Machine learning system and method for intraocular lens selection

Abstract

Particular embodiments disclosed herein provide a device and corresponding method for directing IOL selection. A machine learning model is trained to generate output parameters based on patient attribute data including dimensions of the patient's eye and characteristics of the IOL. The output parameter may be an emmetropia IOL power (EIOL) corresponding to the IOL, which will reduce post-operative ametropia to zero if the implanted IOL is replaced with the emmetropia IOL power. The machine learning model is trained using data from past IOL implant procedures, including patient attribute data (eye size and degree of implanted IOL), and output parameters derived by summing the degree of implanted IOL with post-operative ametropia scaled by a scaling factor.

Description

Technical Field

[0001] The present disclosure generally relates to methods for selecting the refractive power of an intraocular lens (IOL) in order to minimize postoperative refractive error following cataract surgery. Background Art

[0002] Light received by the human eye passes through the transparent cornea that covers the iris and pupil of the eye. The light travels through the pupil and is focused by the lens located behind the pupil in a structure called the capsular bag. The light is focused by the lens onto the retina, which consists of rods and cones, which generate nerve impulses in response to light.

[0003] Due to age or disease, the lens may become cloudy, a condition known as cataracts. Cataracts and other conditions are easily treated by removing the lens and inserting an artificial lens, i.e., an intraocular lens (IOL). An IOL can be manufactured to additionally correct aberrations in the patient's eye, such as myopia, hyperopia, spherical aberration, cylindrical aberration, and astigmatism. Although the refractive error of the patient's eye prior to surgery can be easily determined, it is extremely difficult to estimate the combined refractive error of the patient's eye and the IOL after implantation and healing of the patient's eye.

[0004] Currently, complex IOL power calculation formulas and models are used to select an IOL for a patient's eye. These formulas and models take as input preoperative and / or intraoperative data associated with the patient's eye and output an estimated postoperative refractive error of the patient's eye for a given IOL power. However, existing IOL power calculation formulas and models, while widely used and providing desirable accuracy in many cases, are often not sufficiently accurate.

[0005] It would be a major advance in the art if the selection of IOL refractive properties could be facilitated to minimize post-operative refractive error in a patient's eye following implantation of an IOL. Summary of the invention

[0006] The present disclosure generally relates to a system for selecting an IOL to reduce postoperative refractive error.

[0007] A machine learning model is trained using a plurality of training data entries. Each training data entry may include: (a) past preoperative data of a patient's eye, the past preoperative data including biometric data of the patient's eye and characteristics of an IOL implanted in the patient's eye, and (b) actual output parameters derived from both the characteristics of the IOL and the postoperative refractive error of the patient's eye after implantation of the IOL. The machine learning model is trained to generate predicted output parameters corresponding to a future IOL implantation surgery for the input preoperative data.

[0008] The following description and the associated drawings set forth in detail certain illustrative features of the one or more embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Certain aspects of one or more embodiments are depicted in the drawings, and therefore should not be considered to limit the scope of the disclosure.

[0010] Figure 1 The dimensions of a patient's eye are shown in accordance with certain embodiments.

[0011] Figure 2 Training data entries for training a machine learning model to guide IOL selection according to certain embodiments are shown.

[0012] Figure 3 is a process flow diagram of an example method for processing postoperative data to obtain training data suitable for training a machine learning model, according to certain embodiments.

[0013] Figure 4 is a process flow diagram of a method for guiding IOL selection using a machine learning model according to certain embodiments.

[0014] Figure 5 is a schematic block diagram of an interface for receiving preoperative data and outputting guidance for selecting an IOL according to some embodiments.

[0015] Figure 6 is a process flow diagram of a method for selecting a machine learning model from a plurality of machine learning models to guide IOL selection according to certain embodiments.

[0016] Fig. 7A is a schematic block diagram of a collection of machine learning models for guiding IOL selection according to certain embodiments.

[0017] Figure 7B is a schematic block diagram of machine learning models combined to implement gradient boosting to guide IOL selection in accordance with certain embodiments.

[0018] Figure 8 An example computing device is shown that at least partially implements one or more functions of training and utilizing a machine learning model to guide IOL selection in accordance with certain embodiments.

