analyze different visual acuity testing workflows, such as, for example, reading the first row, reading the entire chart, or reading only a single index.

[0198] As disclosed herein, the initial sampling from the chart library can take several forms selected from, but not limited to, the following: As in the prior art, a non-random, complete test, single standard ETDRS (or Snellen) chart is presented and the patient attempts to read all 14 lines. In this case, the initial sample is simply the same chart that is normally presented to all patients. b. A single standard ETDRS chart is displayed as above, non-random, abbreviated, but the patient reads only the first column. c. Any chart from the library of random, complete exams will be sampled and displayed. There is no need to use the fixed target size of the standard ETDRS or Snellen charts currently presented to all patients.

[0199] In some embodiments, the systems and methods disclosed herein can be used to analyze visual acuity metrics across these three different forms. More specifically, such analysis can be applied to visual acuity data collected on current ETDRS and Snellen charts, generating visual acuity metrics independent of the chart design principles. Alternatively, the patient reads the first column of the initial chart, and then testing continues with another chart or sub-chart based on the adaptive visual acuity chart disclosed herein. Analysis can be applied to both the initial chart and the subsequent test charts.

[0200] (Response Scoring) In visual acuity testing, response scoring is a key component that connects many other component processes, especially the process of estimating visual acuity metrics that decide whether to continue or terminate the test. Response scoring may be explicitly connected to the design features of the test. The decision to terminate the test may be related to the inability to see the indicator. In the prior art, the primary difference between chart-based methods was line assignment versus letter-by-letter scoring. For single-indicator computerized testing, the scoring is the accuracy of the identification: correct or incorrect. In some embodiments, a more fine-grained scoring is used for the number of letters correctly reported in each line. This fine-grained line-based approach, which is the basis of the present invention, differs from the current techniques of line assignment and letter counting.

[0201] Typical scoring of a visual acuity test may follow one or more heuristics. Non-limiting examples of such heuristics include using letter-by-letter scoring to assign each line a pass / fail grade based on correctly reporting any letter in that line, assigning the last line as read correctly on three out of five measures of accuracy, or counting all letters that are correctly reported.

[0202] In some embodiments, the algorithm disclosed herein focuses on the visual acuity metrics collected in a single test condition.It is often more important to measure and compare the visual acuity metrics in two test conditions (left eye vs. right eye, low luminance vs. high luminance).Therefore, it is also possible to consider an extended algorithm that provides scoring and adaptive testing for the change in visual acuity between multiple conditions.

[0203] In some embodiments, the scoring algorithm may be applied in a very simple "adaptive" manner. For predictive analysis during testing with established charts, the scoring algorithm may make a decision to continue testing until the visual acuity estimate reaches a certain level of accuracy. Specifically, the same scoring algorithm may analyze data collected when subjects are tested using ETDRS with static, deterministic paper charts. These charts may be used for repeated testing until the scoring algorithm S returns parameter estimates of a given accuracy and / or reliability. Rather than considering this an "adaptive" application, it may be more accurate to consider it an end-of-test application for quality control.

[0204] In some embodiments, the response format herein depends on the study design, e.g., whether the patient is presented with a full chart, a chart subsample of fewer rows, or a single indicator. Complete charts, column by column, until they can read no more than two letters. The first column, Single characters or combinations of characters can be read. These three types of responses may give different results. Conventional techniques give equal scores to all letters. The technician may register the patient's response through the identity of the indicators presented individually or consecutively, through the correct or incorrect judgment of the presented indicators, or through a combination thereof.

[0205] In existing techniques, these different responses are scored independently. Another method actually considers the full optotype confusion matrix to evaluate the probability of reporting different optotypes based on their identity (e.g., "O" is easier to read than "K"). The systems and methods disclosed herein can handle the three types of responses within the same theoretical framework. A more fine-grained row-by-row scoring can be generated using the relative response distribution. In some embodiments, stochastic behavior in the visual acuity task is considered. In some embodiments, the scoring may be based on the relative distribution of response outcomes as a function of the optotype size. In some embodiments, the visual acuity model considers composite responses, combinations of responses, probabilistic descriptions, and considers and predicts the probability of composite responses, or combinations thereof.

[0206] In some embodiments, scoring may be done via a data table of candidate acuity functions, a look-up table of predicted probabilities, an empirical table of confusion matrices, or a combination thereof.

[0207] In the prior art, many explicit testing strategies are design dependent, since the design principles that define visual acuity characteristics similarly constrain the visual acuity metrics. This explains why these scoring methods yield different visual acuity estimates for the ETDRS optotype set with 10 options and the HOTV / Lea optotype set with 4 options. As a result, visual acuity estimates obtained from different visual acuity tests are not directly comparable. Recent standards have evaluated alternative approaches to assess performance that attempt to correct for different optotype guess levels. These standards recognize the utility of using psychometric functions to describe optotype recognition performance. For example, acuity thresholds are defined as 50% of guess corrected psychometric function: Pc = guess rate + (1-guess rate) * (sigmoid function) The use of psychometric functions recognizes that visual acuity performance is probabilistic and defines the visual acuity threshold to be the 50% threshold of the guess-corrected psychometric function. Due to different guess rates for optotype sets of different sizes, the 50% threshold of the guess-corrected psychometric function corresponds to different levels of objective performance. In summary, the main methods in the existing technology for scoring eye chart data to derive visual acuity metrics can be listed as follows: a. How to lower the limit - row allocation b. Character Count - Each character is counted as 0.02 logMAR units c. An empirical or chart-specific threshold of 50% estimated from least-squares or maximum likelihood fitting of the estimated corrected psychometric function The empirical or equivalent chart-specific thresholds in the specification for the chart-specific or empirical functions may depend heavily on design criteria, such as the number of guess options for different visual acuity labels. Thus, the novel application of signal detection theory to characterize psychometric functions in chart-invariant sensitivity units may provide a potential solution for visual acuity metrics that do not rely on explicit features of the visual acuity chart. The conversion between the sensitivity psychometric functions prescribed by signal detection theory and the empirical chart-specific psychometric functions may depend on the visual acuity chart functions proposed in the visual acuity chart standard.

[0208] Scoring Algorithm (Algorithm S) In some embodiments, the systems and methods disclosed herein include a scoring algorithm or algorithm S, which is a key component of the framework for visual acuity testing. During visual acuity testing, subjects attempt to recognize different sized optotypes displayed on an optotype chart or sub-chart. When applied to retrospective analysis, algorithm S can generate accurate visual acuity metrics from previously collected visual acuity chart data. Although the scoring is primarily based on visual acuity data collected on a full optotype chart, with multiple optotypes displayed on multiple rows, the analysis can be simplified to score visual acuity tests using sub-charts: a series of rows of optotypes, a single row of optotypes, or a single presentation of an optotype.

[0209] In some embodiments, algorithm S exhibits flexibility for analyzing visual acuity data as a function of the particular chart design used for testing, and more generally for analyzing visual acuity data and generating visual acuity metrics independent of chart design. This analysis allows for universal, chart-invariant visual acuity metrics or parameters to which visual acuity results can be compared and adjusted across the many chart designs used in clinical practice.

[0210] FIG. 1 shows a flow chart of an exemplary embodiment of the scoring algorithm 40 disclosed herein. In this particular embodiment, the algorithm S (scoring algorithm) 40 starts and at least one vision chart design is determined (100). In this embodiment, the vision chart design determination (100) is based on information provided by a user and / or information automatically generated by analysis of the vision chart or sub-chart used to test the subject. In the same embodiment, the responses (i.e., vision chart data) collected from the subject are summarized (400), for example, in a vision chart data table. In parallel or in series with the summary of the vision chart responses (400), operation (300) can determine one or more sets of candidate vision parameters (e.g., chart-specific or chart-invariant parameters) that may explain the vision of the subject. Operation (300) can be performed before or after operation (200) in parallel. In operation (200), the visual acuity model may be defined based on chart-specific or chart-invariant parameters (e.g., chart-specific visual acuity threshold, chart-specific visual acuity range, chart-invariant visual acuity threshold, etc.). In some embodiments, a set of candidate visual acuity parameter values ​​is selected as values ​​for the visual acuity parameters in the visual acuity model determined in operation (200). Alternatively, or in combination, the visual acuity model shown in FIG. 6A-6C may be generated in operation (300) based on chart design parameters for predicting the subject's visual acuity chart data. In this case, the defined visual acuity model is applied to retrospectively predict the visual acuity chart data using the selected candidate parameters (500), and the probability of an aggregate response is calculated given the visual acuity model and its two visual acuity parameters, as shown in Table 2. Then, in the same embodiment, each visual acuity model with its candidate parameters is evaluated and ranked (600) based on a table summarizing the visual acuity chart data collected from the subject (400) as shown in Table 1. The estimation of the visual acuity parameters may be based on the ranking of the visual acuity models with the candidate parameters (700). In this case, the visual acuity model with the highest rank is selected as the estimate of the visual acuity parameters for the test subject.The results of those statistical inferences are presented to the user and / or subject 800. Optionally, algorithm S terminates after the inferences are completed.

