Remote self-vision screening method based on visual target presentation device

Abstract

The application discloses a remote self-service vision screening method based on a visual target presentation device and belongs to the technical field of vision detection. The visual target presentation device receives a screening task configuration issued by a cloud, retrieves a personalized visual target presentation sequence according to the identity of a person to be examined, collects a response time and a correct recognition rate in real time during the detection process, dynamically adjusts the size and presentation rhythm of subsequent visual targets, uploads the result after the detection is completed, and generates a vision change trend evaluation. The method uses historical data to set the initial visual target level, adjusts the visual target parameters according to the response performance, effectively shortens the screening time, and improves the accuracy of the detection process and result.

Description

Technical Field

[0001] This invention relates to the field of vision testing technology, specifically a remote self-service vision screening method based on optotype presentation devices. Background Technology

[0002] Remote self-service vision screening methods typically rely on optotype presentation devices to display a series of optotypes in a fixed order. After collecting the subject's identification results, the visual acuity level is determined according to the rules of a standard visual acuity chart. This type of method requires all subjects to start the test from the same initial optotype size, and the presentation rhythm and size change step size of the optotypes are all set to uniform preset values, without any adaptive adjustments based on individual differences of the subjects or their actual performance during the screening process.

[0003] The existing technical solutions have two main drawbacks. First, the standardized initial optotype size does not take into account the visual fluctuations experienced by the examinee during previous screenings. This results in examinees with relatively stable vision still having to endure a large number of invalid optotypes far below their resolution threshold, prolonging the screening time, increasing examinee fatigue, and negatively impacting the cooperation of younger or older examinees. Second, the fixed size adjustment step size and fixed presentation interval cannot match the real-time changes in the examinee's attention state and recognition speed during the test. When the examinee's recognition ability is strong, redundant trials are likely to occur. When the recognition ability is close to the threshold, the fine-level information may be lost due to excessively large step sizes or too fast paces, ultimately affecting the accuracy of the vision screening results and making it difficult to distinguish between accidental errors caused by attention fluctuations and the true resolution boundary.

[0004] The problem to be solved by this invention focuses on how to provide an individualized starting optotype level for each screening based on the examinee's historical screening data, and how to adjust the direction of optotype size change and optotype presentation interval in real time according to the examinee's actual response time and recognition accuracy during the test, so as to achieve efficient and accurate vision assessment in remote self-service scenarios. Summary of the Invention

[0005] The purpose of this invention is to provide a remote self-service vision screening method based on optotype presentation devices. This method can adaptively determine the starting optotype level based on the examinee's previous screening records to generate a personalized optotype presentation sequence. It can also dynamically adjust the size, step size, and presentation rhythm of subsequent optotypes based on the respondent's reaction time and recognition accuracy during the testing process. This reduces invalid trials while capturing the examinee's true vision threshold, thereby achieving a synergistic improvement in screening efficiency and measurement accuracy.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides a remote self-service vision screening method based on a visual target presentation device, comprising: receiving a screening task configuration sent from the cloud through the visual target presentation device, the screening task configuration including a subject's identity identifier; retrieving the subject's previous screening records based on the subject's identity identifier, and generating a personalized visual target presentation sequence based on the previous screening records; during the detection process, collecting the subject's response time and recognition accuracy in real time, and dynamically adjusting the size and presentation rhythm of subsequent visual targets according to the response time and recognition accuracy; after the detection is completed, uploading the detection results to the cloud, and generating a vision change trend assessment based on the detection results and the previous screening records.

[0007] As a preferred embodiment of the present invention, after receiving the screening task configuration sent from the cloud through the visual target presentation device, the method further includes: parsing the screening task configuration to obtain the examinee's identity identifier, screening time window, and visual target presentation parameter set; determining whether the current time is within a valid screening period based on the screening time window; if it is within a valid screening period, initiating screen brightness adaptive calibration of the visual target presentation device; after the screen brightness adaptive calibration is completed, setting the contrast and background illuminance of the visual target according to the visual target presentation parameter set; and after setting, executing the step of retrieving the examinee's previous screening records. Through time window limitations and environmental adaptive calibration, screening of different examinees is ensured under standardized conditions, eliminating ambient light interference.

[0008] Preferably, retrieving past screening records and generating a personalized visual target presentation sequence based on the examinee's identity identifier specifically includes: sending a historical data request to the cloud based on the examinee's identity identifier, and receiving the examinee's past screening records returned by the cloud; extracting the termination visual target level and corresponding timestamp for each screening from the past screening records, and arranging the termination visual target levels into a historical level sequence according to the order of the timestamps; calculating the level change trend value based on the historical level sequence, and determining the starting visual target level for this screening based on the level change trend value; and generating the personalized visual target presentation sequence based on the starting visual target level, according to a preset level increment rule and visual target type rotation rule. This method dynamically determines the starting point using historical records, avoiding starting with a fixed low-level visual target each time, effectively saving screening time and improving examinee compliance.

[0009] A further preferred embodiment of the process for determining the starting visual target level includes: selecting the termination visual target levels from the three most recent screenings in the historical level sequence, calculating the average of the three termination visual target levels as the level change trend value; comparing the level change trend value with a preset standard visual acuity level; if the level change trend value is higher than the standard visual acuity level, then setting the starting visual target level to a first offset level higher than the standard visual acuity level; if it is lower, then setting it to a second offset level lower than the standard visual acuity level; if it is equal, then setting it to the standard visual acuity level. This ensures that the screening starting point closely follows the actual visual acuity change trend of the examinee, avoiding the repeated presentation of a large number of irrelevant visual targets due to improvement or decline in visual acuity.

[0010] The preferred method for generating personalized visual target presentation sequences according to the present invention includes: using the initial visual target level as the current presentation level, and selecting the current presentation type from multiple visual target types according to a visual target type rotation rule; generating the current visual target according to the current presentation type at the current presentation level and presenting it to the subject; after the recognition result of the current visual target is recorded, increasing the current presentation level by one level according to the level increment rule, and switching to the next visual target type according to the visual target type rotation rule to generate the next visual target; repeating the step of generating the next visual target until the number of generated visual targets reaches the total number of single screenings specified in the screening task configuration. Through level increment and type rotation, the detection follows a logic from easy to difficult, and the visual target morphology is continuously changed, reducing subject guessing and memory effects, and improving accuracy.