[0019] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures. It is contemplated that elements and features of one embodiment may be beneficially incorporated in other embodiments without further recitation. DETAILED DESCRIPTION

[0020] As described above, a variety of IOL power calculation formulas and models are currently widely used in the industry. However, existing IOL power calculation formulas and models are not accurate enough when estimating or predicting postoperative refractive error for a given IOL power. For example, various machine learning models have been developed to take certain patient data as input and predict or recommend an IOL power that will minimize the patient's postoperative refractive error. However, many of these machine learning models are trained to predict postoperative refractive error, which means that the postoperative refractive error itself is the output parameter. However, there are technical problems in training machine learning models with postoperative refractive error as an output parameter. Specifically, the technical problems associated with such machine learning models are introduced by the nature of postoperative refractive error, which, as a data point, typically has highly variable values. In this way, training a machine learning model to directly predict postoperative refractive error may lead to overfitting, whereby the machine learning model simply reproduces the random variation of the postoperative refractive error of the training data entries.

[0021] Therefore, the embodiments of the present invention provide a technical solution to the above-mentioned technical problems by providing a system and method for training one or more machine learning models to guide IOL selection, wherein the one or more machine learning models are trained using output parameters derived from postoperative refractive error. For example, a training data set can be used, which includes training data entries, each training data entry including various patient data and output parameters derived from measured postoperative refractive error. Specifically, for each training data entry, the output parameter can be obtained from a combination of one or more parameters describing the implanted IOL and the postoperative refractive error. For example, for each training data entry, the output parameter can be calculated by combining the postoperative refractive error with the IOL power of the IOL implanted in the patient's eye. In this way, the resulting trained machine learning model will provide smoother and more reliable predictions for IOL power changes. In one example, the output parameter can be the emmetropia IOL (EOIL) power, which is calculated as described in more detail below.

[0022] Figure 1 An eye 100 is shown including an outer layer, shown as sclera 102. Cornea 104 is a curved, transparent layer on the front side of the eye. Cornea 104 cooperates with lens 106 to focus light onto retina 108, which includes light-detecting nerve cells. Lens 106 is contained within capsular bag 110. Iris 112 is located between cornea 104 and lens 106.

[0023] The refractive error of the eye 100 is based on the various dimensions of the various optical components of the eye 100 and can be used according to the methods described herein to guide the selection of an IOL that minimizes the postoperative refractive error. These dimensions may include some or all of the following: the average power K of the cornea 104, the white-to-white distance (WTW), the anterior chamber depth (ACD), the lens thickness (LT), the axial length (AL), and / or other dimensions. WTW can be defined as the diameter of the opening in the area occupied by the cornea 104 and the iris 112 in the sclera 102. ACD can be defined as the distance between the anterior pole (outermost point) of the cornea 104 and the lens 106. LT can be defined as the thickness of the lens 106 along the optical axis of the eye 100. AL can be defined as the distance between the anterior pole of the cornea 104 and the retina 108 (specifically, the Bruch's membrane of the retina 108).

[0024] The dimensions of the eye 100 may be measured using one or more preoperative and / or intraoperative imaging devices, such as an optical coherence tomography (OCT) device, a rotational camera (e.g., a Scheimpflug camera), a magnetic resonance imaging (MRI) device, a keratometer, an ophthalmometer, an optical biometer, an intraoperative aberrometer, and / or any other imaging device or method. The dimensions of the eye 100 may also be characterized by evaluating an image of the eye. For example, the image 100 may be an OCT image, which may be processed with a machine learning model to derive an array of values ​​representing the eye 100. The machine learning model may be implemented as a convolutional neural network (CNN), a deep neural network (DNN), or other type of machine learning model.

[0025] The machine learning model can be implemented as an autoencoder that is, for example, trained to generate an array of values. In some embodiments, the array of values ​​is the output of an intermediate layer (i.e., not the first or last layer) of a machine learning model that is trained to perform tasks such as estimating refractive error from an eye image. Additionally or alternatively, the OCT image or the anatomical structure label of the eye 100 in the OCT image can be processed with the aid of a transformation (such as a Fourier transform, a Karhunen-Loeve (KL) transform) to obtain an array of values. Additionally or alternatively, the OCT image or the anatomical structure label of the eye 100 in the OCT image can be processed to obtain the Gaussian degree and / or the principal curvature of the anatomical structure of the eye 100 in order to obtain an array of values. Additionally or alternatively, a task cognitive network (CogNet) can be used to reduce the OCT image or the anatomical structure label of the eye 100 in the OCT image to obtain an array of values.