[0211] Referring to Fig. 3B, in a particular embodiment, the determination of the visual acuity model for predicting the visual acuity chart data or visual acuity test data in Fig. 1 begins (200) with the determination of at least one single-target psychometric function (604), (804) using two chart-specific visual acuity parameters, i.e., a set of candidate visual acuity parameters such as visual acuity threshold and visual acuity range, in operation (230). In the same case, a family of multi-target psychometric functions is calculated by serial multiplication of the single-target psychometric functions in operation (320). In the same embodiment, if necessary, a chart-specific psychometric function is calculated using a weighted sum of the family of multi-target psychometric functions in operation (420).

[0212] Referring to FIG. 3A, in a particular embodiment, the determination of the visual acuity model for predicting the visual acuity chart data or visual acuity test data 200 in FIG. 1 begins with the determination (404) of at least one single-target psychometric function that is chart-invariant using two chart-invariant visual acuity parameters in operation (230), namely, the sensitivity threshold 401 and / or the sensitivity range (403) in FIG. 4A, and the cumulative Gaussian distribution function (cdf). In some embodiments, other chart-invariant visual acuity parameters such as sensitivity slope, sensitivity threshold change, etc. may be used. In some embodiments, there may be two, three, four, five, six, seven, or more single-target psychometric functions. In this embodiment, the single-target psychometric function is converted from the chart-invariant psychometric function (404) to a chart-specific psychometric function (604), (804) in operation (330). In this embodiment, the single-target psychometric functions are converted from the chart-invariant functions (404) to chart-specific psychometric functions in act (330) by generating intermediate chart-invariant functions (504) as shown in Figure 5. In the same embodiment, the family of multi-target psychometric functions is calculated by serial multiplication of the single-target psychometric functions (320) that are chart-specific. In the same embodiment, if desired, the chart-specific psychometric functions are calculated using a weighted sum of the family of multi-target psychometric functions (420).

[0213] Referring to FIG. 5, in certain embodiments, the chart-invariant single-target psychometric functions such as (404) may be presented using different coordinates than the chart-invariant single-target psychometric functions in the conversion to the chart-specific single-target psychometric functions (604). In this embodiment, FIG. 5 shows the probability of correctly reporting a single letter as a function of sensitivity, given the number of options. There may be different functions (504) depending on the number of options (e.g., 2AFC, 10AFC). For example, the function corresponds to the probability of correctly responding to "A" when the target set (the set of guess options) is "A" and "C" (2AFC). The probability of correctly reporting "A" from a target set with 10 options becomes lower.

[0214] Adaptive Algorithm (Algorithm A) In some examples, the systems and methods disclosed herein include adaptive algorithms. Adaptive algorithms can be used to improve accuracy by focusing testing on individuals. To complement the application of algorithm S to retrospective analysis, adaptive algorithm A can improve predictive analysis of, for example, eye chart data.

[0215] In some embodiments, algorithm A (adaptive algorithm) (20) improves predictive analysis of eye chart data by selecting optimal eye chart designs. In other words, algorithm A may select one or more eye chart designs for future eye tests of a particular subject based on the collected eye chart data of that particular subject.

[0216] Standard paper eye charts can be deterministic. All subjects are tested with the same paper eye chart, displayed with the same optotypes, and displayed in the same size. Printing different forms of eye chart designs that present the same optotype sizes but randomize the optotypes can prevent subjects from simply learning or guessing the chart. However, using the same set of limited optotype sizes can reduce the flexibility and accuracy of the test.

[0217] Adaptive visual acuity testing using the adaptive algorithms herein varies based on the subject's response, and in some embodiments, the visual acuity test is dynamically adjusted to focus on each subject. Recognizing the advantages of adaptive testing over deterministic testing, existing technology has moved to adaptive visual acuity testing using a single target. The adaptive testing approaches that dominate existing technology (up / down staircases; Dixon&Mood 1948; Zippy Estimation by Sequential Testing (ZEST); Parameter Estimation by Sequential Testing (PEST); and QUEST method) are only applicable to a single target. As a notable example, the full EDTRS visual acuity chart has been restructured as e-ETDRS, which combines electronic testing, single target presentation, and adaptive letter-by-letter scoring heuristics.

[0218] In some embodiments, algorithm A provides a new, more powerful, and more general approach to adaptive testing of visual acuity. Algorithm A can optimize visual acuity testing with a single visual target, a row of visual targets, or multiple rows of visual targets. Algorithm A can be powerful enough to optimize the complete visual acuity chart presented to the subject. In some embodiments, the adaptive algorithm disclosed herein includes one or more of the following steps: 1) define visual acuity chart design; 2) define visual acuity model for predicting visual acuity chart data; 3) define a set of candidate visual acuity model parameters; and 4) preliminary ranking of candidate visual acuity model parameters. A flow chart of an exemplary embodiment of algorithm A is shown in FIG. 9.

[0219] Unlike Algorithm S, which is applied to the analysis of previously collected visual acuity data, Algorithm A can be applied prospectively to improve and personalize the collection of visual acuity data based on prior information. To personalize the visual acuity chart to improve the quality of the data collected from a particular subject, several steps are common to Algorithm S.

[0220] In some embodiments, in step 2), a visual acuity model used in algorithm S may be defined. Rather than applying the visual acuity model to predict previously collected data, algorithm A may include a visual acuity model to predict optimal conditions for collecting future visual acuity chart data. Algorithm A may provide predictive analysis of which candidate visual acuity chart should be used to test the subject. In some embodiments, in step 3), a table of visual acuity model parameters used in algorithm S may be defined. The table includes candidate parameters that have the potential to describe visual acuity chart data collected from the subject. The visual acuity model parameters may be chart-specific or chart-invariant. In algorithm S, the ranking of visual acuity model parameters may be based on a likelihood value that describes the prediction of the visual acuity chart data as a function of the visual acuity model parameters. In some embodiments, in step 4), the preliminary ranking of visual acuity model parameters may be uniform or defined on any scale. In some embodiments, the ranked values ​​should be normalized to 1. This effectively makes the list of ranked values ​​a probability distribution. Before testing, all candidate parameters may have a uniform probability. Given the candidate parameters, a probability distribution can be constructed that describes the probability of the combination of the parameters describing the subject. To initiate Bayesian adaptive inference, the ranking of the candidate parameters can be converted to a probability distribution. The preliminary ranking of the candidate visual acuity parameters can be uniform. Alternatively, preliminary prior information can be used to seed knowledge about the visual acuity of the subject. Prior information can be obtained from, but is not limited to: Demographics such as age; b. Co-morbid conditions such as cataracts, AMD, or other conditions known to affect the visual system; c. Different vision tests; d. A reduced screening mode of the current test. For example, rather than reading the entire chart, the subject can read down the first column, or a combination thereof.

[0221] Once the ranking of the candidate parameters is converted to prior probabilities, testing can begin. Before the first test trial, there is no visual acuity data to consider and score. Therefore, the first application may consider the history of collected data. In this case, uniform ranking leads to a flat prior distribution.

[0222] In some embodiments, a single-target psychometric function is generated using eye chart data, and then an estimate of a subject's visual acuity threshold and range can be obtained based on the eye chart data predicted by the generated single-target psychometric function, given a goodness of fit to that data.

[0223] Referring to FIG. 9, in a particular embodiment, algorithm A (adaptive algorithm) (20) starts by determining (100) eye chart design parameters, if necessary. After the chart design is determined, candidate parameters (e.g., one or more visual acuity thresholds, visual acuity slopes, and / or visual acuity ranges) are selected to define (200), (300) an eye chart model, as similarly described in FIG. 1 and FIG. 3A and / or FIG. 3B. The eye chart model, as shown in Table 2, may be defined based on the subject's preliminary information for predicting eye chart data. In this case, an initial ranking of the eye chart models is generated (410) based on the eye chart model in operation (200) of FIG. 1, the parameters in operation (300), and / or the responses collected in operation (400). The set of candidate eye charts presented to the subject is at least partially determined (510) by the eye chart design function. Alternatively or in combination, the set of candidate eye charts presented to the subject is determined (510) by the subject's eye chart data or eye test data collected from a previous test. In some embodiments, the set of candidate vision charts presented to the subject is determined by other information of the test subject, such as demographic information, previous vision parameter estimates, previous vision test results, etc. Optionally, in this embodiment, each of the candidate vision charts of 510 is evaluated and ranked (610) based on the preliminary ranking of the vision model and vision parameters of (410). In the same case, one vision chart with the highest ranking (i.e., top 1, top 2, and / or top 3) is selected to be presented to the subject (710) and the adaptation algorithm ends. Optionally, a vision chart can be randomly selected from the top decile or quartile of the ranked vision charts.

