A personalized prediction method for the efficacy of corneal cross-linking surgery, a computer-readable storage medium, and a program product

Abstract

A personalized prediction method for the efficacy of corneal cross-linking surgery, including: obtaining high-resolution OCT images and corneal topography of patients, constructing a personalized three-dimensional eyeball model including cornea and sclera; establishing a biomechanical material model of the cornea, a keratoconus progression model, and a corneal cross-linking mechanism model based on clinical data; constructing a finite element model of the eyeball by meshing and boundary condition determination; simulating corneal cross-linking surgery using finite element method and artificial intelligence technology, constructing a real-time prediction system, and generating postoperative efficacy prediction results. The present invention is based on real patient data, combined with numerical simulation and artificial intelligence technology, to achieve personalized real-time prediction of surgical efficacy, assist preoperative patient science education, doctor treatment decision-making selection, and telemedicine, and significantly improve the accuracy and personalization of surgical planning.

Description

Technical Field

[0001] The present invention relates to the fields of medical image processing and computer simulation, and in particular to a method for personalized prediction of the efficacy of corneal cross-linking surgery. Background Art

[0002] The cornea is the transparent outermost tissue of the eyeball and an important part of the visual system. Corneal ectasia (such as keratoconus) is a degenerative disease characterized by corneal thinning and protrusion, which can lead to decreased vision and worsening astigmatism. In severe cases, it may cause corneal perforation and blindness. Its incidence is increasing worldwide and has become one of the main diseases threatening vision health.

[0003] Corneal crosslinking (CXL) is an important means of treating corneal ectasia. This surgery enhances the cross-linking of corneal collagen fibers through the photochemical reaction induced by riboflavin and ultraviolet A (UVA), thereby improving the mechanical strength of the cornea and delaying the progression of the disease. At present, the Dresden protocol is widely used in clinical practice, which is a standardized surgical procedure. However, due to individual differences in corneal structure, biomechanical properties and degree of lesions, a unified surgical protocol cannot fully meet the needs of all patients, resulting in poor postoperative efficacy and even complications in some patients.

[0004] In recent years, the rapid development of digital medical technology has made personalized medicine possible. In particular, the application of numerical simulation and artificial intelligence has laid the foundation for building a prediction model for corneal cross-linking surgery based on individual patient characteristics. However, current research focuses on standardized models based on average parameters, lacking personalized modeling and dynamic prediction systems based on real patient data. How to achieve personalized efficacy prediction of corneal cross-linking surgery by combining medical imaging has become a key issue that needs to be solved urgently.

[0005] It should be noted that the information disclosed in the above background technology section is only used for understanding the background of the present application, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the invention

[0006] The main purpose of the present invention is to overcome the defects existing in the above-mentioned background technology and provide a personalized prediction method for the efficacy of corneal cross-linking surgery.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] A personalized prediction method for the efficacy of corneal cross-linking surgery, comprising the following steps:

[0009] S1. Build a personalized 3D eyeball model: Obtain the patient's high-resolution OCT images and corneal topography, and build a personalized 3D eyeball model including the cornea and sclera through image segmentation, 3D reconstruction and data registration;

[0010] S2. Establishing corneal biomechanics and cross-linking mechanism model: Based on the three-dimensional eyeball model constructed in step S1 and combined with clinical data, establish corneal biomechanics material model, keratoconus progression model and corneal cross-linking mechanism model;

[0011] S3, constructing an eyeball finite element model: based on the three-dimensional eyeball model constructed in step S1 and the corneal biomechanical material model and keratoconus progression model established in step S2, meshing is performed and boundary conditions are determined to establish a finite element model of the eyeball;

[0012] S4. Perform surgery simulation and efficacy prediction: Based on the three-dimensional eyeball model constructed in step S1, the cross-linking mechanism model established in step S2 and the finite element model established in step S3, the corneal cross-linking surgery is simulated using the finite element method, and a real-time prediction system is built based on a neural network to generate postoperative efficacy prediction results.

[0013] Furthermore, step S1 specifically includes the following sub-steps:

[0014] S11: Obtain high-resolution OCT images of the patient's eyeball, remove speckle noise in the background, and unify the image background through mask processing; build a Resnet-18 neural network to identify the key points of the chamber angle in the OCT image, and use the central point of the cornea as the displacement reference point for registration;

[0015] S12: Use deep learning algorithms to segment the cornea, iris, and lens, extract geometric feature points on the front and back surfaces of the cornea, construct a spatial coordinate change matrix, and use point cloud reconstruction technology to generate a three-dimensional corneal model; based on corneal topography data, use an iterative optimization algorithm to finely align the OCT-reconstructed corneal data to obtain calibrated corneal point cloud data;

[0016] S13: Based on the sclera geometric parameters, a three-dimensional model of the patient's sclera is constructed, and the cornea and sclera are assembled to generate a complete three-dimensional eyeball geometric model.