[0011] More preferably, the level improvement and type rotation specifically include: after the recognition result of the current optotype is recorded, determining whether the current presentation level has reached the highest presentation level specified in the screening task configuration; if not, increasing the current presentation level by one standard logarithmic visual acuity chart level; selecting a next optotype different from the current optotype from the optotype type set including E-type, digit optotype, and letter optotype according to the optotype type rotation rule; and generating the next optotype using the improved presentation level and the next optotype. The rotation of multiple optotype types makes the screening applicable to subjects of different ages and cultural backgrounds, while the maximum level limit prevents excessive presentation of overly small optotypes from causing visual fatigue.

[0012] As another preferred technical solution of the present invention, the steps of real-time acquisition of response time and recognition accuracy, and dynamic adjustment of subsequent visual target size and presentation rhythm, include: after the current visual target is presented, starting a timer to record the time taken for the subject to make a response instruction, and using the time taken as the response time; determining whether the response time exceeds a preset response time threshold, and if it does, determining that the recognition result corresponding to the current visual target is incorrect; counting the total number of visual targets presented and the number of correctly recognized visual targets in the current detection round, and calculating the recognition accuracy; determining the adjustment direction and adjustment step size of subsequent visual targets based on the relationship between the recognition accuracy and a preset accuracy range, and determining the presentation interval of subsequent visual targets based on the average response time. This method combines reaction time and accuracy, not only assessing visual discrimination ability, but also comprehensively considering the subject's response state, achieving dynamic difficulty adaptation, and making the screening process closer to the real visual acuity level.

[0013] Further preferably, determining the size adjustment direction and step size based on the recognition accuracy and the presentation interval duration based on the response time includes: comparing the recognition accuracy with a first accuracy threshold and a second accuracy threshold, wherein the first accuracy threshold is greater than the second accuracy threshold; if the recognition accuracy is greater than the first accuracy threshold, then determining the size adjustment direction as a shrinking direction and setting the adjustment step size as a first shrinking step size; if the recognition accuracy is less than the second accuracy threshold, then determining the size adjustment direction as an enlarging direction and setting the adjustment step size as a first enlarging step size; if the recognition accuracy is between the second accuracy threshold and the first accuracy threshold, then keeping the current target size unchanged, calculating the average duration of the last three response times, and setting the presentation interval duration as the sum of the average duration and a preset buffer duration. Based on the accuracy interval strategy, frequent size fluctuations caused by accidental correct or incorrect answers can be avoided, and the target appearance rhythm can be adjusted according to the subject's natural reaction speed in a stable state, enhancing testing comfort.

[0014] Furthermore, this invention provides an accelerated adjustment mechanism, comprising: when three consecutive adjustments are all in a shrinking direction and the recognition accuracy after each adjustment is greater than the first accuracy threshold, switching the adjustment step size to a second shrinking step size, wherein the second shrinking step size is greater than the first shrinking step size; when three consecutive adjustments are all in a magnifying direction and the recognition accuracy after each adjustment is less than the second accuracy threshold, switching the adjustment step size to a second magnifying step size, wherein the second magnifying step size is greater than the first magnifying step size; after switching the adjustment step size, if the current recognition accuracy falls back to between the second accuracy threshold and the first accuracy threshold, then restoring the adjustment step size to the first shrinking step size or the first magnifying step size. By triggering large step size adjustments through continuous stable performance, the visual threshold boundary can be quickly approached, while small step size adjustments are restored when the state returns to the ambiguous range, ensuring the accuracy and reliability of the final visual acuity measurement value.

[0015] For post-detection processing, the preferred approach is as follows: After detection, the recognition results for each optotype, the response time, and the final termination optotype level are packaged into the detection results and uploaded to the cloud via wireless communication; a trend analysis request is sent to the cloud, the request including the subject's identity and the timestamp of the detection; a vision change curve, fitted from previous screening records and the current detection results, is received from the cloud, containing the trajectory of the termination optotype level over time for each screening; the level change between two adjacent screenings is calculated based on the vision change curve, and the number of times the level change exceeds a preset threshold is counted as the fluctuation frequency, which is used as the vision change trend assessment. Through centralized cloud management and visualization of vision changes, caregivers or users can intuitively understand the long-term trend and fluctuations of vision, and promptly identify potential vision problems.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: Regarding the innovation mentioned in the background technology—personalized starting optotype levels based on historical data—this method requests the subject's previous screening records from the cloud, extracts the ending optotype level and timestamp of each screening, arranges them into a historical level sequence, and calculates the level change trend value to determine the starting optotype level for the current screening. A personalized optotype presentation sequence is generated based on this starting optotype level, replacing the conventional practice of all subjects starting from a fixed initial level. Using this method, subjects with long-term stable vision or clear recent trends can directly start screening from optotype levels close to their resolution threshold, eliminating numerous invalid trials far below or above the threshold. This significantly shortens the time required for a single screening, reduces visual fatigue and operational burden on the subject, and is particularly suitable for individuals who need to frequently monitor changes in vision. Simultaneously, determining the starting optotype level based on real historical trends avoids placing subjects with significantly deteriorated vision at excessively high initial levels, preventing the frustration of consecutive identification failures in the early stages of testing, and improving compliance with remote self-service screening and the effectiveness of screening results.