[0224] For example, in addition to defining an eye chart design and defining a visual acuity model based on that design, it is necessary to define a set of candidate visual acuity model parameters to describe the eye chart data collected from the patient. As shown in FIG. 9, some steps common to algorithm S-eye chart design (100) and visual acuity model definition (200), candidate visual acuity model parameter definition (300), and steps different from those in algorithms S410, 510 and 610-analysis of the candidate visual acuity chart to generate an optimal visual acuity chart design (710) that can be used to test the visual acuity of the subject. For example, the optimal visual acuity chart may focus on a target size that may be close to the visual acuity threshold of the subject or within the visual acuity range of the subject.

[0225] Combinatorial Algorithms The systems and methods disclosed herein may include combination algorithms. Exemplary embodiments of the combination algorithms disclosed herein are shown in Figures 16-17. The systems and methods disclosed herein address the complexity of visual acuity metrics by adding more psychometric features by applying a combination of signal detection theory and probability theory. In some embodiments, the analysis herein considers how an implicit chart-invariant psychometric function defined by two sensitivity parameters, e.g., sensitivity threshold and sensitivity slope or sensitivity threshold and sensitivity range, can generate a complete family of explicit psychometric functions observed in visual acuity testing.

[0226] Referring to FIG. 16, in certain embodiments, the visual acuity test optionally begins (10) by collecting preliminary information about the subject (10) to be tested. In the same embodiment, algorithm A is applied (20), an eye chart is presented to the subject (30), and algorithm S is applied (40) to generate a visual acuity test score. If the results from algorithm S meet a termination condition (e.g., if the model parameters are above a determined threshold probability or if the Bayesian posterior variance is below a determined level of variation), the visual acuity test does not continue (50) and the test results (e.g., the value of the subject's estimated visual acuity parameters) are displayed (60). Optionally, if the results from algorithm S do not meet the termination criteria, the visual acuity test continues (50) by repeating one or more previously performed steps.

[0227] Referring to FIG. 17, in certain embodiments, the visual acuity test optionally begins by collecting preliminary information about the subject 10 to be tested. A preliminary visual acuity chart selected based on the preliminary information may then be presented to the subject (70), and a scoring algorithm S may be applied to generate a score for the subject's visual acuity (40). In the same embodiment, algorithm A is then applied to the user's score and / or the preliminary information (20), and the visual acuity chart adaptively generated by algorithm A is presented to the subject (30). After responses are collected from the subject, algorithm S is applied to generate a visual acuity test score for the subject (40). If the results from algorithm S meet the termination criteria, the visual acuity test does not continue (50), and the test results are displayed (60). Optionally, if the results from algorithm S do not meet the termination criteria, the visual acuity test continues by repeating one or more previously performed steps (50).

[0228] Signal detection theory (SDT) was established to describe the wartime performance of radar operators and to distinguish between perceptual-level factors that made operators sensitive to target detection and judgment-level factors that made operators liberal or conservative in their response to targets (Green & Swets, 1961; McMillan & Creelman, 2004, incorporated herein by reference in their entirety). The SDT framework provides a richer quantification of detection behavior beyond "correct" and "incorrect". SDT analysis considers how perceptual sensitivity (d') parameters and decision-level parameters contribute to predicting the observed probabilities of "hits", "misses", "false alarms", and "correct rejections". The distinction between sensitivity and decision factors, and their quantification, is critical to the development of modern psychophysics. It provides a framework for identifying the contribution of factors independent of the design of the psychophysical task to factors of response bias and decision criteria that are specific to the task design and the subject. Lesmes et al. (2015, incorporated herein by reference in its entirety) combined Bayesian adaptive algorithms and signal detection theory analysis to develop an adaptive method that provides threshold estimates independent of psychophysical tasks such as forced choice, simple detection, cued detection and nominal detection. In the present invention, an adaptive algorithm is elaborated to take into account the complex design of eye charts and provide a design-invariant measure of visual acuity that is independent of task factors and design principles specific to each eye chart.

[0229] When combined with design-dependent adaptive testing algorithms tuned to be optimized for a particular chart design, libraries of eye charts can be optimally sampled to accelerate and improve testing and analysis of visual acuity and changes therein.

[0230] Despite some of the complexities of the predictive or visual acuity model disclosed herein, it can be simplified and presented to patients and physicians as an indication of test results.

[0231] Verification of Combinatorial Algorithms Some exemplary aggregate simulation results of the combination algorithm are shown in FIG. 18. Convergence of adaptive visual acuity testing is reflected in the Bayesian posterior estimates obtained from a large number of simulated test runs. The widest posterior reflects the earliest visual acuity threshold estimate obtained from the patient's reading of the first column of the full 14-row ETDRS chart shown on slide 0. In the next chart shown on slide 1, the patient is presented with a smaller 5-row sub-chart that adheres to the ETDRS principles but represents a sub-chart that is specifically optimized for the patient. The Bayesian posterior distribution that results after testing with slide 1 reflects the Bayesian updating of the parameters of the visual acuity model. The reduction in the posterior variance reflects the success of the information acquisition criteria used for adaptive testing. Convergence of visual acuity estimates continues with the information obtained from the presentation of the optimized slide 2. The inset shows the rapid reduction in test variability with increasing number of test slides.

[0232] Figure 19 shows experimental results for adaptive chart-based visual acuity testing. Individuals with normal visual acuity are tested using a standard constrained adaptive testing algorithm that presents five chart slides in sequence, as needed. Each chart consists of five lines, and each line on the chart consists of five letters sampled without replacement from the Sloan set. Four test runs are completed, and the variability of the test results is calculated across the runs. The successful convergence of this adaptive test is reflected in the reduced variability of visual acuity threshold estimates with increasing numbers of test slides.

[0233] Inspection and response In some embodiments, the visual acuity test herein comprises at least one test run or at least one test slide, in which the subject is presented with a chart or sub-chart of indicators and the subject's appropriate response to the test questions is required to properly record the response, evaluate the response, score the response, estimate visual acuity parameters, and continue or terminate the current test.

[0234] In some embodiments, test responses are recorded separately for each test run within a single test.

[0235] In some embodiments, the visual indicia are sampled repeatedly from the library of visual indicia. In some embodiments, the visual indicia are sampled non-repetitively from the library of visual indicia.

[0236] In some embodiments, at least one optotype in the vision chart or sub-chart is generated using random sampling from an optotype library. In some embodiments, at least one optotype in the vision chart or sub-chart is generated using adaptive sampling from an optotype library.

[0237] In some embodiments, the expected value of information gain is calculated for all possible charts or sub-charts that will be presented at the next examination. In some embodiments, the expected entropy is calculated for all possible charts or sub-charts that will be presented at the next examination. In some embodiments, the chart, sub-chart, or slide with the smallest expected entropy is presented to the subject.

[0238] end In some embodiments, the test is terminated after meeting at least one of the termination criteria. In some embodiments, the visual acuity test is terminated after testing a predetermined number of test charts, subcharts, or slides. In some embodiments, the visual acuity test is terminated after meeting a preset accuracy level.

[0239] In some embodiments, for short test times, the test can be terminated after just one chart. Alternatively, for precision, the test can continue to refine optimal selection as more information is collected, making the test even more focused on the individual. The stopping rule for reading ETDRS charts is a heuristic, and patients are encouraged to progress until they can no longer read three of the five targets per line.

[0240] In some embodiments, rather than terminating the test, the test can continue its focus on the individual. An optimization algorithm can improve the test by focusing the test on the most informative areas of the eye chart. If the test is not terminated after reading the first chart (in the first trial), algorithm A can be applied to display another eye chart. If the test is terminated, the system or method can display the test results to the subject.