[0017] Furthermore, step S2 specifically includes the following sub-steps:

[0018] S21. Establishing corneal material model: Based on the Neo-Hookean hyperelastic material model, a biomechanical material model of the cornea is established, in which the volume strain energy is a quadratic function and the material property is incompressible;

[0019] S22: Establish a keratoconus progression model: set a weak zone in the central area of ​​the cornea, and use a function that attenuates along the center to simulate the corneal intensity distribution of keratoconus patients to reflect the characteristics of the lesion area;

[0020] S23: Establish a corneal cross-linking mechanism model: set the corneal vertex elastic modulus increase to the maximum, and the elastic modulus increment is linearly related to the radiation intensity; the radiation intensity decays exponentially with the central axis distance at a fixed depth, and decays exponentially with the depth at a fixed central axis distance, to simulate the effect of cross-linking surgery on the mechanical properties of the cornea.

[0021] Furthermore, step S3 specifically includes the following sub-steps:

[0022] S31: The eyeball model is divided using a free tetrahedral mesh, where different maximum and minimum unit sizes are set for the sclera and cornea regions;

[0023] S32: Apply a fixed constraint to the optic nerve at the bottom of the sclera, apply a fixed constraint in the z direction to the axial boundary of the equatorial part of the sclera, and apply a uniform pressure load to the inner surface of the eyeball model to simulate the intraocular pressure under physiological conditions;

[0024] S33: Based on clinical data, the optimal proportional factor is iteratively solved through feature point sampling and error calculation to establish a zero-stress state model of the eyeball to improve the accuracy of the simulation.

[0025] Furthermore, in step S31, the maximum unit size of the sclera is 815 , the maximum corneal cell size is 150 , the minimum cell size is 8.15 .

[0026] Furthermore, in step S33, establishing the zero stress state model specifically includes the following sub-steps:

[0027] S331: sampling feature points of the initial eyeball model to obtain spatial position information of key points as a basis for subsequent error calculation;

[0028] S332: calculating the error between the feature points of the deformed model and the initial model by adjusting the scale factor of the model scaling; iteratively adjusting the scale factor to minimize the error and optimize the scale factor of the model;

[0029] S333: According to the error minimization principle, determine the optimal proportional factor, and establish a zero stress state model of the eyeball based on the proportional factor.

[0030] Furthermore, in step S4, the finite element method is used to analyze the stress distribution, deformation and biomechanical property changes after corneal cross-linking; a displacement cloud map of the corneal area before and after corneal cross-linking is generated to intuitively show the impact of the surgery on the mechanical properties of the cornea.

[0031] Furthermore, the method also includes: establishing a real-time interactive virtual surgery platform based on a personalized three-dimensional eyeball model and an efficacy prediction model, wherein the virtual surgery platform supports dynamic adjustment of surgical parameters, updates prediction results in real time, and provides doctors with a reference for preoperative planning and intraoperative adjustments.

[0032] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the personalized prediction method for the efficacy of corneal cross-linking surgery.

[0033] A computer program product comprises a computer program, wherein when the computer program is executed by a processor, the method for personalized prediction of the efficacy of corneal cross-linking surgery is implemented.

[0034] The present invention has the following beneficial effects:

[0035] The present invention provides a personalized prediction method for the efficacy of corneal cross-linking surgery. By combining high-resolution OCT images and corneal topography, a personalized three-dimensional eyeball model is constructed, and a corneal biomechanical material model, a keratoconus progression model and a corneal cross-linking mechanism model are established based on clinical data. The corneal cross-linking surgery is simulated using the finite element method and artificial neural network technology, a real-time prediction system is constructed, and a postoperative efficacy prediction result is generated, providing a scientific basis for preoperative planning and postoperative evaluation. The method can effectively solve the problem of unclear postoperative efficacy of corneal cross-linking surgery in clinical practice, especially the corneal stress of keratoconus patients after corneal cross-linking surgery can be calculated based on numerical simulation and artificial neural network technology, so as to achieve personalized real-time prediction of surgical efficacy, assist clinical preoperative patient popular science education, doctor treatment decision selection and telemedicine. Compared with the prior art, the present invention constructs a personalized eyeball model based on real patient data, combines finite element simulation with neural network technology, and establishes a systematic surgical prediction system, which provides data and platform support for the scientific nature of doctors' surgical decisions and patients' preoperative education, and significantly improves the accuracy and personalization of surgical planning.

