Method and device for predicting vault height after ICL surgery

By constructing a postoperative arch height prediction operator for ICL using a LASSO regression model combined with multidimensional feature data, the problems of insufficient prediction accuracy and interpretability in existing technologies are solved, enabling more accurate prediction of postoperative arch height and preoperative plan evaluation.

CN123058831APending Publication Date: 2026-11-10THE SECOND AFFILIATED HOSPITAL OF ZHEJIANG UNIVERSITY SCHOOL OF MEDICINE (SECOND HOSPITAL OF ZHEJIANG PROVINCE)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611585576.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-14
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In existing technologies, the prediction of ICL postoperative arch height relies on a one-dimensional span parameter, ignoring the anterior-posterior positional relationship of the lens relative to the ciliary sulcus plane and the morphology of the bimeridian posterior chamber, which limits the accuracy of prediction. Furthermore, traditional machine learning methods lack interpretability and are difficult to integrate surgical decision variables.

Method used

A LASSO regression model was used to construct an arch height prediction operator. The operator combined variables such as the lens elevation height in the horizontal ciliary sulcus plane, the lens elevation height in the vertical ciliary sulcus plane, the ICL lens size, and the implantation direction. The parameters were estimated and variables were selected through L1 regularization penalty term. The predicted value of the ICL postoperative arch height was output, and interpretable contribution information was provided.

Benefits of technology

It improves the accuracy and interpretability of ICL postoperative arch height prediction, assists in preoperative plan evaluation, reduces prediction error, and enhances the scientific nature of surgical decisions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123058831A_ABST
    Figure CN123058831A_ABST
Patent Text Reader

Abstract

The application provides an ICL postoperative vault height prediction method and device, and relates to the technical field of data processing.The method comprises the following steps: obtaining preoperative basic data, binocular UBM feature data and surgical decision variables of a target object; performing encoding processing on the preoperative basic data, the binocular UBM feature data and the surgical decision variables to obtain corresponding prediction input variables; inputting the prediction input variables into a pre-constructed vault height prediction operator, wherein the vault height prediction operator comprises an intercept term and a correction term corresponding to each prediction input variable; and outputting an ICL postoperative vault height prediction value of the target object under a candidate ICL lens size and a tentative ICL implantation direction based on a superposition result of the intercept term and the correction terms.The ICL postoperative vault height prediction method and device provided by the application can improve the accuracy and interpretability of ICL postoperative vault height prediction and provide auxiliary information for preoperative scheme evaluation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and device for predicting the arch height after ICL surgery. Background Technology

[0002] Postoperative arch height is a crucial indicator for evaluating the safety and effectiveness of ICL (Implantable Collamer Lens) implantation. Preoperative prediction of postoperative arch height helps in selecting an appropriate ICL lens size and reducing the risk of postoperative ocular damage. Traditional techniques often rely on one-dimensional span parameters such as the diameter from the ciliary sulcus to the sulcus for predicting postoperative arch height, neglecting the anterior-posterior position of the lens relative to the ciliary sulcus plane and the bimeridian posterior chamber morphology, thus limiting prediction accuracy. While traditional machine learning methods are employed, they often suffer from black-box effects, making it difficult to explain the influence of various parameters on the prediction results. Furthermore, they struggle to integrate surgical decision variables such as ICL lens size and implantation direction, further impacting prediction accuracy. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method and device for predicting the arch height after ICL surgery, so as to alleviate the above-mentioned technical problems and improve the accuracy of ICL postoperative arch height prediction.

[0004] In a first aspect, embodiments of the present invention provide a method for predicting the postoperative arch height of an ICL, the method comprising: acquiring preoperative baseline data, bimeridian UBM feature data, and surgical decision variables of a target subject; wherein the bimeridian UBM feature data includes at least horizontal ciliary sulcus plane lens elevation height (HSTSL) and vertical ciliary sulcus plane lens elevation height (VSTSL), and the surgical decision variables include at least candidate ICL lens size and proposed ICL implantation direction; encoding the preoperative baseline data, the bimeridian UBM feature data, and the surgical decision variables to obtain corresponding prediction input variables; inputting the prediction input variables into a pre-constructed arch height prediction operator, wherein the arch height prediction operator includes an intercept term and correction terms corresponding to each prediction input variable; and outputting the predicted postoperative arch height of the target subject under the candidate ICL lens size and the proposed ICL implantation direction based on the superposition result of the intercept term and each correction term by the arch height prediction operator.