[0026] As described above, certain embodiments herein provide one or more machine learning models trained using output parameters derived from postoperative refractive error. To train the one or more machine learning models, one or more training data sets including patient data associated with many different patients (e.g., hundreds, thousands, or more patients) may be used. Each of the one or more training data sets may include many training data records or entries, each training data record or entry being associated with a different patient.

[0027] Figure 2 An example training data entry 200 is shown, which includes patient data that can be used to train a machine learning model to provide IOL selection guidance. An example training data entry 200 can be provided for each eye that undergoes surgery. As shown, the training data entry 200 includes patient attributes 204 (which includes patient demographic data, preoperative biometric data, intraoperative biometric data, and the power of the implanted IOL), postoperative refractive error 202, and derived output parameters 206.

[0028] Preoperative biometric data may include the Figure 1 Any of the dimensions described above, the array of values ​​derived from the image of the eye 100 as described above, and / or any diagnostic / imaging data (or information derived therefrom) provided by the preoperative diagnostic / imaging system and device. Intraoperative biometric data may include the above-mentioned Figure 1 The patient attributes 204 may also include any of the dimensions described, an array of values ​​derived from an image of the eye 100 as described above, and / or any diagnostic / imaging data (or information derived therefrom) provided by the intraoperative imaging system and device. The patient attributes 204 may also include the power and / or type of IOL actually implanted during surgery. The IOL power may be selected before the surgery begins, or may be selected based on one or more measurements of refractive error obtained during surgery (such as after the lens is removed). The IOL power may be measured in diopters. The IOL power may be enhanced with other properties of the IOL, such as the properties of the IOL to correct astigmatism or spherical aberration.

[0029] The patient attributes 204 may further include demographic data. The patient's outcome may have some degree of correlation with the patient's age, gender, ethnicity, race, and other demographic factors. Therefore, some or all of these demographic data items may be included in the patient attributes 204. Other values ​​such as the surgeon's identifier, the clinic where the surgery was performed, the location (country, state, city, etc.) where the surgery was performed, etc. may also be used as patient attributes 204 that may be related to the postoperative refractive error 202.

[0030] The post-operative refractive error 202 may include the spherical equivalent (SE) of the patient's eye after implantation of an IOL of the power specified in the training data entry 200. Other measures of the post-operative refractive error 202 may also be included in the post-operative refractive error 202, such as astigmatism and spherical aberration. Each training data entry 200 may further include a derived output parameter 206, which serves as the desired output of the training data entry, and the machine learning model is trained to generate the desired output. In some embodiments, the output parameter is a function of the post-operative refractive error 202 and one or more data points (such as IOL power) provided as part of the patient attributes 204. For example, the derived output parameter 206 may be an emmetropic equivalent IOL power (EIOL) as defined below.

[0031] Figure 3 An example method 300 is shown for training a machine learning model to guide IOL selection based on derived output parameters (e.g., EIOL). The machine learning model can be trained using a training data set including training data entries (such as training data entries 200). Note that the training data set can be updated regularly. For example, information about new patients can be collected regularly and used to update the training data set. Note that method 300 can be performed separately for each type of IOL or class of types of IOLs, thereby generating a trained machine learning model for each type of IOL (e.g., monofocal, a specific set of monofocal IOL models, multifocal, astigmatism, etc.).

[0032] The method 300 may include, at step 302, receiving patient attributes 204 and postoperative refractive error 202 for each training data entry 200. In the following discussion, the training of a machine learning model will be described, but it should be understood that the machine learning model may be composed of multiple machine learning models that individually or collectively perform the same training data as, for example, Figure 4 , Figure 6 , Fig. 7A and Figure 7B And the tasks described in the corresponding description.