[0036] Other beneficial effects of the embodiments of the present invention will be further described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a patient-personalized corneal model according to an embodiment of the present invention.

[0038] Figure 2 This is a three-dimensional eyeball assembly model according to an embodiment of the present invention.

[0039] Figure 3 This is a corneal intensity distribution diagram of a keratoconus patient according to an embodiment of the present invention.

[0040] Figure 4 It is the optimal scale factor iteration algorithm of the embodiment of the present invention.

[0041] Figure 5 This is a cloud map of the partial displacement of the cornea before cross-linking according to an embodiment of the present invention (average displacement 88.4585 μm).

[0042] Figure 6 This is a cloud map of partial displacement of the cornea after cross-linking according to an embodiment of the present invention (average displacement 19.42465 μm).

[0043] Figure 7 This is an overall flow chart of the personalized prediction method for corneal cross-linking surgery according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following is a detailed description of the embodiments of the present invention. It should be emphasized that the following description is only exemplary and is not intended to limit the scope and application of the present invention.

[0045] The present invention proposes a personalized prediction method for the efficacy of corneal cross-linking surgery, the main steps of which are as follows: 1) Based on optical coherence tomography (OCT) and corneal topography data, a three-dimensional model of the eyeball of a keratoconus patient is constructed, including the cornea and sclera; 2) Based on clinical data, a corneal material model, a keratoconus progression model, and a corneal cross-linking mechanism model are established; 3) A finite element model of the eyeball is constructed; 4) Based on the finite element method and artificial intelligence, the corneal cross-linking surgery is simulated and a prediction system platform for the corneal cross-linking surgery is constructed. The present invention can solve the problem of unclear postoperative efficacy of corneal cross-linking surgery in clinical practice, realize personalized real-time prediction of the efficacy of corneal cross-linking surgery, and assist clinical preoperative patient science education, doctor treatment decision selection, and telemedicine.

[0046] See also Figures 1 to 7 The embodiment of the present invention provides a personalized prediction method for the efficacy of corneal cross-linking surgery, comprising the following steps:

[0047] Step S1, constructing a personalized three-dimensional eyeball model: obtaining a patient's high-resolution OCT image and corneal topography, and constructing a personalized three-dimensional eyeball model including the cornea and sclera through image segmentation, three-dimensional reconstruction and data registration;

[0048] In a preferred embodiment, step S1 specifically includes the following sub-steps:

[0049] S11: Obtain high-resolution OCT images of the patient's eyeball, remove speckle noise in the background, and unify the image background through mask processing; build a Resnet-18 neural network to identify the key points of the chamber angle in the OCT image, and use the central point of the cornea as the displacement reference point for registration;

[0050] S12: Use deep learning algorithms to segment the cornea, iris, and lens, extract geometric feature points on the front and back surfaces of the cornea, construct a spatial coordinate change matrix, and use point cloud reconstruction technology to generate a three-dimensional corneal model; based on corneal topography data, use an iterative optimization algorithm to finely align the OCT-reconstructed corneal data to obtain calibrated corneal point cloud data;

[0051] S13: Based on the sclera geometric parameters, a three-dimensional model of the patient's sclera is constructed, and the cornea and sclera are assembled to generate a complete three-dimensional eyeball geometric model.

[0052] Step S2, establishing a corneal biomechanics and cross-linking mechanism model: Based on the three-dimensional eyeball model constructed in step S1 and combined with clinical data, a corneal biomechanical material model, a keratoconus progression model, and a corneal cross-linking mechanism model are established to simulate the mechanical properties of the cornea and the mechanism of action of the cross-linking surgery.

[0053] In a preferred embodiment, step S2 specifically includes the following sub-steps:

[0054] S21. Establishing corneal material model: Based on the Neo-Hookean hyperelastic material model, a biomechanical material model of the cornea is established, in which the volume strain energy is a quadratic function and the material property is incompressible;

[0055] S22: Establish a keratoconus progression model: set a weak zone in the central area of ​​the cornea, and use a function that attenuates along the center to simulate the corneal intensity distribution of keratoconus patients to reflect the characteristics of the lesion area;

[0056] S23: Establish a corneal cross-linking mechanism model: set the corneal vertex elastic modulus increase to the maximum, and the elastic modulus increment is linearly related to the radiation intensity; the radiation intensity decays exponentially with the central axis distance at a fixed depth, and decays exponentially with the depth at a fixed central axis distance, to simulate the effect of cross-linking surgery on the mechanical properties of the cornea.