[0005] In conjunction with the first aspect, the present invention provides a first possible implementation of the first aspect, wherein the arch height prediction operator is constructed based on the LASSO regression model; the LASSO regression model is trained based on a training set containing clinical data, and L1 regularization penalty term is used for parameter estimation when training the LASSO regression model.

[0006] In conjunction with the first possible implementation of the first aspect, this embodiment of the invention provides a second possible implementation of the first aspect, wherein the method further includes: obtaining an initial feature set, the initial feature set including multiple sets of clinical data, and a label corresponding to each set of clinical data, the label being used to characterize the actual postoperative arch height value corresponding to the clinical data; performing parameter estimation using a cost function containing the L1 regularization penalty term, and determining the optimal penalty parameter through a cross-validation mechanism; under the optimal penalty parameter, shrinking the variable coefficients corresponding to each candidate variable in the clinical data, and selecting candidate input variables based on the shrunken variable coefficients, so as to construct the arch height prediction operator based on the candidate input variables, wherein the candidate input variables include the preoperative baseline data, the bimeridian UBM feature data, and the surgical decision variables.

[0007] In conjunction with the first aspect, this embodiment of the invention provides a third possible implementation of the first aspect, wherein the step of obtaining the bimeridian UBM feature data of the target object includes: obtaining UBM images of the target object along a horizontal meridian and a vertical meridian; identifying the two endpoints of the ciliary sulcus in the UBM images of the horizontal meridian and the vertical meridian, and constructing a horizontal ciliary sulcus-to-sulcus reference line and a vertical ciliary sulcus-to-sulcus reference line respectively; measuring the rise height of the lens relative to the corresponding ciliary sulcus-to-sulcus reference line to obtain the horizontal ciliary sulcus plane lens rise height HSTSL and the vertical ciliary sulcus plane lens rise height VSTSL; and obtaining the horizontal ciliary sulcus-to-sulcus diameter HSTS and the vertical ciliary sulcus-to-sulcus diameter VSTS based on the two endpoints of the ciliary sulcus.

[0008] In conjunction with the first aspect, this embodiment of the invention provides a fourth possible implementation of the first aspect, wherein the step of obtaining surgical decision variables of the target object includes: obtaining preoperative surgical parameters of the target object, and obtaining surgical decision variables from the preoperative surgical parameters, wherein the surgical decision variables include candidate ICL lens size and proposed ICL surgical implantation direction.

[0009] In conjunction with the first aspect, this embodiment of the invention provides a fifth possible implementation of the first aspect, wherein the aforementioned preoperative basic data includes the demographic parameters and ocular measurement parameters of the target subject.

[0010] In conjunction with the first aspect, this embodiment of the invention provides a sixth possible implementation of the first aspect, wherein the step of outputting the predicted ICL postoperative arch height of the target object under the candidate ICL lens size and the proposed ICL implantation direction based on the superposition result of the intercept term and each of the correction terms by the arch height prediction operator includes: superimposing the correction terms corresponding to the preoperative baseline data, the bimeridian UBM feature data and the surgical decision variables respectively by the arch height prediction operator; adding the superposition result to the intercept term to obtain the predicted ICL postoperative arch height.

[0011] In conjunction with the second possible implementation of the first aspect, this embodiment of the invention provides a seventh possible implementation of the first aspect, wherein the above method further includes: outputting contribution information of each prediction input variable to the predicted arch height value after ICL surgery based on the arch height prediction operator; wherein the contribution information includes correction terms corresponding to each prediction input variable and / or contribution values ​​obtained based on the SHAP method.

[0012] In conjunction with the first aspect, the present invention provides an eighth possible implementation of the first aspect, wherein the above method further includes: generating predicted ICL postoperative arch height values ​​corresponding to multiple candidate surgical plans by changing the candidate ICL lens size and / or the proposed ICL implantation direction, and outputting preoperative simulation results to assist in evaluating the predicted arch height under different ICL lens sizes and ICL implantation directions.