[0033] The method 300 may include, for each set of patient attributes 204 and post-operative refractive error 202 in the corresponding data entry 200, deriving an output parameter 206 from the post-operative refractive error 202 and one or more patient attributes 204, such as the IOL power of the implanted IOL, in step 304. For example, the derived output parameter 206 may be an EIOL, which is an approximation of the IOL power that, if used instead of the IOL power of the implanted IOL, would correct the post-operative refractive error 202 of the corresponding patient.

[0034] In one example, let D IOLis the IOL power (in diopters), and let R be the postoperative refractive error in spherical equivalent form 202. The EIOL can then be calculated as D IOL +p*R, where p is the scaling factor. The value of p can be a fixed predetermined amount. For example, p can be a value between 0.64 and 0.72, between 0.66 and 0.7, or between 0.675 and 0.685. Experiments conducted by the inventors found that a p of 0.68 is suitable for most applications. The value of p is an adjustable parameter and can be adjusted to improve the results. The value of p can be D IOL The result of a more complex function of one or both of and R and other patient attributes 204. Once the EIOL is calculated, it is added to the corresponding patient data record.

[0035] Postoperative values ​​of astigmatism or spherical aberration alone (rather than representing the spherical equivalent of these two values) or any other type of refractive error can be processed in a similar manner: the output parameter can be derived as a combination (e.g., summation) of the characteristics of the IOL that compensates for this type of refractive error and the postoperative value of this type of refractive error multiplied by a scaling factor. The scaling factor can be selected using a known relationship between this type of refractive error and the characteristics of the IOL that compensates for this type of refractive error, or can be determined experimentally. In the following description, the spherical equivalent will be discussed, but it should be understood that any other type of refractive error can be substituted and the characteristics of the IOL that compensates for this type of refractive error can be selected using the embodiments disclosed herein.

[0036] Each training data entry 200 may include as input a set of patient attributes 204 and as a desired output an EIOL derived from the patient attributes 204 and the corresponding post-operative refractive error 202. Thousands of training data entries may be used to train the machine learning model. Then, in step 306, the machine learning model may be trained by processing the patient attributes 204 using the machine learning model for each training data entry to obtain an estimated EIOL. The estimated EIOL may be compared to the EIOL of the corresponding training data entry, and one or more weights of the machine learning model may be modified by the training algorithm based on the difference between the estimated EIOL and the actual EIOL of the training data entry. Once the machine learning model has been trained so that the difference between the estimated EIOL and the EIOL of the training data entry is minimized or close to zero, the machine learning model may be deployed and used, such as with respect to Figure 4 as described.

[0037] Figure 4An example method 400 is shown for utilizing a machine learning model trained according to method 300 to assist in selecting an IOL power to be used in a future surgery on a patient's eye. Method 400 may include, at step 402, receiving patient attributes 204 of the patient's eye, which do not include an IOL power. Method 400 may include, at step 404, selecting an initial IOL power. The initial IOL power may be selected based on preoperative and / or intraoperative biometric data using any method known in the art (including any formula-based selection criteria, such as Barrett's formula, refractive convergence formula, etc.).

[0038] Then, method 400 may include, in step 406, generating an IOL set. The IOL set may include a range of IOL powers, including IOL powers that are less than or greater than the initial IOL power. In the case of sufficient computing and storage capacity, step 404 may be omitted, and the IOL set may simply be the entire set of available IOL powers. The IOL powers of the IOL set may be limited to commercially available IOL powers, which are typically in increments of 0.5 diopters.

[0039] Method 400 may include, at step 408, predicting postoperative refractive error for each IOL power in the IOL set based on the EIOL predicted by the machine learning model. At step 408, for each IOL power in the IOL set, the input to the machine learning model may include the patient attributes 204 from step 402 and each IOL power in the set of IOL powers. For example, the predicted postoperative refractive error may be calculated as (EIOL-D IOL ) / p, where D IOL is the IOL power from the IOL, and EIOL is the value predicted by the machine learning model. The predicted postoperative refractive error can then be displayed along with the corresponding IOL power, such as on a display of a computing device, in step 410. In the case where the IOL group is very large, step 410 can include displaying only the N IOL powers in the IOL group that have the lowest N postoperative refractive errors, where N is an integer, such as a value between three and twenty.