[0057] Step S3, constructing a finite element model of the eyeball: based on the three-dimensional eyeball model constructed in step S1 and the corneal biomechanical material model and keratoconus progression model established in step S2, meshing is performed and boundary conditions are determined to establish a finite element model of the eyeball to provide a computational framework for simulation.

[0058] In a preferred embodiment, step S3 specifically includes the following sub-steps:

[0059] S31: The eyeball model is divided using a free tetrahedral mesh, where different maximum and minimum unit sizes are set for the sclera and cornea regions;

[0060] S32: Apply a fixed constraint to the optic nerve at the bottom of the sclera, apply a fixed constraint in the z direction to the axial boundary of the equatorial part of the sclera, and apply a uniform pressure load to the inner surface of the eyeball model to simulate the intraocular pressure under physiological conditions;

[0061] S33: Based on clinical data, the optimal proportional factor is iteratively solved through feature point sampling and error calculation to establish a zero-stress state model of the eyeball to improve the accuracy of the simulation.

[0062] In a more preferred embodiment, in step S31, the maximum unit size of the meshed sclera is 815 , the maximum corneal cell size is 150 , the minimum cell size is 8.15 .

[0063] In a more preferred embodiment, in step S33, establishing the zero stress state model specifically includes the following sub-steps:

[0064] S331: sampling feature points of the initial eyeball model to obtain spatial position information of key points as a basis for subsequent error calculation;

[0065] S332: calculating the error between the feature points of the deformed model and the initial model by adjusting the scale factor of the model scaling; iteratively adjusting the scale factor to minimize the error and optimize the scale factor of the model;

[0066] S333: According to the error minimization principle, determine the optimal proportional factor, and establish a zero stress state model of the eyeball based on the proportional factor.

[0067] Step S4, perform surgery simulation and efficacy prediction: Based on the three-dimensional eyeball model constructed in step S1, the cross-linking mechanism model established in step S2 and the finite element model established in step S3, the corneal cross-linking surgery is simulated using the finite element method, and a real-time prediction system is built based on a neural network to generate a postoperative efficacy prediction result.

[0068] In a preferred embodiment, in step S4, the finite element method is used to analyze the stress distribution, deformation and biomechanical property changes after corneal cross-linking; a displacement cloud map of the corneal area before and after corneal cross-linking is generated to intuitively show the impact of the surgery on the mechanical properties of the cornea.

[0069] In some embodiments, the personalized prediction method for the efficacy of corneal cross-linking surgery also includes: establishing a real-time interactive virtual surgery platform based on a personalized three-dimensional eyeball model and an efficacy prediction model, wherein the virtual surgery platform supports dynamic adjustment of surgical parameters, updates prediction results in real time, and provides doctors with a reference for preoperative planning and intraoperative adjustments.

[0070] In some embodiments, a personalized corneal cross-linking surgery efficacy prediction method comprises the following steps:

[0071] 1. Constructing a 3D model of the eye of a patient with keratoconus

[0072] Obtain high-resolution OCT images and corneal topography of the patient, reconstruct the 3D model of the cornea and sclera through image segmentation, 3D reconstruction, data registration and other technologies, and perform smoothing and assembly operations. Specifically, it includes: using the Unet neural network to segment and register the OCT image; using the height data in the corneal topography to correct and calibrate the 3D model.

[0073] 2. Establish corneal material model, keratoconus progression model and corneal cross-linking mechanism model

[0074] Based on clinical data, a corneal biomechanical model, a keratoconus progression model and a corneal cross-linking mechanism model were established, including: Corneal material model: the Neo-Hookean hyperelastic material model is used to describe the biomechanical properties of the cornea; Keratoconus progression model: a weak zone is set in the central area of ​​the cornea, and a function with attenuation along the center is used to simulate the corneal strength distribution of patients with keratoconus; Corneal cross-linking mechanism model: the increase in the elastic modulus of the corneal vertex (center point) is set to the maximum, and the elastic modulus increment is linearly related to the radiation intensity; the radiation intensity decays exponentially with depth and distance from the central axis.

[0075] 3. Constructing the finite element model of the eye

[0076] Based on the three-dimensional eyeball model, mesh division is performed, fixed constraints and boundary loads are determined, and a zero-stress state model of the eyeball is established, including: mesh division: the maximum unit size of the sclera is 815 μm, the maximum unit size of the cornea is 150 μm, and the minimum unit size is 8.15 μm; constraint and load application: a fixed constraint is applied to the optic nerve at the bottom of the sclera, a fixed constraint in the z direction is applied to the axial boundary of the equatorial part of the sclera, and a uniform pressure load of 2000 Pa is applied to the inner surface of the eyeball model to simulate the intraocular pressure in the physiological state; zero-stress state model: through feature point sampling and error calculation, the proportional factor is iteratively optimized to establish a zero-stress state model of the eyeball.