[0013] Secondly, embodiments of the present invention also provide an ICL postoperative arch height prediction device. The device includes an acquisition module for acquiring preoperative baseline data, bimeridian UBM feature data, and surgical decision variables of the target object; wherein the bimeridian UBM feature data includes at least horizontal ciliary sulcus plane lens elevation (HSTSL) and vertical ciliary sulcus plane lens elevation (VSTSL), and the surgical decision variables include at least candidate ICL lens size and proposed ICL implantation direction; an encoding module for encoding the preoperative baseline data, the bimeridian UBM feature data, and the surgical decision variables to obtain corresponding prediction input variables; and a prediction output module for inputting the prediction input variables into a pre-constructed arch height prediction operator, wherein the arch height prediction operator includes an intercept term and correction terms corresponding to each prediction input variable; and outputs the predicted ICL postoperative arch height of the target object under the candidate ICL lens size and the proposed ICL implantation direction based on the superposition result of the intercept term and each correction term.

[0014] Thirdly, embodiments of the present invention also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the steps of the method of the first aspect.

[0015] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, performs the steps of the method described in the first aspect.

[0016] The embodiments of the present invention bring the following beneficial effects: This invention provides a method and apparatus for predicting postoperative ICL arch height. It acquires preoperative baseline data, bimeridian UBM feature data, and surgical decision variables of the target subject. The preoperative baseline data, bimeridian UBM feature data, and surgical decision variables are encoded to obtain corresponding prediction input variables. These prediction input variables are then input into a pre-constructed arch height prediction operator. Based on the superposition of the intercept term and various correction terms, the arch height prediction operator outputs the predicted postoperative ICL arch height of the target subject under the candidate ICL lens size and the proposed ICL implantation direction. During the prediction process, the horizontal ciliary sulcus plane lens elevation height (HSTSL), vertical ciliary sulcus plane lens elevation height (VSTSL), ICL lens size, and ICL implantation direction are incorporated into the same interpretable prediction framework. This helps improve the accuracy and interpretability of postoperative arch height prediction. Furthermore, by changing the candidate ICL lens size and / or the proposed ICL implantation direction, preoperative simulation results can be output, providing auxiliary information for preoperative plan evaluation.

[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0019] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1A flowchart illustrating a method for predicting the arch height after ICL surgery, provided in an embodiment of the present invention; Figure 2 A schematic diagram illustrating the contribution level of an input variable provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an ICL postoperative arch height prediction device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions 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.

[0022] Currently, the prediction of postoperative arch height after ICL surgery has the following limitations: (1) Incomplete spatial representation: In related technologies, only the horizontal or vertical span is often considered, ignoring the longitudinal height of the lens protruding into the posterior chamber (Lens Rise), which leads to a large bias in the prediction model for patients with complex anatomical structures.

[0023] (2) Black box effect: Traditional machine learning models lack interpretability, and doctors cannot understand the specific contribution of each anatomical parameter to the prediction results.

[0024] (3) Weak clinical adaptability: Traditional prediction methods are often difficult to effectively predict surgical decision variables such as ICL lens size and ICL implantation direction; different implantation directions will change the relative position of ICL foot loops in the horizontal and vertical ciliary groove areas, thus affecting the postoperative arch height.

[0025] Based on this, the present invention provides a method and device for predicting the arch height after ICL surgery, which can effectively alleviate the above-mentioned technical problems and improve the accuracy of ICL postoperative arch height prediction.

[0026] To facilitate understanding of this embodiment, a method for predicting the arch height after ICL surgery disclosed in this embodiment will be described in detail first.

[0027] In one possible implementation, embodiments of the present invention provide a method for predicting the arch height after ICL surgery, such as... Figure 1 The flowchart shown is a method for predicting the arch height after ICL surgery. The method includes the following steps: Step S102: Obtain the preoperative baseline data, bimeridian UBM feature data, and surgical decision variables of the target object; In this embodiment of the invention, the preoperative basic data includes demographic parameters and ocular measurement parameters of the target subject. The bimeridian UBM feature data includes at least the horizontal ciliary sulcus plane lens elevation height (HSTSL) and the vertical ciliary sulcus plane lens elevation height (VSTSL). The surgical decision variables include at least the candidate ICL lens size and the proposed ICL implantation direction.

[0028] Step S104: Encode the preoperative baseline data, bimeridian UBM feature data, and surgical decision variables to obtain the corresponding predictive input variables; In practice, since the original data forms of preoperative baseline data, bimeridian UBM feature data, and surgical decision variables are different, they may include text, numerical data, image data, etc. Therefore, in actual use, it is necessary to encode the preoperative baseline data, bimeridian UBM feature data, and surgical decision variables to eliminate the differences in form and dimension between different data. The specific encoding method can be set according to the actual use situation, and the embodiments of the present invention do not limit this.