[0040] In this way, the surgeon or other professional can easily see which available IOL power will produce the lowest expected postoperative refractive error. Experiments performed by the inventors have shown that simply using postoperative refractive error as the desired output when training a machine learning model is ineffective due to the presence of a large amount of noise in the measured postoperative refractive error. Therefore, by using derived output parameters, such as EIOL derived from the characteristics of the IOL and the postoperative refractive error, improved results can be obtained.

[0041] Figure 5 Demonstrated use of Figure 3 and Figure 4 Example interface 500 of a machine learning model described. Interface 500 may include some or all of the following: a field 502 for entering a patient identifier; a field 504 for entering which eye (right or left) received surgery; a field 506 for entering the type of IOL; a field 508 for selecting a machine learning model from a plurality of available machine learning models; one or more fields 510 for entering biometric data of the patient's eye (and possibly selecting which size or other biometric data to use); and a field 512 for entering a target postoperative refractive error. Fields may be provided to enter other patient attributes, such as age, sex, ethnicity, eye identifier (right or left), surgeon name, clinic name, and / or location. Additionally, other fields may be provided for entering intraoperative data parameters, including intraoperative measurements of aphakic eyes. The fields of interface 500 may be populated manually or automatically with data obtained from other systems or devices (e.g., imaging systems, clinic services, health databases, electronic medical records (ERM) systems, surgical consoles, digital microscopes, etc.).

[0042] The interface 500 may include a results section 514 that presents an IOL power 516 and a predicted postoperative refractive error 518 for each IOL power in the IOL set. The predicted refractive error 518 may be measured in diopters, such as a value (EIOL-D IOL ) / p.

[0043] refer to Figure 6 , Fig. 7A and Figure 7B The machine learning model trained according to method 300 and utilized according to method 400 can have various forms, such as Figure 6 , Fig. 7A and Figure 7B However, these are merely exemplary, and any machine learning or artificial intelligence model known in the art can be trained to perform the described tasks.

[0044] Figure 6 Demonstrated the use of clustering to improve Figure 3 and Figure 4 The method 600 of measuring the accuracy of the machine learning model described herein can include clustering the training data entries based on the values ​​of one or more of the patient attributes 204, the postoperative refractive error 202, and the derived output parameter 206 (e.g., EIOL) in each data entry in step 602. The clustering can include using a k-nearest neighbor (KNN) algorithm. Other clustering methods can be used, such as k-means, Gaussian mixture models, centroid-based clustering, density-based clustering, distribution-based clustering, and hierarchical clustering.

[0045] Then, the method 600 may include, at step 604, receiving patient attributes 204 of a patient's eye for which an IOL is to be selected for future surgery. The patient attributes 204 may include the power of the IOL selected from the set of IOLs, as described above with respect to Figure 4 The method may include, in step 606, identifying clusters associated with the patient attributes 204. The patient attributes 204 may then be processed in step 608 using a machine learning model specific to the cluster identified in step 606 to generate a predicted EIOL (or another derived output parameter). By using a machine learning model trained using only training data entries in a cluster associated with the patient attributes 204, the prediction accuracy of the machine learning model may be improved relative to a machine learning model for a different cluster or a machine learning model trained on a larger, less specific set of training data entries.

[0046] For example, step 606 may include processing the patient attributes 204 according to KNN (which is itself a machine learning model) to identify clusters of training data items in the training data set (i.e., K nearest neighbors), and then in step 608, processing the patient attributes 204 according to a second machine learning model trained using the clusters of training data items to output a predicted EIOL. The second machine learning model may include a multivariate linear regression (MLR) model (KNN+MLR) or a random sampling consensus (RANSAC) regression model (KNN+RAN). The second machine learning model may also include any machine learning model known in the art that is trained using clusters of training data items, such as DNN, CNN, multivariate polynomial regression (MPR) (2nd order, 3rd order or higher), support vector regression model (SVM), or SVM radial basis function (SVM-RBF).