[0017] Regarding the innovative aspect of dynamically adjusting the size and presentation rhythm of visual targets based on response time and recognition accuracy, the system records the response time and recognition result of each visual target in real time during the testing process. It then calculates the recognition accuracy in the current testing round and compares it with a preset accuracy threshold range to determine the direction and step size for subsequent visual target size adjustments. The presentation interval for subsequent visual targets is determined based on the average response time. When the subject's recognition accuracy remains consistently high, the system automatically switches to a larger shrinking step size to accelerate approaching the threshold region; when the recognition accuracy remains consistently low, a larger enlarging step size is used to quickly exit the recognition difficulty zone; when the recognition accuracy returns to the middle range, the initial step size is restored, achieving fine-tuned adjustment of size changes. Simultaneously, the presentation interval is dynamically calculated based on the most recent response time, automatically extending the interval for slower responses and appropriately shortening it for faster responses, ensuring that the visual target presentation rhythm matches the subject's current cognitive processing speed. This method enables the screening process to accurately capture real-time fluctuations in the examinee's recognition ability, reducing misjudgments caused by instantaneous changes in attention or differences in reaction speed. While ensuring detection accuracy, it controls the total number of visual targets presented, further improving the efficiency of remote self-service screening and the stability of test results. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1This is a flowchart of a remote self-service vision screening method based on visual target presentation devices; Figure 2 This is a flowchart illustrating the adaptive adjustment of optotype size and presentation interval in vision testing. Figure 3 This is a flowchart of the personalized visual representation sequence generation process; Figure 4 This is a diagram illustrating the trend of visual acuity changes and the initial optotype level for screening; Figure 5 It is an adaptive calibration curve of ambient light intensity and screen brightness. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] See Figure 1 This invention provides a remote self-service vision screening method based on a visual target presentation device. The method is executed by a visual target presentation device equipped with a display screen, input function, and wireless communication capabilities. The device interacts with a cloud server through installed software applications. The method includes: receiving a screening task configuration from the cloud via the visual target presentation device. The screening task configuration is generated by the cloud according to a screening plan and includes at least the examinee's identity identifier, and may also include information such as a screening time window and a set of visual target presentation parameters. Based on the examinee's identity identifier in the screening task configuration, the device retrieves the examinee's previous screening records from the cloud and generates a personalized visual target presentation sequence that matches the examinee's vision change characteristics based on these records. During the testing phase, the device sequentially presents visual targets on the screen, collecting real-time data on the examinee's response time for each instruction, while simultaneously calculating the recognition accuracy. Based on the response time and recognition accuracy, the device dynamically adjusts the size of subsequent visual targets and the presentation rhythm between them. After the test, the device packages and uploads the test results to the cloud. The cloud performs a comprehensive analysis based on the previous screening records, generates a vision change trend assessment, and returns or pushes the assessment results to users and professionals. Example 1

[0022] The visual target presentation device, based on the subject's identity identifier parsed from the screening task configuration sent from the cloud, sends a historical data request to the cloud, carrying the subject's identity identifier in the request. The cloud retrieves stored previous screening records based on the subject's identity identifier and returns these records to the visual target presentation device. The visual target presentation device receives the previous screening records returned from the cloud; each record includes the final termination visual target level measured for that screening and a timestamp indicating the screening was completed.

[0023] The target presentation device sequentially extracts the termination target level and corresponding timestamp for each screening from the previous screening records, and arranges the extracted termination target levels into a historical level sequence according to the order of the timestamps. In the historical level sequence, the order of the termination target levels reflects the time interval between each screening.

[0024] The visual target presentation device calculates the visual target change trend value based on a historical grading sequence. The calculation method is as follows: Select the termination visual target grading levels corresponding to the three most recent screenings from the historical grading sequence, calculate the arithmetic mean of these three termination visual target grading levels, and use this arithmetic mean as the grading trend value. The reason for using the average of the termination visual target grading levels from the three most recent screenings is that these records reflect the continuous state of the subject's recent visual acuity changes. These three data points maintain the smoothness of trend judgment while avoiding the introduction of too much early historical data that could cause a lag in trend estimation. The formula for calculating the grading trend value is expressed as follows: ; in, Indicates the trend value of grade change; This indicates the termination target level of the earliest screening out of the three most recent screenings; This indicates the termination target level for the screening that occurred in the middle of the three most recent screenings; This indicates the termination target level of the most recent screening out of the three most recent screenings. , and The specific values ​​are taken from the terminating optotype levels corresponding to the third-to-last, second-to-last, and last elements in the historical grading sequence, respectively. Each terminating optotype level is a logarithmic visual acuity value on the standard logarithmic visual acuity chart.

[0025] The visual target display device compares the calculated trend value of the level change with the preset standard visual acuity level. The standard visual acuity level is preset to 5.0, which represents normal visual acuity on the standard logarithmic visual acuity chart and corresponds to 1.0 decimal visual acuity. The comparison results are processed in three ways.

[0026] When the trend of visual acuity changes exceeds the standard visual acuity level, the optotype presentation device sets the starting optotype level for this screening as the first offset level. The first offset level is the standard visual acuity level increased by a preset offset, which is set to one standard logarithmic level interval of 0.1, i.e., the first offset level is 5.1. The basis for using one level interval as the offset is that the multiple of optotype size change between adjacent rows in the standard logarithmic visual acuity chart is constant. An offset of one level can safely and effectively adjust the screening starting point to match the subject's recent visual acuity trend of above-normal levels, while avoiding the initial optotype being too small, which could lead to initial identification difficulties or error accumulation.

[0027] If the trend value of the visual acuity level is less than the standard visual acuity level, the optotype presentation device will set the starting optotype level for this screening to the second offset level. The second offset level is the standard visual acuity level reduced by a preset offset amount of 0.1, that is, the second offset level is 4.9.

[0028] When the trend value of the grade change is equal to the standard visual acuity grade, the optotype presentation device will directly set the starting optotype grade for this screening to the standard visual acuity grade of 5.0.