[0241] Displaying the inspection results With reference to FIG. 1, the results including the model parameters may be displayed to the user and / or the subject (800). In some embodiments, the results of the chart-based visual acuity test are visualized in a way that allows the subject, the physician, and / or the caregiver to intuitively understand the results. With reference to FIGS. 22A-22D, in some embodiments, the Bayesian posterior probability functions of the visual acuity parameters, i.e., visual acuity thresholds and / or ranges, generated across different visual acuity model parameters, are overlaid on the ETDRS visual acuity chart as a heat map (FIGS. 22A-22B). In some embodiments, the heat map of visual acuity parameters represents the average estimate of the visual acuity threshold parameters in the darkest gray, with decreasing shades of gray as the optotype size differs increasingly from the average optotype size. In some embodiments, the gradient of color change in the heat map reflects the gradient of visual acuity sensitivity. In some embodiments, disease progression or improvement with treatment is represented by multiple heat maps presented on the same chart. In certain embodiments, as in FIG. 22A, the results reflect lower confidence in the visual acuity threshold parameters compared to the results presented as in FIG. 22B. This may provide a useful presentation of visual acuity results corresponding to sizes between those presented in the ETDRS standard. In another specific embodiment, the Bayesian posterior probability is represented by a graphical box plot (FIG. 22C), which shows the mean, interquartile range, and finally the 95% confidence interval of the visual acuity threshold estimate. In another embodiment, as in FIG. 22D, the progression of visual acuity loss is shown by visual acuity metrics taken at different assessment times. "Time 1" represents the visual acuity threshold tested at the first time point, and "Time 2" represents the visual acuity threshold tested at the second time point.

[0242] With reference to FIG. 23, in certain embodiments, the predicted performance is overlaid on the eye chart for display. In this embodiment, the probability of correctly identifying fewer than three of the five total signs in each row is shown in black directly above each row of the eye chart. In this embodiment, the probability of correctly identifying three of the five total signs in each row is shown in dark gray. In this embodiment, the probability of correctly identifying more than three of the five total signs in each row is shown in light gray, and the probability of correctly identifying exactly three of the five total signs in each row is shown in medium gray. The probability is quantified by dividing the length of each gray bar by the length of all three bars under each column of signs.

[0243] In some embodiments, the visual acuity parameter estimates may be presented on an eye chart. Additionally, the results of several tests over time may be presented to conveniently show the progression of visual acuity over time.

[0244] For patient engagement and education, visualization of test results is important. In the case of progressive vision loss, for example, the degree and rate of vision loss are important factors. Especially for diabetes-related vision loss, since blood glucose compliance and control directly affect vision, visualization of test results may include rich vision-related statistics. These statistics may include one or more of the following: posterior probability distribution, parameter box plot, mean plus confidence, vision sensitivity parameter, vision threshold parameter / mean plus confidence, empirical or psychometric threshold and sensitivity threshold, mean / confidence interval, vision threshold and vision range, vision threshold and vision slope, degree of change, overlap showing probability of change, overlap of vision probability distribution.

[0245] Digital Processing Device In some embodiments, the platforms, systems, media, and methods described herein include a digital processing device, or the use thereof. In further embodiments, the digital processing device includes one or more hardware central processing units (CPUs) or general purpose graphics processing units (GPGPUs) that perform the functions of the device. In still further embodiments, the digital processing device further comprises an operating system configured to execute executable instructions. In some embodiments, the digital processing device is optionally connected to a computer network. In further embodiments, the digital processing device is optionally connected to the Internet so that it accesses the World Wide Web. In still further embodiments, the digital processing device is optionally connected to a cloud computing infrastructure. In other embodiments, the digital processing device is optionally connected to an intranet. In other embodiments, the digital processing device is optionally connected to a data storage device.

[0246] In accordance with the description herein, suitable digital processing devices include, by way of non-limiting example, server computers, desktop computers, laptop computers, notebook computers, subnotebook computers, netbook computers, netpad computers, set-top computers, media streaming devices, handheld computers, Internet appliances, mobile smart phones, tablet computers, personal digital assistants, video game consoles, and vehicles. Those skilled in the art will recognize that many smart phones are suitable for use in the systems described herein. Those skilled in the art will also recognize that selected televisions, video players, and digital music players, with optional computer network connectivity, are suitable for use in the systems described herein. Suitable tablet computers include computers having booklet, slate, and convertible configurations known to those skilled in the art.

[0247] In some embodiments, the digital processing device includes an operating system configured to execute executable instructions, which is software, including, for example, programs and data, that manages the device's hardware and provides services for the execution of applications.

[0248] In some embodiments, the device includes a storage and / or memory device. A storage and / or memory device is one or more physical devices used to temporarily or permanently store data or programs. In some embodiments, the device is a volatile memory and requires power to maintain the stored information. In some embodiments, the device is a non-volatile memory and retains the stored information when the digital processing device is not powered. In further embodiments, the non-volatile memory includes a flash memory. In some embodiments, the non-volatile memory includes a dynamic random access memory (DRAM). In some embodiments, the non-volatile memory includes a ferroelectric random access memory (FRAM®). In some embodiments, the non-volatile memory includes a phase change random access memory (PRAM). In other embodiments, the device is a storage device, including, by way of non-limiting examples, CD-ROMs, DVDs, flash memory devices, magnetic disk drives, magnetic tape drives, optical disk drives, and cloud computing-based storage. In further embodiments, the storage and / or memory device is a combination of devices such as those disclosed herein.

[0249] In some embodiments, the digital processing device includes a display for transmitting visual information to a user. In some embodiments, the display is a liquid crystal display (LCD). In further embodiments, the display is a thin film transistor liquid crystal display (TFT-LCD). In some embodiments, the display is an organic light emitting diode (OLED) display. In various further embodiments, the OLED display is a passive matrix OLED (PMOLED) or active matrix OLED (AMOLED) display. In some embodiments, the display is a plasma display. In other embodiments, the display is a video projector. In still other embodiments, the display is a head mounted display in communication with the digital processing device, such as a VR headset.

[0250] In some embodiments, the digital processing device comprises an input device for receiving information from a user. In some embodiments, the input device is a keyboard. In some embodiments, the input device is a pointing device, including, by way of non-limiting examples, a mouse, a trackball, a trackpad, a joystick, a game controller, or a stylus. In some embodiments, the input device is a touch screen or a multi-touch screen. In other embodiments, the input device is a microphone for capturing voice or other audio input. In other embodiments, the input device is a video camera or other sensor for capturing motion or visual input. In further embodiments, the input device is a Kinect, Leap Motion, or the like. In still further embodiments, the input device is a combination of devices, such as those disclosed herein.

[0251] Referring to FIG. 24, in certain embodiments, an exemplary digital processing device 2401 is programmed or configured to assess the visual acuity of a subject. The device 2401 can regulate various aspects of the algorithms and method steps of the present disclosure. In this embodiment, the digital processing device (2401) includes a central processing unit (CPU, also referred to herein as "processor" and "computer processor") (2405), which may be a single-core or multi-core processor, or multiple processors for parallel processing. The digital processing device (2401) also includes memory or memory locations (2410) (e.g., random access memory, read-only memory, flash memory), an electronic storage unit (2415) (e.g., hard disk), a communication interface (2420) (e.g., a network adapter) for communicating with one or more other systems, and peripherals (2425), such as cache, other memory, data storage, and / or electronic display adapters. The memory (2410), electronic storage unit (2415), interface (2420), and peripherals (2425) communicate with the CPU (2405) through a communication bus (solid lines) such as a motherboard. The electronic storage unit (2415) may be a data storage unit (or data repository) for storing data. The digital processing device (2401) may be operatively coupled to a computer network ("network") (2430) with the aid of a communication interface (2420). The network (2430) may be the Internet, an Internet and / or an extranet, or an intranet and / or an extranet in communication with the Internet. The network (2430) in some cases is a telecommunications and / or data network. The network (2430) may include one or more computer servers, which may enable distributed computing such as cloud computing.The network (2430), possibly with the help of the device (2401), may implement a peer-to-peer network, which may enable devices coupled to the device (2401) to operate as clients or servers.

[0252] Continuing to refer to FIG. 24, the CPU (2405) can execute a sequence of machine-readable instructions, which may be embodied in a program or software. The instructions may be stored in a memory location, such as memory (2410). The instructions may be directed to the CPU (2405), which may then program or configure the CPU (2405) to perform the methods of the present disclosure. Examples of operations performed by the CPU (2405) may include fetch, decode, execute, and write back. The CPU (2405) may be part of a circuit, such as an integrated circuit. One or more other components of the device (2401) may be included in the circuit. In some embodiments, the circuit is an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

[0253] 24, the storage unit (2415) can store files such as drivers, libraries, and saved programs. The storage unit (2415) can store user data, such as user preferences and user programs. The digital processing device (101) in some cases can include one or more additional data storage units that are external, such as located on a remote server communicating via an intranet or the Internet.