[0029] Step S106: Input the predicted input variables into the pre-constructed arch height prediction operator, wherein the arch height prediction operator includes an intercept term and a correction term corresponding to each predicted input variable; based on the superposition result of the intercept term and each correction term, the arch height prediction operator outputs the predicted arch height of the target object after ICL under the candidate ICL lens size and the proposed ICL implantation direction.

[0030] In practical use, the target object in this embodiment of the invention is the patient undergoing preoperative evaluation for ICL implantation. Before surgery, the postoperative ICL arch height prediction method provided in this embodiment can be used to predict the postoperative ICL arch height. Furthermore, in this embodiment, in addition to combining the target object's preoperative baseline data, the influence of bimeridian UBM characteristic data and surgical decision variables on the postoperative arch height is fully considered, which helps to improve the accuracy and interpretability of the postoperative arch height prediction and provides auxiliary information for preoperative plan evaluation.

[0031] In specific implementation, in this embodiment of the invention, the above-mentioned preoperative basic data includes the demographic parameters of the target object, which usually refer to age, sex, preoperative equivalent spherical power (SE), etc., and the ocular measurement parameters refer to conventional ocular measurement parameters, which usually include axial length (AL), lens thickness (LT), corneal diameter (WTW), anterior chamber depth (ACD), pupil diameter (PD), etc.

[0032] Furthermore, the aforementioned bimeridian UBM (Ultrasound Biomicroscopy) feature data is extracted from UBM images. Specifically, UBM images of the target object along the horizontal and vertical meridians can be acquired; the two endpoints of the ciliary sulcus can be identified in the horizontal and vertical UBM images, and horizontal and vertical ciliary sulcus-to-sulcus reference lines can be constructed respectively; the elevation of the lens relative to the corresponding ciliary sulcus-to-sulcus reference line can be measured to obtain the horizontal ciliary sulcus plane lens elevation height HSTSL (Horizontal Sulcus to Lens Sagittal Height / Elevation) and the vertical ciliary sulcus plane lens elevation height VSTSL (Vertical Sulcus to Lens Sagittal Height / Elevation); and the horizontal ciliary sulcus to sulcus diameter HSTS (Horizontal Sulcus to Sulcus Diameter) and the vertical ciliary sulcus to sulcus diameter VSTS (Vertical Sulcus to Sulcus Diameter) can be obtained based on the two endpoints of the ciliary sulcus. Among them, HSTSL and VSTSL are the core UBM features that reflect the height of the bimeridian lens relative to the ciliary sulcus plane.

[0033] Furthermore, the surgical decision variables in this embodiment are extracted from preoperative surgical parameters. Specifically, preoperative surgical parameters of the target object can be obtained, and surgical decision variables can be derived from these parameters. In this embodiment, the surgical decision variables include the candidate ICL lens size and the proposed ICL implantation direction. Specifically, the candidate ICL lens size can include at least one of 12.1 mm, 12.6 mm, 13.2 mm, and 13.7 mm, and the proposed ICL implantation direction can include at least one of horizontal, vertical, or oblique orientations. The specific settings depend on the actual usage, and this embodiment does not impose any limitations on these settings. Moreover, during preoperative plan evaluation, multiple candidate surgical plans can be generated by changing the candidate ICL lens size and / or the proposed ICL implantation direction, resulting in predicted ICL postoperative arch height values, and the preoperative simulation results are output. These preoperative simulation results are used to assist in evaluating the predicted arch height under different ICL lens sizes and ICL implantation directions, and are not used to directly determine the final surgical plan or perform the surgery.

[0034] Furthermore, the arch height prediction operator in the embodiments of the present invention is constructed based on the LASSO (Least Absolute Shrinkage and Selection Operator) regression model; and the LASSO regression model is trained on a training set containing clinical data, and the L1 regularization penalty term is used for parameter estimation when training the LASSO regression model, and the L1 regularization penalty term can also shrink the variable coefficients when training the LASSO regression model.

[0035] In practical use, the aforementioned clinical data refers to data collected clinically, typically including age, sex, intraocular pressure, comprehensive spherical power, central corneal thickness, average corneal curvature, anterior chamber depth, anterior chamber volume, angle width, pupil diameter, axial length, lens thickness, horizontal ciliary sulcus-to-sulcus diameter, vertical ciliary sulcus-to-sulcus diameter, horizontal ciliary body plane lens elevation, vertical ciliary body plane lens elevation, ICL lens size, ICL implantation direction, ICL type, and postoperative monitoring data on the actual arch height, etc. That is, the preoperative baseline data, bimeridian UBM characteristic data, and surgical decision variables obtained in this embodiment of the invention are all data included in the clinical data.