[0029] After determining the initial optotype level, the optotype presentation device generates a personalized optotype presentation sequence based on the initial optotype level, according to preset level increment rules and optotype type rotation rules. The level increment rule is as follows: starting from the initial optotype level, after each optotype is generated, the level of the next optotype increases by one standard logarithmic visual acuity level based on the current optotype level; that is, the level increment is fixed at 0.1. The optotype type rotation rule is as follows: optotype types are selected cyclically in a fixed order from a preset set of optotype types. The optotype type set includes three types: E-shaped optotypes, number optotypes, and letter optotypes. The rotation order is set to E-shaped optotype → number optotype → letter optotype → E-shaped optotype. The specific process for generating a personalized visual target presentation sequence is as follows: The initial visual target level is used as the presentation level of the first visual target. Following a rotation rule, a visual target type is selected from the set of visual target types as the presentation type of the first visual target, generating the first visual target. After the first visual target is generated, the presentation level is incremented by 0.1 to become the presentation level of the second visual target, and the visual target type is switched to the next type according to the rotation order, generating the second visual target. This process of incrementing the level and rotating the type continues, generating subsequent visual targets sequentially, until the total number of generated visual targets equals the total number of screenings specified in the screening task configuration. This yields the personalized visual target presentation sequence required for this screening. Each visual target in the personalized visual target presentation sequence has a defined visual target level and visual target type.

[0030] See Figure 4In the graph, the horizontal axis represents the screening sequence number, ranging from 0 to 500, and the vertical axis represents the logarithmic visual acuity value, with a range of approximately 4.7 to 5.3. The three curves in the graph represent the termination optotype level (thin gray line), the level change trend value (thick blue line), and the starting optotype level for this screening (dashed green line), respectively, with the standard visual acuity level of 5.0 represented by a red dashed line.

[0031] The thin gray line representing the termination optotype grade curve shows considerable fluctuations, reflecting a certain degree of dispersion in the specific visual acuity levels measured by the examinees in each screening. The thick blue line representing the grade change trend value is the moving average of the termination optotype grades across all screenings. The curve shows an overall upward trend, indicating that the examinees' visual acuity levels gradually improve as the screening sequence increases. The trend line value slowly rises from approximately 4.8 to around 5.2, reflecting a stable improvement in recent visual acuity.

[0032] The green dashed line indicates that the starting visual acuity level for this screening is 5.1, which is significantly higher than the standard visual acuity level of 5.0 (red dashed line). This conforms to the provision in Example 1 that when the trend value of the level change is greater than the standard visual acuity level, the starting visual acuity level is set to the first offset level of 5.1. This setting of the starting level effectively matches the recent improvement trend of the examinee's vision, avoiding a decrease in judgment efficiency due to an excessively low starting visual acuity level. Example 2

[0033] During the testing process, refer to Figure 2 The moment the visual target presentation device displays the current visual target on the screen, an internal timer is started. The visual target presentation device continuously listens to the subject's response commands made by touching the screen or issuing voice commands, and records the time elapsed from the completion of the presentation of the current visual target to the receipt of the response command by the visual target presentation device. This time elapsed is recorded as the response time of the current visual target.

[0034] The visual target presentation device compares the response time with a preset reaction time threshold. The preset reaction time threshold is set at 3000 milliseconds. This is based on the following: In a standard vision test environment, normal reaction times typically fall between 500 and 2000 milliseconds. Setting the threshold at 3000 milliseconds covers the vast majority of normal reaction scenarios with sufficient margin. A reaction time exceeding 3000 milliseconds indicates hesitation, distraction, or visual recognition difficulties. Responses in such cases are influenced by non-visual factors and should not be considered valid. When the response time exceeds the 3000-millisecond threshold, the visual target presentation device directly classifies the recognition result corresponding to the current visual target as incorrect, without further determining whether the response matches the actual visual target direction. When the response time does not exceed the 3000-millisecond threshold, the device compares the direction of the visual target opening indicated by the subject's response command with the actual opening direction of the current visual target. If the directions match, the recognition result is considered correct; otherwise, it is considered incorrect.

[0035] During the detection process, the target presentation device continuously maintains the total number of targets presented in the current detection round and the number of targets with correct recognition results. Each time a target recognition result is recorded, the total number of presented targets increases by one; if the recognition result is correct, the number of correctly recognized targets increases by one. The target presentation device calculates the recognition accuracy rate based on the total number of presented targets and the number of correctly recognized targets. The formula for calculating the recognition accuracy rate is expressed as: ; in, Indicates the recognition accuracy rate; This indicates the number of visual targets that were correctly identified in the current detection round. The value is the cumulative count of all visual targets whose identification results are judged to be correct from the start of detection to the calculation time; This indicates the total number of targets that have been presented in the current testing round. The value is the cumulative count of all targets that have been presented from the start of detection to the calculation time, and Always greater than or equal to Recognition accuracy The value range is between 0 and 1.

[0036] The optotype presentation device determines the direction and step size of subsequent optotype size adjustments based on the relationship between the recognition accuracy rate and a preset accuracy range, and determines the presentation interval of subsequent optotypes based on the average response time. The preset accuracy range is defined by a first accuracy threshold and a second accuracy threshold, with the first accuracy threshold set at 80% and the second accuracy threshold set at 60%. The basis for setting the first accuracy threshold of 80% and the second accuracy threshold of 60% is as follows: In the stepwise approach to the threshold process of standard vision tests, when the subject's recognition accuracy rate is consistently higher than 80%, it usually indicates that the current optotype size is too easy for the subject to recognize, and there is still room for further improvement in the optotype level; when the recognition accuracy rate is consistently lower than 60%, it indicates that the optotype size has approached or fallen below the subject's discrimination threshold, and the recognition state tends to be at the level of random guessing. At this time, the optotype size needs to be increased to ensure the effectiveness of the test.

[0037] The target presentation device will calculate the recognition accuracy. Compare with the first accuracy threshold of 80% and the second accuracy threshold of 60%.