[0254] 24, the digital processing device (2401) can communicate with one or more remote computer systems over a network (2430). For example, the device (2401) can communicate with a user's remote computer system. Examples of remote computer systems include a personal computer (e.g., a portable PC), a slate or tablet PC (e.g., Apple® iPad®, Samsung® Galaxy Tab), a phone, a smartphone (e.g., Apple® iPhone®, Android-enabled devices, Blackberry®), or a personal digital assistant.

[0255] The methods described herein may be implemented by machine (e.g., computer processor) executable code stored on an electronic storage location of the digital processing device (2401), such as on a memory (2410) or electronic storage unit (2415). The machine executable or machine readable code may be provided in the form of software. During use, the code may be executed by the processor (2405). In some embodiments, the code may be retrieved from the storage unit (2415) and stored on the memory (2410) for immediate access by the processor (105). In some circumstances, the electronic storage unit (2415) may be omitted and the machine executable instructions are stored on the memory (2410).

[0256] Non-transitory computer-readable storage medium In some embodiments, the platform, system, medium, and method disclosed herein includes one or more non-transitory computer-readable storage media encoded with a program including instructions executable by an operating system of an optionally networked digital processing device. In further embodiments, the computer-readable storage medium is a tangible component of the digital processing device. In still further embodiments, the computer-readable storage medium is optionally removable from the digital processing device. In some embodiments, the computer-readable storage medium includes, as non-limiting examples, CD-ROMs, DVDs, flash memory devices, solid-state storage devices, magnetic disk drives, magnetic tape drives, optical disk drives, cloud computing systems and services, and the like. In some embodiments, the programs and instructions are encoded permanently, substantially permanently, semi-permanently, or non-transitory on the medium.

[0257] Computer Programs In some embodiments, the platform, system, medium, and method disclosed herein include at least one computer program, or the use thereof. A computer program is executable in a CPU of a digital processing device and includes a set of instructions written to perform a specific task. The computer-readable instructions may be implemented as program modules, such as functions, objects, application programming interfaces (APIs), data structures, etc., that perform a specific task or implement a specific abstract data type. In light of the disclosure provided herein, those skilled in the art will recognize that computer programs may be written in a variety of languages ​​and in a variety of versions.

[0258] The functionality of the computer readable instructions may be combined or distributed as desired in various environments. In some embodiments, the computer program comprises one sequence of instructions. In some embodiments, the computer program comprises multiple sequences of instructions. In some embodiments, the computer program is provided from one location. In other embodiments, the computer program is provided from multiple locations. In various embodiments, the computer program comprises one or more software modules. In various embodiments, the computer program comprises, in part or in whole, one or more web applications, one or more mobile applications, one or more standalone applications, one or more web browser plug-ins, extensions, add-ins, or add-ons, or combinations thereof.

[0259] Web Applications In some embodiments, the computer program comprises a web application. In light of the disclosure provided herein, one skilled in the art will recognize that web applications, in various embodiments, utilize one or more software frameworks and one or more database systems. In some embodiments, the web application is created on a software framework such as Microsoft® .NET or Ruby on Rails (RoR). In some embodiments, the web application utilizes one or more database systems, including, by way of non-limiting examples, relational, non-relational, object-oriented, associative, and XML database systems. In further embodiments, suitable relational database systems include, by way of non-limiting examples, Microsoft® SQL Server, mySQL®, and Oracle®. One skilled in the art will recognize that web applications, in various embodiments, are written in one or more versions of one or more languages. Web applications may be written in one or more markup languages, presentation definition languages, client-side scripting languages, server-side coding languages, database query languages, or combinations thereof. In some embodiments, a web application is written in part in a markup language such as Hypertext Markup Language (HTML), Extensible Hypertext Markup Language (XHTML), or eXtensible Markup Language (XML). In some embodiments, a web application is written in part in a presentation definition language such as Cascading Style Sheets (CSS). In some embodiments, a web application is written in part in client-side scripting such as Asynchronous Javascript and XML (AJAX), Flash® Actionscript, Javascript, or Silverlight®.In some embodiments, the web application is written in part in a server-side coding language, such as Active Server Pages (ASP), ColdFusion®, Perl, Java®, JavaServer Pages (JSP), Hypertext Preprocessor (PHP), Python®, Ruby, Tcl, Smalltalk, WebDNA®, or Groovy. In some embodiments, the web application is written in part in a database query language, such as Structured Query Language (SQL). In some embodiments, the web application integrates with an enterprise server product, such as IBM® Lotus Domino®. In some embodiments, the web application includes a media player element. In various further embodiments, the media player element utilizes one or more suitable multimedia technologies, including, by way of non-limiting examples, Adobe® Flash®, HTML 5, Apple® QuickTime®, Microsoft® Silverlight®, Java®, and Unity®.

[0260] Referring to FIG. 25, in a particular embodiment, the application provisioning system comprises one or more databases (2500) accessed by a relational database management system (RDBMS) (2510). Suitable RDBMSs include Firebird, MySQL, PostgreSQL, SQLite, Oracle Database, Microsoft SQL Server, IBM DB2, IBM Informix, SAP Sybase, Teradata, and the like. In this embodiment, the application provisioning system further comprises one or more application servers (2520) (e.g., Java server, .NET server, PHP server, and the like) and one or more web servers (2530) (e.g., Apache, IIS, GWS, and the like). The web servers optionally expose one or more web services via app application programming interfaces (APIs) (2540). Over a network such as the Internet, the system provides a browser-based and / or mobile native user interface.

[0261] Referring to FIG. 26, in a particular embodiment, the application delivery system alternatively has a distributed cloud-based architecture (2600) with elastically load balanced, auto-scaling web server resources (2610) and application server resources (2620), and a synchronously replicated database (2630).

[0262] Software Module In some embodiments, the platforms, systems, media, and methods disclosed herein include software, server, and / or database modules, or the use thereof. In view of the disclosure provided herein, the software modules are created by techniques known to those skilled in the art using machines, software, and languages ​​known in the art. The software modules disclosed herein are implemented in a number of ways. In various embodiments, the software modules include files, sections of code, programming objects, programming structures, or combinations thereof. In further various embodiments, the software modules include multiple files, multiple sections of code, multiple programming objects, multiple programming structures, or combinations thereof. In various embodiments, the one or more software modules include, by way of non-limiting examples, web applications, mobile applications, and standalone application applications. In some embodiments, the software modules are in one computer program or application. In other embodiments, the software modules are in more than one computer program or application. In some embodiments, the software modules are hosted on one machine. In other embodiments, the software modules are hosted on more than one machine. In further embodiments, the software modules are hosted on a cloud computing platform. In some embodiments, the software modules are hosted on one or more machines in one location. In other embodiments, the software modules are hosted on one or more machines in more than one location.

[0263] Database In some embodiments, the platforms, systems, media, and methods disclosed herein include one or more databases or their use. In view of the disclosure provided herein, one skilled in the art will recognize that many databases are suitable for storing and retrieving vision charts, vision subcharts, subject preliminary information, subject chart data, input and / or output of the algorithms herein, and the like. In various embodiments, suitable databases include, by way of non-limiting examples, relational databases, non-relational databases, object-oriented databases, object databases, entity-relationship model databases, associative databases, and XML databases. Further non-limiting examples include SQL, PostgreSQL, MySQL, Oracle, DB2, and Sybase. In some embodiments, the database is Internet-based. In further embodiments, the database is web-based. In yet further embodiments, the database is cloud computing-based. In other embodiments, the database is based on one or more local computer storage devices.