[0047] See also Fig. 7A , the machine learning model trained in step 306 may include the displayed ensemble machine learning model 700a. The ensemble machine learning model 700a is a combination of multiple machine learning models 702a-702e. Each machine learning model 702a-702e may be of a different type. Non-limiting examples of these types include KNN, MLR, KNN+MLR, KNN+RAN, DNN, multivariate polynomial regression (MPR) (2nd order, 3rd order or higher), support vector regression model (SVM), SVM radial basis function (SVM-RBF),

[0048] Each machine learning model 702a-702e can be trained separately according to the method 300. The output of the machine learning models 702a-702e is passed to the hybrid algorithm 704 along with the input (patient attributes 204). The hybrid algorithm 704 itself is any type of machine learning model, such as any type of machine learning model discussed herein. The hybrid algorithm 704 can be trained to (a) select from the outputs of the machine learning models 702a-702e, or (b) combine the outputs of the machine learning models 702a-702e to produce a final result, e.g., a predicted EIOL.

[0049] Figure 7B A gradient boosting machine learning model 700b is shown, which is also composed of multiple machine learning models 702a-702e of different types (such as any of the machine learning model types mentioned herein). The first machine learning model 702a takes the patient attributes 204 as input and produces a predicted EIOL and a residual 706a. Each of the other machine learning models 702b to 702e takes the patient attributes 204 and the residuals 706a to 706d from the previous stage as input. The output of the final machine learning model 702e can then be the final EIOL prediction 708.

[0050] The ensemble machine learning model 700a and the gradient boosting machine learning model 700b can provide marginal benefits relative to any of the individual machine learning models 702a-702e. Experiments conducted by the inventors found that the ensemble machine learning model 700a or the gradient boosting machine learning model 700b can increase the percentage of correct predictions within 0.5 diopters by up to about 2%.

[0051] Tables 1 and 2 list the results of various individual machine learning models as the percentage of predicted EIOLs that are within 0.5 diopters of the actual EIOL as provided in the data entry. The results of Tables 1 and 2 were obtained using patient attributes 204 and postoperative refractive error 202 of patients who received Acrysof monofocal IOLs. The results of Table 1 were obtained using data from 17,500 patients, including three biometric variables (AL, K, and WTW) for each patient. The results of Table 2 were obtained using data from 5,277 patients. Entries marked "(Ex)" additionally used ACD and LT for each patient. Another metric used to characterize the accuracy of a machine learning model is the root mean square (RMS) of the error (predicted EIOL - EIOL in the training data entry).

[0052] Table 1: Experimental results for three variables (AL, K, WTW)

[0053] Machine Learning Models % <0.5 diopters MLR 84.9 MPR (2nd order) 87.8 SVR-RBF 87.7 DNN 88.0 KNN+MLR 87.6

[0054] Table 2: Experimental results for five variables (AL, K, WTW, ACD, LT)

[0055] Machine Learning Models % <0.5 diopters KNN+MLR 89.1 KNN+MLR(Ex) 90.5 KNN+RAN 89.9 KNN+RAN(Ex) 91.5

[0056] Figure 8 An example computing system 800 is shown that implements, at least in part, one or more functions described herein in response to input to interface 500. Computing system 800 may also implement methods 300, 400, and / or 600.

[0057] As shown, the computing system 800 includes a central processing unit (CPU) 802, one or more I / O device interfaces 804 that can allow various I / O devices 814 (e.g., keyboard, display, mouse device, pen input, etc.) to be connected to the computing system 800, a network interface 806 through which the computing system 800 is connected to a network 890 (which can be a local network, an intranet, the Internet, or any other group of computing systems communicatively connected to each other), a memory 808, a storage device 810, and an interconnect 812.

[0058] CPU 802 can retrieve and execute programming instructions stored in memory 808. Similarly, CPU 802 can retrieve and store application data residing in memory 808. Interconnect 812 transmits programming instructions and application data between CPU 802, I / O device interface 804, network interface 806, memory 808, and storage device 810. CPU 802 is included to represent a single CPU, multiple CPUs, a single CPU with multiple processing cores, etc.