[0038] In terms of recognition accuracy In more than 80% of cases, the optotype presentation device determines that the subsequent optotype size adjustment direction is the reduction direction and sets the adjustment step size as the first reduction step size. The first reduction step size is set to reduce the optotype size by 1 standard logarithmic visual acuity class, that is, the logarithmic visual acuity value corresponding to the optotype height or stroke width decreases by 0.1. This ensures a high recognition accuracy. In cases where the accuracy is less than 60%, the visual target presentation device determines the subsequent visual target size adjustment direction as the magnification direction and sets the adjustment step size as the first magnification step size. The first magnification step size is set to magnify the visual target size by one standard logarithmic visual acuity level, that is, the logarithmic visual acuity value corresponding to the visual target height or stroke width increases by 0.1. This ensures a high recognition accuracy. In cases where the percentage is between 60% and 80% (inclusive), the visual target presentation device maintains the current visual target size without adjustment. Simultaneously, it collects the response times for the three most recent visual targets, calculates the average duration of these three responses, and sets the subsequent visual target presentation interval as the sum of the average duration and a preset buffer time. The preset buffer time is set to 500 milliseconds. This is based on the principle that a 500-millisecond rest period after a response allows the subject's gaze to reset and refocus on the next visual target, preventing visual fatigue buildup, while also preventing excessive delays in the screening process.

[0039] During the size reduction adjustment process, the target display device records the adjustment direction and the recognition accuracy after each adjustment. When the target display device detects three consecutive size adjustments in the same direction (reduction), and calculates the recognition accuracy after each of these three adjustments... When all three reduction rates consistently exceed 80%, the visual target presentation device switches the reduction direction adjustment step size from the first reduction step size to the second reduction step size. The second reduction step size is set to reduce the visual target size by two standard logarithmic visual acuity grades, i.e., a decrease in the logarithmic visual acuity value of 0.2. The rationale for using the second reduction step size is that if the recognition accuracy remains above 80% after three consecutive reductions, it indicates that the test subject's visual acuity is significantly better than the currently presented visual target grade. Increasing the reduction range can accelerate the approach to the threshold without reducing test accuracy. During the size enlargement adjustment process, when the visual target presentation device detects that the direction of three consecutive size adjustments is the enlargement direction, and the recognition accuracy is calculated after each of these three adjustments... When all values ​​remain below 60%, the visual target display device switches the magnification step size from the first magnification step size to the second magnification step size. The second magnification step size is set to magnify the visual target size by two standard logarithmic visual acuity levels, i.e., the logarithmic visual acuity value increases by 0.2.

[0040] After the target presentation device switches the adjustment step size to either the second shrinking step size or the second enlarging step size, it continues to present subsequent targets and calculate the recognition accuracy. The target presentation device then recalculates the recognition accuracy. When the accuracy drops to between 60% and 80% (inclusive), the target presentation device will adjust the step size back to the first shrinking or enlarging step size used before the switch. This adjustment is based on the fact that the accuracy drop to the transition range indicates that the current target size is close to the subject's discrimination threshold, requiring a smaller step size for finer approximation to ensure the accuracy of the final measured target level. Example 3

[0041] In practice, after a complete testing process, the visual target presentation device packages all the data collected during the testing process. The package includes the recognition result for each presented visual target, the response time for each presented visual target, and the final determined termination visual target level. The recognition result for each presented visual target is recorded as correct or incorrect, the response time for each presented visual target is recorded as an integer value in milliseconds, and the final determined termination visual target level is recorded as a logarithmic visual acuity value to one decimal place. The visual target presentation device packages the above data into a test result data package, which is organized using a structured data format.

[0042] The visual target presentation device uploads the detection result data packet to the cloud server via its built-in wireless communication module. This module can be either Wi-Fi or cellular, and the communication transmission is protected by an encryption protocol. After the detection result data packet is uploaded, the visual target presentation device sends a trend analysis request to the cloud server. This request includes the subject's identification identifier and the timestamp indicating the completion of the detection. The subject's identification identifier is a unique string, and the timestamp indicating the completion of the detection is the Unix timestamp or ISO8601 format time string corresponding to the last time the visual target presentation device recorded the response command received.

[0043] In some embodiments, after receiving a trend analysis request, the cloud server retrieves all previous screening records corresponding to the examinee's identity from the cloud database based on the examinee's identity carried in the trend analysis request. Each record in the previous screening records includes the termination visual target level and the timestamp of the screening completion for that screening. The cloud server merges and sorts the termination visual target level and its corresponding timestamp from the currently uploaded test results with the termination visual target levels and their corresponding timestamps from all previous screening records, according to the chronological order of the timestamps, to form time series data containing all termination visual target levels from previous screenings and the current test. The cloud server fits the change of the termination visual target level in the time series data with the screening time to generate a visual acuity change curve. The fitting uses linear interpolation, connecting the termination visual target levels of two adjacent time points with line segments to form a continuously changing broken line. The horizontal axis of the visual acuity change curve represents the screening completion time, and the vertical axis represents the logarithmic visual acuity value corresponding to the termination visual target level. The cloud server returns the generated visual acuity change curve data to the visual target presentation device.

[0044] In other embodiments, the visual acuity display device receives visual acuity change curve data returned from the cloud. This data includes several data points sorted by time, each containing a timestamp of a screening and the corresponding termination visual acuity level. The device calculates the level change between two adjacent screenings based on the visual acuity change curve. The level change is calculated as the difference between the termination visual acuity level of the later screening and the termination visual acuity level of the earlier screening. The device takes the absolute value of each calculated level change and compares it with a preset change threshold. The preset change threshold is set to 0.2, based on the fact that a complete level interval on a standard logarithmic visual acuity chart is 0.1. A level change of 0.2 between two adjacent screenings indicates a change in visual acuity of more than two levels, spanning two standard logarithmic level intervals, which falls within a clinically significant range of visual acuity fluctuation and excludes minor fluctuations caused by random errors in a single screening.