[0264] Although specific embodiments and examples are provided in the foregoing description, the subject matter of the present invention extends beyond the specifically disclosed embodiments to other alternative embodiments and / or uses, as well as modifications and equivalents thereof. Thus, the claims appended hereto are not limited by any of the specific embodiments described herein. For example, in the methods disclosed herein, operations may be performed in any suitable sequence and are not necessarily limited to the specific disclosed sequence. Various operations may be described as multiple separate operations in sequence, in a manner that may be helpful in understanding a particular embodiment. However, the order of description should not be construed to imply that these operations are order-dependent. In addition, the systems and / or devices described herein may be embodied as integrated components or as separate components. EXAMPLES

[0265] The following illustrative examples are representative of embodiments of the software applications, systems, and methods described herein and are not intended to be limiting. Example 1

[0266] The computer-based medical system disclosed herein is used for testing and analysis of visual acuity in diabetic patients by presenting the subject with three of ten Sloan letters at a time on a digital screen. The prior information used to initiate the test is a population-based study of visual acuity. Using these population estimates, algorithm A is applied to present the subject with the top single-line subchart of each test slide. After applying algorithm S to evaluate the visual acuity chart data collected on each slide, with respect to the visual acuity models predicting the respective probabilities of correctly reporting 0, 1, 2, or 3 letters on each slide, the next slide is selected based on the iterative application of algorithm A. When the patient returns for the next test, the test can begin with information from the population-based study of visual acuity or alternatively consider data from the previous test session. Example 2

[0267] A computer-based medical device that applies combination algorithms to test visual acuity with a personalized version of the full Snellen or ETDRS eye chart. Current testing with paper eye charts is based on a single read or "run" through the full eye chart as the subject reads from the top of the chart to the bottom. In practice, when testing visual acuity in two eyes, it is common to ask the subject to read a single complete eye chart, first with one eye and then with the other. This confusion, which prevents independent testing of the two eyes, would be addressed and corrected by the personalization of different eye charts for testing each eye.

[0268] As with the above applications, general or specific clinical information may be used to initiate algorithm A. Based on that information, algorithm A is applied to select a personalized Snellen or ETDRS design. Following scoring by algorithm S, the test may end after display of one complete chart, or the test may continue with the generation of a second complete chart based on iterative application of algorithm A.

[0269] By presenting a complete visual acuity chart that follows the design principles of the ETDRS or Snellen chart, this adaptive visual acuity test conforms to current standards and established practices, personalizing visual acuity testing and enabling novel algorithms that allow high-resolution sampling of optotype sizes, thereby improving its accuracy. Example 3

[0270] A web-based software service that enables retrospective analysis of eye chart data. Customers use this web interface to apply algorithm S to analyze visual acuity data previously collected using an established vision chart design. Current analysis of these data provides qualitative results with limited precision. This application of the scoring algorithm provides the richer information necessary to detect changes in visual acuity.

[0271] This is achieved by detecting changes in sensitivity parameters that are independent of the details of the chart design. When examining changes in contrast sensitivity, Hou et al. 2016 (incorporated herein by reference in its entirety) proposed a new Bayesian signal detection analysis that uses Bayesian posterior distributions estimated in independent experimental conditions to calculate the area under the ROC to characterize the change between two experimental conditions. They calculated the Bayesian posterior distribution for a one-dimensional summary metric provided by the area under the logarithmic contrast sensitivity function (AULCSF).

[0272] The user evaluates the change in visual acuity using the area under the ROC analysis to calculate the probability that the visual acuity threshold in one test is greater than the visual acuity threshold in the second test. In some applications, the results of the visual acuity model being evaluated are obtained in different eye tests, in pre-treatment vs. post-treatment tests, or in visual acuity tests at low and high luminance. If necessary, the signal detection analysis can be extended to two dimensions to evaluate visual changes in two parameters of a chart-invariant psychometric function. This two-dimensional signal detection would detect changes in visual acuity threshold, or visual acuity range, or both. In addition, the scoring algorithm can be applied to score data from other tasks, including contrast sensitivity tests and reading function, including letter discrimination as a function of other stimulus parameters. Rather than defining the visual acuity chart design as a function of optotype size, these visual and cognitive tasks measure performance as a combination of other design parameters that can be quantified and estimated. Example 4

[0273] In the mobile-based application, algorithm S is applied to analyze visual acuity data during its active collection in the clinic with established chart designs. To align the definition of the visual acuity chart design required for scoring with the chart design in clinical use, the application presents the user with a library of visual acuity chart designs. When presented with the library, the user indicates which chart is being used to collect the data. Given the limited number of visual acuity charts currently on the market, and the tendency of clinicians to use the same paper charts for years, it is possible to build a small to medium-sized library of visual acuity chart designs in active circulation.

[0274] After the eye chart design is defined for scoring, the app and mobile device act as a response remote to mark the correctly or incorrectly reported eyes. In another implementation, given the appropriate eye chart design by the user, the subject can use an interface that allows for unsupervised testing.

[0275] A population database is constructed to record eye chart data. Once eye chart data is collected and scored, analysis can be uploaded to server. The resulting database provides the basis for machine learning and big data analysis, improves eye chart model, and improves preliminary information used to start combination algorithm in examples 3 and 4. In these two examples, scoring and analysis of retrospective data, and prospective collection of new data greatly improves computational framework. Example 5

[0276] The software service aggregates and analyzes retrospective and predictive analysis of the eye chart data generated by the hardware, software, and services defined in Examples 1-4. Analysis of these big data using algorithm S provides a foundation to better inform the preliminary information used to initiate algorithm A. For longitudinal visual acuity data collected over time, the software service calculates change indices for visual acuity. For example, in addition to calculating a running index that estimates the probability of visual acuity decline between sessions, a global change index is calculated to estimate the probability of visual acuity change relative to an initial baseline. Example 6

[0277] A computer-based medical device that executes algorithms S and A in combination with a visual acuity model that directly measures interocular differences in visual acuity. The eyes are stimulated independently, such as in a phoropter, stereoscope, or virtual reality headset. The visual acuity model is implemented with a total of three or four visual acuity model parameters, as desired. For example, the visual acuity model includes a visual acuity threshold and a visual acuity range for the left eye. Additionally, the model includes two difference parameters that define the difference in visual acuity threshold between the left and right eyes, respectively.

[0278] Unless otherwise defined, all technical terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs. As used herein and in the appended claims, the singular forms "a", "an" and "the" include plural references unless the context clearly dictates otherwise. References to "or" herein are intended to encompass "and / or" unless otherwise indicated. As used herein and in the claims, unless otherwise indicated, the terms "about" and "approximately" refer to a variation of + / -1%, + / -2%, + / -3%, + / -4%, + / -5%, + / -6%, + / -7%, + / -8%, + / -9%, + / -10%, + / -11%, + / -12%, + / -14%, + / -15%, or + / -20%, depending on the embodiment. As a limiting example, approximately 100 meters may represent a range of 95 meters to 105 meters, 90 meters to 110 meters, or 85 meters to 115 meters, depending on the embodiment. As a non-limiting example, approximately 100 meters may represent a range of 95 meters to 105 meters, 90 meters to 110 meters, or 85 meters to 115 meters, depending on the embodiment.

[0279] All publications, patents, and patent applications mentioned in this specification are incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.

[0280] While preferred embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions will occur to those skilled in the art without departing from the invention. It should be understood that various alternatives to the embodiments of the invention described herein can be used in practicing the invention.

Claims

1. 1. A computer-implemented system for generating a visual acuity model for scoring a subject's visual acuity, comprising: The computer-implemented system includes: a digital processing device including an operating system and memory configured to execute executable instructions; and a computer program, the computer program comprising: a) obtaining one or more eye chart design parameters; b) collecting subject vision test data; c) selecting a first set of chart-specific visual acuity parameters from one or more sets of candidate visual acuity parameters; d) generating a visual acuity model, i. generating a psychometric function for a single optotype including a first set of chart-specific visual acuity parameters; ii. Using the psychometric function of a single target to calculate a family of psychometric functions of multiple targets; iii. generating a visual acuity model using the family of psychometric functions of the plurality of visual targets to generate a chart-specific psychometric function including a second set of chart-specific visual acuity parameters; a scoring application including a software module for performing a scoring method according to claim 1, A computer-implemented system, wherein the visual acuity model is configurable to estimate the probability of obtaining visual acuity test data for a subject based on one or more visual acuity chart design parameters.

2. 2. The computer-implemented system of claim 1, further comprising after (c) and before (i), generating a sensitivity-based psychometric function including a first set of table-invariant sensitivity parameters, wherein the first set of table-invariant sensitivity parameters include a sensitivity threshold and a sensitivity range, and wherein the first set of table-invariant sensitivity parameters are generated based on the first set of chart-specific visual acuity parameters.

3. 3. The computer-implemented system of claim 2, wherein the sensitivity-based psychometric function is independent of one or more eye chart design parameters.

4. 3. The computer-implemented system of claim 2, wherein the sensitivity-based psychometric function is generated based on one or more eye chart design parameters and one or more additional parameters of the subject, the one or more additional parameters being table invariant.

5. The computer-implemented system of claim 2 , wherein the sensitivity-based psychometric functions are generated based on signal detection theory.