[0036] In this embodiment of the invention, during the training of the LASSO regression model, LASSO shrinks the coefficients of each candidate variable using an L1 regularization penalty term, and then selects variables based on the shrunken coefficients. Specifically, the training process of the LASSO regression model includes the following steps: (1) Obtain the initial feature set; The initial feature set includes multiple sets of clinical data, and a label corresponding to each set of clinical data, which is used to characterize the actual postoperative arch height value corresponding to the clinical data. (2) The cost function containing L1 regularization penalty term is used for parameter estimation, and the optimal penalty parameter is determined by cross-validation mechanism; (3) Under the optimal penalty parameter, the coefficients of each candidate variable in the clinical data are shrunk, and candidate input variables are selected based on the shrunk variable coefficients to construct an arch height prediction operator based on the candidate input variables. The candidate input variables include the above-mentioned preoperative basic data, bimeridian UBM feature data and surgical decision variables.

[0037] In practice, the core rule of the LASSO regression model (hereinafter referred to as the model) lies in shrinking the variable coefficients through an L1 regularization penalty term. That is, based on ordinary linear regression, a penalty weight λ is applied to the sum of the absolute values ​​of the variable coefficients. As λ increases, some variable coefficients can be shrunk to 0, thereby achieving variable screening. Variables with non-zero coefficients can be used as candidate input variables for subsequent evaluation. The final input variables can be determined based on non-zero coefficients, combined with clinical relevance, biological rationality, and model interpretability.

[0038] In specific implementation, in order to determine the penalty weight λ, the present invention employs a cross-validation mechanism to determine the optimal penalty parameter. Specifically, a 10-fold cross-validation mechanism can be used to determine λmin. That is, the training data in the initial feature set is divided into 10 equal parts for data iteration. Nine of these parts are used to train the model, and one part is used to verify the effect. The deviation change curves under different λ values ​​are recorded. The λ value that minimizes the mean square error is selected as λmin. Under λmin, the final candidate input variables are determined based on the non-zero coefficients and in combination with clinical relevance, biological rationality, and model interpretability, forming the final arch height prediction operator.

[0039] In other words, the LASSO regression model shrinks and filters the coefficients of each candidate input variable, reducing model complexity and preventing overfitting, while retaining clinically relevant and interpretable input variables.

[0040] That is, during the training of the LASSO regression model, the input parameters include labeled clinical data, which are collected based on the patients' clinical data. The input parameters also include the actual postoperative arch height value corresponding to each group of clinical data. The output is the arch height prediction operator in this embodiment of the invention, which determines which data in the clinical data are retained.

[0041] Furthermore, the arch height prediction operator in this embodiment of the invention includes an intercept term and a correction term calculated based on the aforementioned preoperative baseline data, UBM feature data, and surgical decision variables. In this embodiment of the invention, the correction terms corresponding to the preoperative baseline data, bimeridian UBM feature data, and surgical decision variables can be superimposed using the arch height prediction operator. The superposition result is added to the intercept term to obtain the predicted arch height value after ICL surgery.

[0042] Specifically, the formula for the arch height prediction operator in this embodiment of the invention can be expressed as: Predicted Vault = V0 + 60.07 × ACD + 2.56 × AL + 3.48 × Age -39.36 × HSTS - 361.4 × HSTSL - 110.02 × ICLsize.12.1 + 110.24 ×ICLsize.13.2 + 248.27 × ICLsize.13.7 - 131.68 × LT + 14.43 × Orientation.Horizontal - 68.53 × Orientation.Vertical + 11.81 × PD - 12.31 × SE + 36.34 × Sex.Female + 19.88 × VSTS - 260.12 × VSTSL + 41.8 × WTW; In this context, V0 represents the intercept term of the arch height prediction operator, also known as the baseline arch height value; ACD (Anterior Chamber Depth) refers to the anterior chamber depth; AL (Axial Length) refers to the axial length; Age represents age; HSTS represents the horizontal ciliary sulcus-to-sulcus diameter; VSTS represents the vertical ciliary sulcus-to-sulcus diameter; HSTSL represents the lens elevation at the horizontal ciliary sulcus plane; VSTSL represents the lens elevation at the vertical ciliary sulcus plane; LT (Lens Thickness) represents lens thickness; PD (Pupil Diameter) represents pupil diameter; WTW (White-to-White Distance) represents corneal diameter; and SE (Spherical Equivalent) represents the equivalent spherical power. The units for ACD, AL, LT, WTW, HSTS, VSTS, HSTSL, VSTSL, and PD are all mm; the unit for SE is D; the unit for Age is years; and the unit for the post-ICL arch height prediction value is μm.