[0045] The visual target presentation device counts the number of times the absolute value of the grade change exceeds a preset change threshold of 0.2, and records this count as the fluctuation frequency. The fluctuation frequency is calculated by iterating through all the absolute values ​​of grade changes between two adjacent screenings on the visual acuity change curve, and accumulating the counts for cases where the absolute value of the grade change is greater than 0.2. The visual target presentation device outputs the fluctuation frequency value as the result of the visual acuity change trend assessment. This assessment represents the stability of the examinee's visual state throughout the screening process; a higher fluctuation frequency indicates more unstable visual acuity changes, while a lower fluctuation frequency indicates a more stable visual state. Example 4

[0046] In specific implementation, please refer to Figure 3 The optotype presentation device sets the determined starting optotype level as the current presentation level. The current presentation level is represented by a numerical value with one decimal place, corresponding to a specific level on the standard logarithmic visual acuity chart, such as 5.1, 5.0, or 4.9. The optotype presentation device selects an optotype from a preset set of optotypes as the current presentation level according to an optotype type rotation rule. The preset set of optotypes includes three types: E-shaped optotypes, number optotypes, and letter optotypes. The optotype type rotation rule is set to cyclically select E-shaped optotypes, number optotypes, and letter optotypes in a fixed order: the first time, an E-shaped optotype is selected; the second time, a number optotype is selected; the third time, a letter optotype is selected; the fourth time, an E-shaped optotype is selected, and so on. In the optotype type rotation rule, a remainder of 1 when the sequence number is divided by 3 corresponds to an E-shaped optotype, a remainder of 2 corresponds to a number optotype, and a remainder of 0 corresponds to a letter optotype.

[0047] The optotype presentation device generates the current optotype according to the current presentation level and presentation type. The process of generating the current optotype is as follows: The optotype presentation device calculates the standard size of the current optotype based on the logarithmic visual acuity value corresponding to the current presentation level. The calculation of the standard size follows the design specifications of the standard logarithmic visual acuity chart, meaning there is a defined conversion relationship between the optotype height or key stroke width and the logarithmic visual acuity value. The optotype presentation device determines the randomization method of the optotype's shape and orientation according to the current presentation type. For the E-shaped optotype, the device randomly selects one of the four opening directions to generate the E-shaped graphic; for the digit optotype, the device randomly selects a digit from a preset digit set to generate the digit graphic; for the letter optotype, the device randomly selects a letter from a preset letter set to generate the letter graphic. The optotype presentation device displays the generated current optotype on the screen and presents it to the examinee.

[0048] After the optotype presentation device records the subject's recognition result of the current optotype, it increases the current presentation level by one level according to the level increment rule. Before increasing, the optotype presentation device determines whether the current presentation level has reached the highest presentation level specified in the screening task configuration. The highest presentation level specified in the screening task configuration is 5.3, which is set based on the fact that 5.3 on the standard logarithmic visual acuity chart corresponds to decimal visual acuity of 2.0, which is the recognized upper limit of superior visual acuity in routine vision screening. Setting the highest presentation level to 5.3 can cover the range of normal to excellent visual acuity while avoiding the presentation of optotypes that are too small, thus making recognition meaningless. If the current presentation level has not yet reached the highest presentation level of 5.3, the optotype presentation device increases the current presentation level by one standard logarithmic visual acuity chart level, that is, the logarithmic visual acuity value increases by 0.1. If the current presentation level has already reached the highest presentation level of 5.3, the optotype presentation device will not increase the current presentation level, and the presentation level of subsequent optotypes will remain unchanged at 5.3.

[0049] While the target presentation device upgrades the current presentation level by one level according to the level increment rule, it also switches the current presentation type to the next target type according to the target type rotation rule. The switching method for the current presentation type is as follows: determine the position of the current target in the target type rotation sequence, and select the next target type in the sequence as the new current presentation type. Specifically, if the current presentation type is an "E" target, it switches to a number target; if the current presentation type is a number target, it switches to a letter target; and if the current presentation type is a letter target, it switches to an "E" target.

[0050] The target rendering device generates the next target using the improved rendering level and the switched target type. The size calculation and graphics generation methods for generating the next target are the same as those for generating the current target. The rendering level of the next target is the improved rendering level, and the rendering type of the next target is the switched target type.

[0051] The optotype presentation device repeatedly performs the operation of generating the next optotype. In each repetition, the optotype generated in the previous round is presented as the new current optotype, and the recognition result is recorded. The presentation level is incremented and adjusted, and the optotype type is rotated before generating the next optotype. The optotype presentation device maintains an optotype generation counter, which increments by one for each optotype generated. The device compares the current count of the optotype generation counter with the total number of screenings specified in the screening task configuration. The total number of screenings per session is set to 20, based on the following criteria: presenting and recognizing 20 optotypes achieves a balance between examination time and testing accuracy; 20 optotypes cover approximately 2.0 logarithmic visual acuity levels, matching the span between the highest presentation level of 5.3 and the most likely starting level of 5.0, ensuring a stable termination decision for the screening. When the optotype generation counter reaches the total number of screenings per session of 20, the optotype presentation device stops generating new optotypes, and the personalized optotype presentation sequence is complete. The personalized target presentation sequence consists of 20 targets, each with a defined target level and target type. The target level increases or remains constant in the order of generation, and the target types are arranged in a cyclical order of E-shaped targets, number targets, and letter targets. Example 5

[0052] In practice, the target presentation device receives the screening task configuration from the cloud via a wireless communication module. The screening task configuration is encapsulated in a structured data format, and the target presentation device parses it. The parsing process includes reading various fields from the screening task configuration and extracting the subject identification field, the screening time window field, and the target presentation parameter set field. The subject identification field contains a unique string of letters and numbers, used to uniquely identify the subject in this screening. The screening time window field includes a screening start timestamp and a screening end timestamp, both expressed in Coordinated Universal Time (UTC) millisecond-level Unix timestamp format. The screening start timestamp defines the earliest time the screening process can be started, and the screening end timestamp defines the latest time the screening process can be started. The target presentation parameter set field includes target contrast parameters and background illuminance parameters. The target contrast parameter represents the brightness contrast between the target graphic and the background as a percentage value, and the background illuminance parameter represents the target illuminance that the target presentation device's screen background should achieve as a lux value.

[0053] The target display device reads its own system clock to obtain the current Coordinated Universal Time (UTC) Unix timestamp in milliseconds. It then compares this current timestamp with the screening start and end timestamps. If the current timestamp is greater than or equal to the screening start timestamp and less than or equal to the screening end timestamp, the device is considered to be within a valid screening period. If the current timestamp is less than the screening start timestamp or greater than the screening end timestamp, the device is considered not to be within a valid screening period. The target display device then displays a prompt on the screen, including a description of the screening time window, and aborts the screening process, refraining from further steps.