6. 3. The computer-implemented system of claim 2, wherein the sensitivity-based psychometric function is configured to describe a subject's visual acuity performance as a function d' of one or more target sizes and independent of one or more chart design parameters.

7. 3. The computer-implemented system of claim 2, further comprising, prior to (i), converting the sensitivity-based psychometric function to a single target psychometric function.

8. The computer-implemented system of any one of claims 1-7, wherein the first set of chart-specific visual acuity parameters includes a visual acuity threshold and a visual acuity range.

9. The computer-implemented system of any one of claims 1-8, wherein the psychometric function of a single target is chart specific.

10. The computer-implemented system of any one of claims 1-9, wherein the second set of chart-specific visual acuity parameters includes a visual acuity threshold and a visual acuity range.

11. 11. The computer-implemented system of claim 1, wherein the first set or the second set of chart-specific visual acuity parameters comprises a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof.

12. The computer-implemented system of any one of claims 1-11, wherein the first set or the second set of chart-specific visual acuity parameters comprises one or more numerical values.

13. The computer-implemented system of any one of claims 1-12, wherein the first set or the second set of chart-specific visual acuity parameters comprises at least one posterior probability density function for that parameter.

14. The computer-implemented system of any one of claims 1-13, wherein the subject's vision testing data includes data from at least two different testing conditions.

15. 15. The computer-implemented system of claim 14, wherein the first set or the second set of chart-specific visual acuity parameters comprises posterior probability density functions for visual acuity thresholds from at least two different testing conditions.

16. 16. The computer-implemented system of any one of claims 1-15, wherein the one or more optotype chart design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof.

17. The computer-implemented system of any one of claims 1-16, wherein the psychometric function of the single target is based on at least one target size.

18. 18. The computer-implemented system of claim 1, wherein each function in the family of psychometric functions of a plurality of optotypes is based on two or more different optotype sizes, and the family of psychometric functions of a plurality of optotypes includes psychometric functions of 2, 3, 4, 5, 6, or any other integer number of optotypes.

19. The computer-implemented system of any one of claims 1-18, wherein the psychometric function of the single target includes a visual acuity threshold and a visual acuity range of the subject.

20. 20. The computer-implemented system of claim 1, wherein one or more functions in the family of psychometric functions for multiple optotypes are based on the number of optotypes per row in the eye chart or sub-chart.

21. The computer-implemented system of any one of claims 1-20, wherein (iii) further comprises using a weighted sum of a family of psychometric functions of a plurality of optotypes.

22. 22. The computer-implemented system of claim 21, wherein the weighting of the family of psychometric functions is based on a subject's eye chart data.

23. 22. The computer implemented system of claim 21, wherein the weightings are determined manually by a user or automatically by a computer program.

24. 24. The computer-implemented system of claim 1, further comprising: (b) displaying one or more eye charts generated based on the one or more eye chart design parameters to collect eye test data.

25. 25. The computer-implemented system of claim 24, wherein each of the one or more eye charts includes at least one optotype, the at least one optotype selected from a library of optotypes.

26. 26. The computer-implemented system of any one of claims 24-25, wherein the one or more eye charts are of contrast that is approximately 25%, 2.5%, or 1.5% of normal contrast on a standard ETDRS or Sloane chart.

27. 27. The computer implemented system of any one of claims 24-26, wherein at least one indicium is a letter, a number, or a symbol.

28. The computer-implemented system of any one of claims 1-27, wherein the one or more sets of candidate visual acuity parameters are generated based on one or more eye chart design parameters, visual acuity test data of the subject, or both.

29. 1. A computer system for generating a visual acuity model for scoring a subject's visual acuity, comprising: The computer system includes: a digital processing device including an operating system and memory configured to execute executable instructions; and a computer program comprising: a) obtaining one or more eye chart design parameters; b) collecting subject vision test data; c) generating one or more sets of candidate visual acuity parameters based on one or more eye chart design parameters, visual acuity test data of the subject, or both; d) generating a visual acuity model including generating one or more chart-specific psychometric functions, each chart-specific psychometric function including a set of chart-specific visual acuity parameters of the one or more sets of chart-specific visual acuity parameters; e) generating a probability of observing the vision test data using the visual acuity model and one or more sets of candidate visual acuity parameters; f) ranking one or more sets of candidate visual acuity parameters of the subject based on probability; and g) selecting a set from the one or more sets of candidate visual acuity parameters for the subject based on the rank; A computer system comprising instructions executable by a digital processing device to create a scoring application including a software module, the computer system comprising:

30. 30. The computer system of claim 29, wherein each set of candidate visual acuity parameters includes a visual acuity threshold and a visual acuity range.

31. A computer system according to any one of claims 29-30, wherein each set of candidate visual acuity parameters includes: a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or any combination thereof.

32. A computer system according to any one of claims 29-31, wherein each set of candidate visual acuity parameters comprises a posterior probability density function for one or more visual acuity parameters.

33. A computer system according to any one of claims 29-32, wherein each set of candidate visual acuity parameters includes a posterior probability density function for each of the one or more visual acuity parameters.

34. 34. The computer system of any one of claims 29-33, wherein (b) comprises summarizing the visual acuity test data from the subjects and storing the visual acuity test data in a data table in memory.

35. The computer system of any one of claims 29-34, further comprising, after (g), displaying results of the visual acuity test to the subject or user using a digital display, the results of the visual acuity test including the selected set of candidate visual acuity parameters.

36. 36. The computer system of claim 29, wherein the subject's vision testing data includes data from only the first testing condition.

37. 37. The computer system of any one of claims 29-36, wherein the subject's vision testing data includes data from first and second testing conditions.

38. 38. The computer system of claim 37, wherein the selected set of candidate visual acuity parameters comprises a first posterior probability density function of a visual acuity threshold or a visual acuity range.

39. 40. The computer system of claim 38, wherein b) comprises selecting data having the first test condition.

40. The computer system of claim 39, further comprising, after g), selecting data having a second test condition from the subject's visual acuity test data; and repeating c) to g).

41. 41. The computer system of claim 40, wherein the selected set of candidate visual acuity parameters includes a second posterior probability density function for a visual acuity threshold or visual acuity range.

42. 42. The computer system of claim 41, further comprising using the first and second posterior probability density functions to obtain a difference distribution.

43. 43. The computer system of claim 42, further comprising calculating a change in acuity threshold or acuity range between the first and second test conditions based on the difference distribution.

44. 36. The computer system of claim 35, wherein the results of the vision test are table invariant or chart specific.

45. The computer system of any one of claims 29-44, wherein (e) is based on Bayes' theorem or maximum likelihood estimation.

46. 34. The computer system of claim 32 or 33, wherein the posterior probability density function is a Bayesian posterior probability density function.

47. 42. The computer system of claim 38 or 41, wherein the first and second posterior probability density functions are Bayesian posterior probability density functions.

48. (d) is: generating a psychometric function of a single optotype including a first set of chart-specific visual acuity parameters; Using the psychometric function of a single target, calculate a family of psychometric functions of multiple targets; and A computer system according to any one of claims 29-47, comprising generating a chart-specific psychometric function comprising a second set of chart-specific visual acuity parameters using a family of psychometric functions of a plurality of optotypes.

49. 49. The computer system of claim 48, further comprising generating a sensitivity-based psychometric function including a first set of table-invariant sensitivity parameters prior to generating the psychometric function for the single visual target.

50. 50. The computer system of claim 49, wherein the sensitivity-based psychometric function is independent of one or more optotype design parameters.

51. A computer system according to any one of claims 49-50, wherein the first set of chart-invariant sensitivity parameters comprises a sensitivity threshold and a sensitivity range.

52. 52. The computer system of any one of claims 49-51, wherein the sensitivity-based psychometric function is generated based on one or more optotype chart design parameters and / or one or more additional subject-based parameters, the one or more additional parameters being table-invariant.

53. 53. The computer system of any one of claims 49-52, wherein the sensitivity-based psychometric functions are generated based on signal detection theory.

54. 54. A computer system according to any one of claims 50-53, wherein the sensitivity-based psychometric function is configured to describe a subject's visual acuity performance as a function d' of one or more target sizes and independent of one or more chart design parameters.

55. 55. The computer system of any one of claims 49-54, further comprising converting the sensitivity-based psychometric function to a single target psychometric function.

56. A computer system according to any one of claims 29-55, wherein the set of chart-specific visual acuity parameters includes a visual acuity threshold and a visual acuity range.

57. 49. The computer system of claim 48, wherein the psychometric functions of a single target are chart specific.