[0043] Furthermore, ICLsize.12.1, ICLsize.13.2, and ICLsize.13.7 are dummy variables for ICL lens size, with 12.6 mm serving as the reference group; Orientation.Horizontal and Orientation.Vertical are dummy variables for ICL implantation direction, with oblique implantation direction serving as the reference group; Sex.Female is a dummy variable for gender, with male serving as the reference group. The corresponding dummy variable values ​​are determined as follows: 1 is taken when the category is selected, and 0 is taken otherwise.

[0044] Furthermore, based on the aforementioned prediction algorithm, in this embodiment of the invention, the predicted arch height can be output in real time based on the preoperative baseline data, bimeridian UBM feature data, and surgical decision variables. Furthermore, in this embodiment of the invention, the contribution information of each prediction input variable to the predicted arch height after ICL surgery can also be output based on the arch height prediction operator; wherein, the contribution information includes correction terms corresponding to each prediction input variable and / or contribution values ​​obtained based on the SHAP method. For example, a SHAP (SHapley Additive exPlanations) interpretation diagram can be generated to display the contribution of each prediction input variable to the predicted arch height after ICL surgery.

[0045] For ease of understanding, Figure 2 This illustrates the contribution level of an input variable, such as... Figure 2 As shown, in addition to the various input variables included in the expression formula of the arch height prediction operator mentioned above, Figure 2 The paper also shows the orientation (Oblique) of ICL implantation, as well as the contribution of candidate variables such as Km (Mean Keratometry), IOP (Intraocular Pressure), ACA (Anterior Chamber Angle), ACV (Anterior Chamber Volume), and CCT (Central Corneal Thickness), based on... Figure 2 It can display the positive or negative contributions of different input variables to the predicted arch height after ICL surgery, among which, Figure 2 In the graph, red indicates positive contribution and blue indicates negative contribution. The horizontal axis represents the degree of contribution, and the vertical axis represents each candidate variable.

[0046] Furthermore, for ease of understanding, the following example illustrates the process of predicting the arch height after ICL surgery. Specifically, the input data includes: Preoperative baseline data: Demographic parameters: Sex: Female; Age: 21; Comprehensive spherical lens: -10.25 D; Ocular measurement parameters: AL = 28.56, LT = 3.35, WTW = 11.9, ACD = 3.67; Bimeridian UBM characteristic data: HSTS = 12.17, VSTS = 12.57, HSTSL = 0.34 mm, VSTSL = 0.40 mm (belonging to low height characteristics, with ample space); Surgical decision variables: ICL lens size: 13.2mm, ICL implantation direction: Vertical implantation; The intercept term was determined to be 534 μm; During the prediction process, based on the above formula for the arch height prediction operator, the positive contributions to the intercept term include: when the candidate ICL lens size is 13.2mm, the corresponding ICL lens size dummy variable is set to 1 and a corresponding positive correction is generated; HSTSL and VSTSL are used to reflect the bimeridian elevation of the lens relative to the ciliary sulcus plane and generate corresponding corrections; the negative contributions include: when the proposed ICL implantation direction is selected as "vertical", the Orientation.Vertical dummy variable is set to 1 and a corresponding negative correction is generated for the predicted arch height after ICL surgery.

[0047] Output results: The final predicted arch height after ICL surgery was 827 μm, while clinical observation showed that the actual postoperative arch height in this case was 749 μm. The error between the predicted arch height and the actual postoperative arch height was less than the preset error threshold.

[0048] In practical use, the effectiveness of the above-mentioned ICL postoperative arch height prediction method can also be evaluated in the embodiments of the present invention. Specifically, the difference between the above-mentioned ICL postoperative arch height prediction value and the actual postoperative arch height value observed clinically can be calculated, and the mean absolute error (MAE) can be used for measurement.