[0054] If the current moment is determined to be within a valid screening period, the target display device initiates adaptive screen brightness calibration. The adaptive calibration process is as follows: the target display device uses its built-in ambient light sensor to collect the current ambient light intensity, measured in lux. Based on the collected ambient light intensity, the target display device calculates the target screen brightness according to a preset mapping relationship between ambient light intensity and target screen brightness. This mapping relationship is expressed using a linear formula, as follows: ; in, This indicates the target screen brightness, expressed in candela per square meter. This indicates the current ambient light intensity as collected by the ambient light sensor, in lux. Indicates the brightness scaling factor. The value ranges from 0.8 to 1.2, and the unit is candela per square meter per lux. The specific value is set to 1.0, based on the following: within the typical indoor ambient illuminance range of 200 to 500 lux, Setting it to 1.0 can make the target screen brightness correspond to the ambient light level in a close 1:1 ratio, and match the screen brightness with the brightness that the human eye adapts to, avoiding glare caused by excessive screen brightness or difficulty in visual target recognition due to insufficient screen brightness; This indicates the brightness offset, measured in candela per square meter. The value ranges from 0 to 50 candela per square meter. The specific value is set to 0. The basis for this setting is that, under standard vision screening conditions without special requirements for ambient light compensation, a direct linear correspondence between the target screen brightness and ambient light illuminance is sufficient to meet the display brightness requirements of the standard logarithmic visual acuity chart, without the need for additional brightness offset.

[0055] The target display device calculates the brightness of the target screen. Next, the output power of the screen backlight drive circuit is adjusted so that the actual brightness of the screen reaches the target screen brightness. The target display device continuously reads the ambient light sensor values ​​and compares them with the target screen brightness to form a closed-loop adjustment. The screen brightness adaptive calibration is considered complete when the deviation between the actual screen brightness and the target screen brightness is less than a preset deviation threshold of 5 candela per square meter.

[0056] After the screen brightness adaptive calibration is completed, the visual target presentation device reads the visual target contrast parameter and background illuminance parameter from the visual target presentation parameter set. The device sets the visual target contrast parameter to the value specified in the parameter set, which is 100%. This setting is based on the principle that 100% contrast represents the maximum brightness difference between visual target pixels and background pixels, meeting the requirements of the standard logarithmic visual acuity chart for visual target-background contrast and eliminating recognition interference caused by low contrast. The device also sets the background illuminance parameter to the value specified in the parameter set, which is 200 lux. This setting is based on the principle that 200 lux provides a uniform and stable lighting environment for the screen display area, consistent with the center value of the recommended room illuminance range of 150 to 300 lux in standard visual acuity testing, avoiding excessively dark backgrounds that increase eye strain or excessively bright backgrounds that cause screen glare.

[0057] After the visual target contrast parameters and background illuminance parameters are set, the visual target presentation device obtains the parsed subject identity identifier, uses the subject identity identifier as an input parameter, and enters the step of retrieving the subject's previous screening records based on the subject identity identifier, sends a historical data request to the cloud, and waits for the cloud to return the previous screening records.

[0058] See Figure 5 In the graph, the horizontal axis represents ambient illuminance in lux, ranging from 0 to 1000; the vertical axis represents screen brightness in candela per square meter, also ranging from 0 to 1000. The blue solid line represents the target screen brightness, which is the ideal screen brightness calculated by the target display device based on the ambient illuminance using a linear mapping relationship; the orange dots represent the actual screen brightness, which is the screen brightness value actually achieved by the target display device through adjusting the backlight drive circuit.

[0059] The curve shows a highly linear positive correlation between the target screen brightness and ambient light intensity, which conforms to the linear mapping formula described in Example 5. The proportionality coefficient Take 1.0, offset The actual screen brightness data points are uniformly distributed around the target screen brightness curve, indicating that the ambient light sensor data acquisition and backlight adjustment closed-loop control process have achieved high-precision adaptive calibration of screen brightness. The deviation between the actual screen brightness and the target brightness is less than the preset threshold requirement of 5 candela per square meter.

[0060] The ambient light intensity gradually increases from 0 to 1000 lux, and the target screen brightness and the actual screen brightness increase synchronously to ensure that the screen brightness can dynamically match the ambient light intensity under different ambient light conditions, avoiding visual difficulties or eye fatigue caused by brightness mismatch.

[0061] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A remote self-service vision screening method based on a visual target presentation device, characterized in that, include: The screening task configuration, which includes the examinee's identity identifier, is received from the cloud through the visual target presentation device. Based on the examinee's identity, retrieve the examinee's previous screening records, and generate a personalized visual target presentation sequence based on the previous screening records; During the detection process, the response time and recognition accuracy of the subject are collected in real time, and the size and presentation rhythm of subsequent visual targets are dynamically adjusted based on the response time and recognition accuracy. After the test is completed, the test results are uploaded to the cloud, and a vision change trend assessment is generated based on the test results and the previous screening records.

2. The remote self-service vision screening method based on a visual target presentation device according to claim 1, characterized in that, The step of retrieving the subject's previous screening records based on the subject's identity identifier and generating a personalized visual target presentation sequence based on the previous screening records includes: Based on the examinee's identity identifier, a historical data request is sent to the cloud, and the examinee's screening records are returned by the cloud. Extract the termination target level and corresponding timestamp for each screening from the historical screening records, and arrange the termination target levels into a historical level sequence according to the order of the timestamps; The grade change trend value is calculated based on the historical grade sequence, and the starting target grade for this screening is determined based on the grade change trend value. Based on the initial target level, the personalized target presentation sequence is generated according to the preset level increment rules and target type rotation rules.

3. The remote self-service vision screening method based on a visual target presentation device according to claim 2, characterized in that, The step of calculating the level change trend value based on the historical level sequence and determining the starting target level for this screening based on the level change trend value includes: Select the termination target level of the three most recent screenings from the historical level sequence, and calculate the average of the termination target levels of the three most recent screenings as the level change trend value; The level change trend value is compared with the preset standard visual acuity level. If the level change trend value is higher than the standard visual acuity level, the starting optotype level is set to a first offset level that is higher than the standard visual acuity level. If the trend value of the grade change is lower than the standard visual acuity grade, then the starting optotype grade is set to a second offset grade that is lower than the standard visual acuity grade; If the trend value of the grade change is equal to the standard visual acuity grade, then the starting optotype grade is set to the standard visual acuity grade.