[0059] The memory 808 represents a volatile memory such as a random access memory and / or a non-volatile memory such as a non-volatile random access memory, a phase change random access memory, etc. As shown, the memory 808 can store a data preparation module 816 for calculating a derived output parameter (e.g., EIOL) based on patient attribute data and a training algorithm 818 for training a machine learning model to predict the derived output parameter. The memory 808 can store a prediction module 820 that uses a machine learning model to predict the derived output parameter based on the patient attributes and the IOL power.

[0060] The storage device 810 can be a non-volatile memory, such as a disk drive, a solid-state drive, or a collection of storage devices distributed across multiple storage systems. The storage device 810 can optionally store one or more machine learning models 822 trained as described above and training data 824 for training the machine learning model 822.

[0061] Additional considerations

[0062] The foregoing description is provided to enable any person skilled in the art to practice the various embodiments described herein. Various modifications to these embodiments are clear to those skilled in the art, and the general principles defined herein may be applied to other embodiments. For example, without departing from the scope of the present disclosure, the functions and arrangements of the elements discussed may be changed. Various examples may appropriately omit, replace or add various programs or components. In addition, the features described with respect to some examples may be combined in some other examples. For example, any number of aspects set forth herein may be used to implement a device or practice method. In addition, the scope of the present disclosure is intended to cover devices or methods practiced using other structures, functions, or structures and functions in addition to or different from the various aspects of the present disclosure set forth herein. It should be understood that any aspect of the present disclosure disclosed herein may be embodied by one or more elements of the claims.

[0063] As used herein, the phrase "at least one of" a list of items refers to any combination of those items, including single members. For example, "at least one of a, b, or c" is intended to cover any combination of a, b, c, ab, ac, bc, and abc, as well as multiples of the same element (e.g., aa, aaa, aab, aac, abb, acc, bb, bbb, bbc, cc, and ccc or any other order of a, b, and c).

[0064] As used herein, the term "determining" encompasses a wide variety of actions. For example, "determining" may include calculating, computing, processing, deriving, investigating, searching (e.g., searching in a table, database, or other data structure), ascertaining, etc. Also, "determining" may include receiving (e.g., receiving information), accessing (e.g., accessing data in a memory), etc. Also, "determining" may include resolving, selecting, choosing, establishing, etc.

[0065] The method disclosed herein includes one or more steps or actions for implementing the method. Without departing from the scope of the claims, the method steps and / or actions can be interchangeable with each other. In other words, unless a specific step or action sequence is specified, the order and / or use of specific steps and / or actions can be modified without departing from the scope of the claims. Further, the various operations of the above-mentioned method can be performed by any suitable device that can perform the corresponding function. These devices may include various hardware and / or software components and / or modules, including but not limited to circuits, application specific integrated circuits (ASICs) or processors. Generally, in the case of operations shown in the figure, those operations can have corresponding corresponding devices with similar numbers plus functional components.

[0066] The various illustrative logical blocks, modules, and circuits described in conjunction with the present disclosure may be implemented or executed with a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device (PLD), discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in an alternative, the processor may be any commercially available processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.

[0067] The processing system can be implemented with a bus architecture. Depending on the specific application and overall design constraints of the processing system, the bus may include any number of interconnecting buses and bridges. The bus may link together various circuits including a processor, a machine-readable medium, and an input / output device. A user interface (e.g., a keyboard, a display, a mouse, a joystick, etc.) may also be connected to the bus. The bus may also link various other circuits such as timing sources, peripherals, voltage regulators, power management circuits, etc., which are well known in the art and therefore will not be further described. The processor may be implemented with one or more general and / or special processors. Examples include microprocessors, microcontrollers, DSP processors, and other circuit systems that can execute software. Those skilled in the art will recognize how to best implement the described functions of the processing system according to the specific application and the overall design constraints imposed on the entire system.