[0049] For example, a portion of the training set can be used as a validation set. That is, for each sample, such as a set of clinical data and corresponding labels, candidate input variables are extracted from the clinical data and fed into the arch height prediction operator to obtain the predicted arch height after ICL surgery. The actual postoperative arch height value is extracted from the labels, and the difference between the two is calculated and its absolute value is taken to obtain the absolute error. The formula is as follows: e i = | Vault_predicted, i – Vault_actual, i |; Where i represents the first sample, Vault_predicted, i represents the predicted arch height of the first sample after ICL surgery, and Vault_actual, i represents the label of the first sample, i.e., the actual postoperative arch height value.

[0050] The mean absolute error (MAE) is obtained by summing the absolute errors of all samples (n=284) in the validation set and then dividing by the sample size. The formula for calculating MAE is as follows:

[0051] The validation set shows that the MAE is 93.2 μm. That is, the average prediction error of the arch height prediction operator in this embodiment is only 93.2 μm, demonstrating good prediction accuracy.

[0052] In addition, clinical reliability can also be calculated. Similarly, the calculation is performed using a validation set. The proportion of samples with an absolute error of less than or equal to 150 μm between the predicted ICL postoperative arch height and the actual postoperative arch height can be calculated. If this proportion is less than a preset threshold, such as less than 80%, it indicates that the ICL postoperative arch height prediction method provided in this embodiment of the invention has good prediction reliability.

[0053] Furthermore, based on the above embodiments, this invention also provides a device for predicting the arch height after ICL surgery, such as... Figure 3 The diagram shows a structural schematic of an ICL postoperative arch height prediction device, which includes: The acquisition module 30 is used to acquire the preoperative basic data, bimeridian UBM feature data, and surgical decision variables of the target object; wherein, the bimeridian UBM feature data includes at least the horizontal ciliary sulcus plane lens elevation height HSTSL and the vertical ciliary sulcus plane lens elevation height VSTSL, and the surgical decision variables include at least the candidate ICL lens size and the proposed ICL implantation direction; Encoding module 32 is used to encode the preoperative basic data, the bimeridian UBM feature data, and the surgical decision variables to obtain the corresponding predictive input variables; The prediction output module 34 is used to input the prediction input variables into a pre-constructed arch height prediction operator, wherein the arch height prediction operator includes an intercept term and a correction term corresponding to each prediction input variable; based on the superposition result of the intercept term and each correction term, the arch height prediction operator outputs the predicted arch height of the target object after ICL under the candidate ICL lens size and the proposed ICL implantation direction.

[0054] The ICL postoperative arch height prediction device provided in this embodiment of the invention has the same technical features as the ICL postoperative arch height prediction method provided in the above embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0055] Furthermore, embodiments of the present invention also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above method.

[0056] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the above-described method.

[0057] Furthermore, embodiments of the present invention also provide a schematic diagram of the structure of an electronic device, such as... Figure 4The diagram shows the structure of the electronic device, which includes a processor 41 and a memory 40. The memory 40 stores computer-executable instructions that can be executed by the processor 41, and the processor 41 executes the computer-executable instructions to implement the above-described method.

[0058] exist Figure 4 In the illustrated embodiment, the electronic device further includes a bus 42 and a communication interface 43, wherein the processor 41, the communication interface 43, and the memory 40 are connected via the bus 42.

[0059] The memory 40 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 43 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 42 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 42 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0060] Processor 41 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 41 or by software instructions. Processor 41 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this invention can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in the memory, and the processor 41 reads the information in the memory and uses its hardware to complete the aforementioned method.

[0061] The computer program product of the method and device for predicting the arch height after ICL surgery provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0062] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0063] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0064] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0065] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0066] Finally, it should be noted that the above embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for predicting the arch height after ICL surgery, characterized in that, The method includes: Acquire preoperative baseline data, bimeridian UBM feature data, and surgical decision variables for the target subject; wherein, the bimeridian UBM feature data includes at least the horizontal ciliary sulcus plane lens elevation height (HSTSL) and the vertical ciliary sulcus plane lens elevation height (VSTSL), and the surgical decision variables include at least the candidate ICL lens size and the proposed ICL implantation direction; The preoperative baseline data, the bimeridian UBM feature data, and the surgical decision variables are encoded to obtain the corresponding predictive input variables; The predicted input variables are input into a pre-constructed arch height prediction operator, wherein the arch height prediction operator includes an intercept term and a correction term corresponding to each predicted input variable; based on the superposition result of the intercept term and each correction term, the arch height prediction operator outputs the predicted arch height of the target object after ICL under the candidate ICL lens size and the proposed ICL implantation direction.