58. 58. The computer system of any one of claims 29-57, wherein the set of chart-specific visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof.

59. A computer system according to any one of claims 29-58, wherein the set of chart-specific visual acuity parameters comprises one or more numerical values.

60. 60. A computer system according to any one of claims 29-59, wherein the set of chart-specific visual acuity parameters includes at least one posterior probability density function for that parameter.

61. 61. The computer system of any one of claims 29-60, wherein the subject's vision testing data includes data from at least two different testing conditions.

62. 62. The computer system of claim 61, wherein the set of chart-specific visual acuity parameters includes posterior probability density functions for visual acuity thresholds for at least two different testing conditions.

63. 63. The computer system of any one of claims 29-62, wherein the one or more optotype chart design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof.

64. 49. The computer system of claim 48, wherein the psychometric function of the single target is based on at least one target size.

65. 65. A computer system according to any one of claims 48-64, wherein each function in the family of psychometric functions for a plurality of targets is based on two or more different target sizes.

66. 66. A computer system according to any one of claims 48-65, wherein the psychometric function of the single target comprises a visual acuity threshold and a visual acuity range of the subject.

67. 67. The computer system of any one of claims 48-66, wherein one or more functions in the family of psychometric functions for a plurality of optotypes are based on the number of optotypes per row in the eye chart or sub-chart.

68. 68. The computer system of any one of claims 48-67, wherein generating the chart-specific psychometric function further comprises using a weighted sum of a family of psychometric functions of multiple optotypes.

69. 69. The computer system of claim 68, wherein the weighted sum of the family of psychometric functions is based on a subject's eye chart data.

70. 69. The computer system of claim 68, wherein the weightings are determined manually by a user or automatically by a computer program.

71. 71. The computer system of claim 29, further comprising: (b) displaying to the subject one or more eye charts generated based on the one or more eye chart design parameters.

72. 72. The computer system of claim 71, wherein the one or more eye charts include at least one optotype, the at least one optotype selected from a library of optotypes.

73. 72. The computer system of claim 71, wherein the one or more eye charts are of contrast that is approximately 25%, 2.5%, or 1.5% of normal contrast on a standard ETDRS or Sloane chart.

74. 73. The computer system of claim 72, wherein at least one of the indicia is a letter, a number, or a symbol.

75. 1. A computer-implemented method for adaptively assessing visual acuity of a subject, comprising: The system is a digital processing device including an operating system and memory configured to execute executable instructions; and a computer program comprising: a) obtaining one or more eye chart design parameters; b) collecting subject vision test data; c) generating one or more sets of candidate visual acuity parameters based on the one or more eye chart design parameters and the subject's visual acuity testing data, each of the one or more sets of candidate visual acuity parameters including a rank obtained by analyzing the subject's visual acuity testing data; d) generating a plurality of candidate eye charts based on the one or more test chart design parameters; e) ranking the plurality of candidate visual acuity charts based on a rank of each of the one or more sets of candidate visual acuity parameters; and f) selecting one or more eye charts from a plurality of candidate eye charts based on a rank presented to the subject for the eye chart; A computer-implemented method comprising instructions executable by a digital processing device to create a scoring application comprising a software module, the computer-implemented method comprising:

76. 76. The system of claim 75, further comprising, after f), presenting the subject with one or more selected eye charts using a digital display.

77. 77. The system of claim 76, further comprising, after f), updating the vision test data with new data collected from the subject using one or more selected vision charts.

78. presenting one or more selected eye charts to the subject using a digital display; 78. The system of any one of claims 75-77, further comprising: updating the vision testing data with new data collected from the subject using one or more selected vision charts; and repeating b) through f); until a stopping criterion is met.

79. A system according to any one of claims 75-78, wherein each set of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold and a visual acuity range.

80. 80. The system of any one of claims 75-79, wherein each set of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof.

81. A system according to any one of claims 75-80, wherein each set of the one or more sets of candidate visual acuity parameters comprises a posterior probability density function for the one or more visual acuity parameters.

82. A system according to any one of claims 75-81, wherein each set of the one or more sets of candidate visual acuity parameters comprises a posterior probability density function for each of the one or more visual acuity parameters.

83. The system of any one of claims 75-82, wherein (b) summarizes the subject's visual acuity test data.

84. 84. The system of any one of claims 75-83, wherein the subject's vision testing data includes data from only the first testing condition.

85. A system according to any one of claims 75-84, wherein the subject's vision testing data includes data from first and second testing conditions.

86. 86. A system as described in any one of claims 75-85, wherein each set of the one or more sets of candidate visual acuity parameters includes a visual acuity threshold, a visual acuity range, a visual acuity slope, a change in visual acuity threshold between two test conditions, a change in visual acuity range between two test conditions, a change in visual acuity slope between two test conditions, or a combination thereof.

87. A system according to any one of claims 75-86, wherein the subject's vision testing data includes data from at least two different testing conditions.

88. A system according to any one of claims 75-87, wherein each set of the one or more sets of candidate visual acuity parameters comprises a posterior probability density function for a visual acuity threshold or visual acuity range.

89. 88. The system of any one of claims 75-87, wherein the one or more optotype chart design parameters include: total number of optotypes, number of optotypes per row, method of optotype sampling, optotype size, number of rows, and subject response, or a combination thereof.

90. 90. A system according to any one of claims 75-89, wherein each of the candidate eye charts includes at least one optotype, the at least one optotype being selected from a library of optotypes.

91. 91. A system according to any one of claims 75-90, wherein one or more of the candidate eye charts are of contrast that is approximately 25%, 2.5%, or 1.5% of normal contrast on a standard ETDRS or Sloane chart.

92. 91. The system of claim 90, wherein at least one indicium is a letter, a number, or a symbol.

93. a) generating a plurality of candidate eye charts based on one or more test chart design parameters; and 30. The system of claim 29, further comprising: b) selecting one or more of the plurality of candidate eye charts for visual acuity testing of the subject based on a ranking of the plurality of candidate eye charts, the ranking being based on a selected set of candidate visual acuity parameters for the subject.

94. The system of claim 93, further comprising: after (b); collecting data from the subject using the selected one or more candidate eye charts; and if the stopping criterion is not met, repeating steps c) to g) of claim 29 and steps a) to b) of claim 93 until the stopping criterion is met.

95. 1. A computer-implemented method for generating a visual acuity model for scoring a subject's visual acuity, comprising: The method comprising: a) obtaining one or more eye chart design parameters; b) collecting visual acuity test data for the subject; c) selecting a first set of chart-specific visual acuity parameters from one or more sets of candidate visual acuity parameters; d) generating a visual acuity model, i) generating a psychometric function for a single optotype including a first set of chart-specific visual acuity parameters; ii) calculating a family of psychometric functions of multiple targets using the psychometric function of a single target; and iii) generating a chart-specific psychometric function using the family of psychometric functions of a plurality of optotypes, the chart-specific psychometric function including a second set of chart-specific visual acuity parameters; A computer-implemented method, wherein the visual acuity model is configurable to estimate the probability of obtaining visual acuity test data for a subject based on one or more eye chart design parameters.

96. 1. A computer-implemented method for scoring visual acuity in a subject, comprising: The method comprising: a) obtaining one or more eye chart design parameters; b) collecting visual acuity test data for the subject; c) generating one or more sets of candidate visual acuity parameters based on one or more eye chart design parameters, visual acuity test data of the subject, or both; d) generating a visual acuity model including generating one or more chart-specific psychometric functions, each chart-specific psychometric function including a set of chart-specific visual acuity parameters of the one or more sets of chart-specific visual acuity parameters; e) generating a probability of observing the vision test data using the visual acuity model and one or more sets of candidate visual acuity parameters; f) ranking the one or more sets of candidate visual acuity parameters of the subject based on probability; and g) selecting a set from the one or more sets of candidate vision parameters for the subject based on the ranking.

97. 1. A computer-implemented method for adaptively assessing visual acuity of a subject, comprising: The method comprising: a) obtaining one or more eye chart design parameters; b) collecting visual acuity test data for the subject; c) generating one or more sets of candidate visual acuity parameters based on the one or more vision chart design parameters and the subject's visual acuity test data, each of the one or more sets of candidate visual acuity parameters comprising a rank obtained by analyzing the subject's visual acuity test data; d) generating a plurality of candidate eye charts based on one or more test chart design parameters; e) ranking the plurality of candidate visual acuity charts based on the rank of each of the one or more sets of candidate visual acuity parameters; and f) selecting one or more eye charts from a plurality of candidate eye charts based on a rank presented to the subject for the eye test.

Images