[0068] If implemented in software, the function may be stored as one or more instructions or codes on a computer-readable medium or transmitted via a computer-readable medium. Software should be broadly interpreted as instructions, data, or any combination thereof, whether referred to as software, firmware, middleware, microcode, hardware description language, or other. Computer-readable media include both computer storage media and communication media (such as any medium that facilitates the transfer of computer programs from one place to another). The processor may be responsible for managing the bus and general processing, including the execution of software modules stored on a computer-readable storage medium. A computer-readable storage medium may be coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. In an alternative, the storage medium may be integrated with the processor. For example, a computer-readable medium may include a transmission line, a carrier modulated by data, and / or a computer-readable storage medium on which instructions separated from a wireless node are stored, all of which may be accessed by a processor via a bus interface. Alternatively or in addition, a computer-readable medium or any portion thereof may be integrated into a processor, such as a cache and / or a general register file. For example, examples of machine-readable storage media may include RAM (random access memory), flash memory, ROM (read-only memory), PROM (programmable read-only memory), EPROM (erasable programmable read-only memory), EEPROM (electrically erasable programmable read-only memory), registers, magnetic disks, optical disks, hard drives, or any other suitable storage media, or any combination thereof. Machine-readable media may be embodied in a computer program product.

[0069] A software module may include a single instruction or multiple instructions, and may be distributed over several different code segments, between different programs, and across multiple storage media. A computer-readable medium may include multiple software modules. A software module includes instructions that, when executed by a device such as a processor, cause a processing system to perform various functions. A software module may include a transmission module and a reception module. Each software module may exist in a single storage device, or may be distributed across multiple storage devices. For example, when a triggering event occurs, a software module may be loaded from a hard drive into a RAM. During the execution of a software module, a processor may load some instructions into a cache to increase access speed. Then, one or more cache lines may be loaded into a general register file for execution by the processor. When referring to the function of a software module, it should be understood that such function is implemented by the processor when executing instructions from the software module.

[0070] The following claims are not intended to be limited to the embodiments shown herein, but are given the full scope consistent with the language of the claims. In the claims, unless otherwise specified, reference to a singular element is not intended to mean "one and only one", but "one or more". Unless otherwise specifically stated, the term "some" refers to one or more. According to 35 U.S.C. § 112 (f), the elements of any claim will not be interpreted unless the phrase "device for..." is used to expressly describe these elements, or in the case of a method claim, the phrase "step for..." is used to describe these elements. All structural and functional equivalents of the elements of the various aspects described throughout this disclosure that are known or will be known to those of ordinary skill in the art are expressly incorporated herein by reference and are intended to be covered by the claims. In addition, whether or not such disclosure is explicitly described in the claims, the content disclosed herein is not intended to be committed to the public.

Claims

1. A system for guiding intraocular lens (IOL) selection, the system comprising: one or more processing devices; One or more memory devices operably coupled to the one or more processing devices, the one or more memory devices storing executable code that, when executed by the one or more processing devices, causes the one or more processing devices to: receiving patient attribute data; receiving a set of one or more IOL characteristics describing one or more IOLs; as well as For each of the one or more IOL characteristics: Processing the patient attribute data and each IOL characteristic using a machine learning model to obtain output parameters; as well as calculating a predicted refractive error based on a combination of each IOL characteristic and the output parameter; as well as The predicted refractive error for each of the one or more IOLs is output.

2. The system of claim 14, wherein: The combination of each IOL characteristic and the output parameter comprises a difference between the output parameter and the IOL characteristic.

3. The system of claim 15, wherein: Each combination of the IOL characteristic and the output parameter comprises a difference between the output parameter and the IOL characteristic, the difference being scaled by a parameter between 0.66 and 0.

7.

4. The system of claim 15, wherein: The patient attribute data includes the dimensions of the patient's eyes.

5. The system of claim 17, wherein: The dimensions of the patient's eye are any of the following: white-to-white distance (WTW), axial length (AL), mean corneal power (K), lens thickness (LT), and anterior chamber depth (ACD).

6. The system of claim 18, wherein: The set of one or more IOL characteristics of the one or more IOLs includes a plurality of IOL characteristics of a plurality of IOLs; Wherein, when the executable code is executed by the one or more processing devices, the one or more processing devices further cause the one or more processing devices to: calculating initial IOL characteristics based on the dimensions of the patient's eye; and The plurality of IOL characteristics are selected based on the initial IOL characteristics.

7. The system of claim 14, wherein: The machine learning models include one or more machine learning models selected from the group consisting of: K nearest neighbor model; Multiple linear regression model; Random sampling consistency model; Multivariate polynomial regression models; Support Vector Machine-Radial Basis Function; and Deep neural network models.

Images