2. The method according to claim 1, characterized in that, The arch height prediction operator is constructed based on the LASSO regression model; the LASSO regression model is trained on a training set containing clinical data, and L1 regularization penalty term is used for parameter estimation when training the LASSO regression model.

3. The method according to claim 2, characterized in that, The method further includes: An initial feature set is obtained, which includes multiple sets of clinical data and a label corresponding to each set of clinical data. The label is used to characterize the actual postoperative arch height value corresponding to the clinical data. The parameters are estimated using a cost function containing the L1 regularization penalty term, and the optimal penalty parameters are determined through a cross-validation mechanism. Under the optimal penalty parameters, the coefficients of each candidate variable in the clinical data are shrunk, and candidate input variables are selected based on the shrunk coefficients to construct the arch height prediction operator based on the candidate input variables. The candidate input variables include the preoperative baseline data, the bimeridian UBM feature data, and the surgical decision variables.

4. The method according to claim 1, characterized in that, The steps for obtaining the bimeridian UBM feature data of the target object include: Acquire UBM images of the target object along the horizontal and vertical meridians; Identify the two endpoints of the ciliary sulcus in the UBM images of the horizontal meridian and the vertical meridian, and construct the horizontal ciliary sulcus-to-sulcus reference line and the vertical ciliary sulcus-to-sulcus reference line respectively; The lens elevation relative to the corresponding ciliary sulcus-to-sulcus reference line is measured to obtain the horizontal ciliary sulcus plane elevation height HSTSL and the vertical ciliary sulcus plane elevation height VSTSL; and the horizontal ciliary sulcus-to-sulcus diameter HSTS and the vertical ciliary sulcus-to-sulcus diameter VSTS are obtained based on the two endpoints of the ciliary sulcus.

5. The method according to claim 1, characterized in that, The steps to obtain the surgical decision variables for the target subject include: Obtain the preoperative surgical parameters of the target object, and obtain surgical decision variables from the preoperative surgical parameters, wherein the surgical decision variables include the candidate ICL lens size and the proposed ICL implantation direction.

6. The method according to claim 1, characterized in that, The preoperative baseline data includes the demographic parameters and ocular measurement parameters of the target subject.

7. The method according to claim 1, characterized in that, The step of outputting the predicted ICL postoperative arch height of the target object under the candidate ICL lens size and the proposed ICL implantation direction based on the superposition result of the intercept term and each of the correction terms using the arch height prediction operator includes: The correction terms corresponding to the preoperative baseline data, the bimeridian UBM feature data, and the surgical decision variables are superimposed using the arch height prediction operator. The superposition result is added to the intercept term to obtain the predicted arch height after ICL surgery.

8. The method according to claim 3, characterized in that, The method further includes: Based on the arch height prediction operator, output the contribution information of each prediction input variable to the predicted arch height value after ICL surgery; The contribution information includes correction terms corresponding to each predicted input variable and / or contribution values ​​obtained based on the SHAP method.

9. The method according to claim 1 or 5, characterized in that, The method further includes: By changing the candidate ICL lens size and / or the proposed ICL implantation direction, predicted ICL postoperative arch height values ​​are generated for multiple candidate surgical plans, and preoperative simulation results are output to assist in evaluating the predicted arch height under different ICL lens sizes and ICL implantation directions.

10. A device for predicting the arch height after ICL surgery, characterized in that, The device includes: The acquisition module is used to acquire the preoperative basic data, bimeridian UBM feature data, and surgical decision variables of the target object; wherein, the bimeridian UBM feature data includes at least the horizontal ciliary sulcus plane lens elevation height (HSTSL) and the vertical ciliary sulcus plane lens elevation height (VSTSL), and the surgical decision variables include at least the candidate ICL lens size and the proposed ICL implantation direction; The encoding module is used to encode the preoperative baseline data, the bimeridian UBM feature data, and the surgical decision variables to obtain the corresponding predictive input variables. The prediction output module is used to input the prediction input variables into a pre-constructed arch height prediction operator, wherein the arch height prediction operator includes an intercept term and a correction term corresponding to each prediction input variable; based on the superposition result of the intercept term and each correction term, the arch height prediction operator outputs the predicted arch height of the target object after ICL surgery under the candidate ICL lens size and the proposed ICL implantation direction.

11. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, performs the steps of the method according to any one of claims 1-9.