4. The remote self-service vision screening method based on a visual target presentation device according to claim 1, characterized in that, During the detection process, the subject's response time and recognition accuracy are collected in real time, and the size and presentation rhythm of subsequent visual targets are dynamically adjusted based on the response time and recognition accuracy, including: After the current visual target is presented, a timer is started to record the time taken for the subject to make a response instruction, and the time taken is used as the response time. Determine whether the response time exceeds a preset response time threshold. If it exceeds the response time threshold, determine that the recognition result corresponding to the current target is incorrect. The total number of visual targets presented in the current detection round and the number of correctly identified visual targets are counted, and the recognition accuracy rate is calculated. Based on the relationship between the recognition accuracy and the preset accuracy range, the adjustment direction and adjustment step size of the subsequent visual targets are determined, and the presentation interval of the subsequent visual targets is determined based on the average value of the response time.

5. The remote self-service vision screening method based on a visual target presentation device according to claim 4, characterized in that, The step of determining the adjustment direction and step size of subsequent visual targets based on the relationship between the recognition accuracy and the preset accuracy range, and determining the presentation interval of subsequent visual targets based on the average response time, includes: The recognition accuracy is compared with a first accuracy threshold and a second accuracy threshold, wherein the first accuracy threshold is greater than the second accuracy threshold; If the recognition accuracy is greater than the first accuracy threshold, then the size adjustment direction is determined to be the shrinking direction, and the adjustment step size is set as the first shrinking step size; If the recognition accuracy is less than the second accuracy threshold, then the size adjustment direction is determined to be the magnification direction, and the adjustment step size is set to the first magnification step size; If the recognition accuracy is between the second accuracy threshold and the first accuracy threshold, the current target size remains unchanged, and the average duration of the three most recent responses is calculated. The presentation interval duration is then set to the sum of the average duration and the preset buffer duration.

6. The remote self-service vision screening method based on a visual target presentation device according to claim 5, characterized in that, The step of determining the adjustment direction and step size of subsequent visual targets based on the relationship between the recognition accuracy rate and the preset accuracy rate range, and determining the presentation interval of subsequent visual targets based on the average response time, further includes: When the direction of the adjustment is the shrinking direction for three consecutive times and the recognition accuracy after each adjustment is greater than the first accuracy threshold, the adjustment step size is switched to the second shrinking step size, which is greater than the first shrinking step size. When the direction of the adjustment is the same as the magnification direction for three consecutive times and the recognition accuracy after each adjustment is less than the second accuracy threshold, the adjustment step size is switched to the second magnification step size, which is larger than the first magnification step size. After switching the adjustment step size, if the current recognition accuracy falls back to between the second accuracy threshold and the first accuracy threshold, the adjustment step size is restored to the first shrinking step size or the first enlarging step size.

7. The remote self-service vision screening method based on a visual target presentation device according to claim 1, characterized in that, After the test is completed, the test results are uploaded to the cloud, and a vision change trend assessment is generated based on the test results and the previous screening records, including: After the detection is completed, the recognition result corresponding to each visual target in this detection, the response time, and the final termination visual target level are packaged into the detection result and uploaded to the cloud via wireless communication; A trend analysis request is sent to the cloud, the trend analysis request including the subject's identity identifier and the timestamp of this test; Receive the visual acuity change curve returned from the cloud, which is generated by fitting the previous screening records and the current test results. The visual acuity change curve includes the trajectory of the change of the termination optotype level of each screening over the screening time. The change in level between two adjacent screenings is calculated based on the vision change curve, and the number of times the change in level exceeds a preset change threshold is counted as the fluctuation frequency. The fluctuation frequency is used as the assessment of the vision change trend.

8. The remote self-service vision screening method based on a visual target presentation device according to claim 2, characterized in that, The process of generating the personalized visual object presentation sequence based on the initial visual object level, according to preset level increment rules and visual object type rotation rules, includes: The initial target level is used as the current presentation level, and the current presentation type is selected from multiple target types according to the target type rotation rules; The current visual target is generated according to the current presentation type under the current presentation level and presented to the examinee. After the recognition result of the current target is recorded, the current presentation level is increased by one level according to the level increment rule, and the target type is switched to the next target type according to the target type rotation rule to generate the next target; Repeat the step of generating the next target until the number of targets generated reaches the total number of single screenings specified in the screening task configuration.

9. The remote self-service vision screening method based on a visual target presentation device according to claim 8, characterized in that, After the recognition result of the current visual target is recorded, the current presentation level is increased by one level according to the level increment rule, and the visual target type is switched to the next visual target type according to the visual target type rotation rule to generate the next visual target, including: After the current target recognition result is recorded, it is determined whether the current presentation level has reached the highest presentation level specified in the screening task configuration; If the current presentation level does not reach the highest presentation level, then the current presentation level is increased by one standard logarithmic visual acuity chart level; According to the aforementioned target type rotation rule, select the next target type that is different from the current target type from the target type set, which includes E-type targets, number targets, and letter targets; The next target is generated based on the improved rendering level and the next target type.

10. The remote self-service vision screening method based on a visual target presentation device according to claim 1, characterized in that, After receiving the screening task configuration from the cloud via the visual target presentation device, the process also includes: The screening task configuration is parsed to obtain the subject's identity identifier, screening time window, and visual target presentation parameter set; Determine whether the current time is within the valid screening period based on the screening time window. If the current time is within the valid screening period, then initiate the screen brightness adaptive calibration of the visual target presentation device. After the screen brightness adaptive calibration is completed, the contrast and background illuminance of the target are set according to the target presentation parameter set; After the setup is complete, execute the step of retrieving the examinee's previous screening records based on the examinee's identity identifier.

Images