[0077] 4. Construction of corneal cross-linking surgery simulation and prediction system

[0078] Based on the finite element method and artificial intelligence technology, the corneal cross-linking surgery is simulated, and a real-time prediction system platform is constructed, which specifically includes: surgical simulation: using the finite element method to calculate the stress distribution, deformation and changes in biomechanical properties after corneal cross-linking surgery; prediction system: based on a personalized three-dimensional eyeball model and efficacy prediction model, a real-time interactive virtual surgery platform is developed to support dynamic adjustment of surgical parameters and real-time update of prediction results; efficacy prediction: based on the patient's preoperative data and personalized simulation results, a postoperative efficacy prediction report is generated to guide doctors in formulating the optimal surgical plan.

[0079] The specific embodiments of the present invention and algorithm examples thereof are further described below.

[0080] 1. Constructing a 3D model of the eyeball of a keratoconus patient

[0081] I. Data Acquisition and Preprocessing

[0082] Obtain high-resolution OCT (optical coherence tomography) images of the patient's eyeballs, use median filtering to remove speckle noise in the background, and then use a black mask to cover the yellow label values ​​in the image to unify the background. A Resnet-18 neural network was built to identify and detect the key points of the chamber angle in the OCT image, and the central point of the cornea (the highest point on the anterior surface of the cornea) was used as the displacement reference point for displacement registration.

[0083] II. 3D Reconstruction and Model Calibration

[0084] Deep learning algorithms (such as U-Net network) are used to segment key structures such as cornea, iris and lens; geometric feature points of the front and back surfaces of the cornea are extracted, and a spatial coordinate change matrix is ​​constructed. The three-dimensional corneal model is generated using point cloud reconstruction technology and Alpha shapes algorithm. The feature points of the front and back cornea in the corneal topography are extracted, and the change matrix is ​​calculated based on the feature points. The corneal data reconstructed by OCT is roughly aligned, and the iterative optimization algorithm is used to finely align it to obtain the corneal point cloud data after iterative alignment. Figure 1 Shown is a three-dimensional reconstructed personalized corneal model.

[0085] III. Construction of sclera and assembly of eyeball model

[0086] Based on the data in Table 1, a three-dimensional model of the patient's sclera was constructed, and the cornea and sclera were assembled using SolidWorks software to construct a three-dimensional geometric model of the patient's eyeball. Figure 2 Shown is the assembled three-dimensional eyeball model.

[0087] Table 1

[0088]

[0089] 2. Establishment of a mathematical model for corneal cross-linking

[0090] I. Corneal material model

[0091] As shown in Formula 1, the biomechanical material model of the cornea is established based on the Neo-Hookean hyperelastic material model, in which the volumetric strain energy is quadratic and the compression property is incompressible;

[0092] Formula 1

[0093] II. Keratoconus Progression Model

[0094] As shown in Formula 2 and Formula 3, a corneal material model of a keratoconus patient is established, wherein a weak zone is set in the central area of ​​the cornea and a function of attenuation along the center is set to simulate the corneal strength of the keratoconus patient.

[0095] Formula 2

[0096] Formula 3

[0097] Corneal intensity distribution in patients with keratoconus Figure 3 shown.

[0098] III. Model of corneal cross-linking mechanism

[0099] As shown in Formula 4-7, a corneal cross-linking surgery mechanism model was established.

[0100] Formula 4

[0101] Formula 5

[0102] Formula 6

[0103] Formula 7

[0104] Among them, the basic setting of the model is that the increase in elastic modulus of the corneal vertex (center point) is the largest, the elastic modulus increment is linearly related to the radiation intensity, at the same depth, the radiation intensity is exponentially attenuated with the distance from the central axis, and when the distance from the central axis is the same, the radiation intensity is exponentially attenuated with the depth.

[0105] 3. Constructing the finite element model of the eyeball

[0106] I. Meshing

[0107] Mesh type: free tetrahedron mesh, number of meshes: 149965, maximum cell size of sclera: 815 , Maximum corneal cell size: 150 , minimum unit size: 8.15 .

[0108] II. Applying Constraints and Loads

[0109] A fixed constraint was applied to the optic nerve at the bottom of the sclera, a fixed constraint was applied to the axial boundary of the equatorial part of the sclera in the z direction, and a uniform pressure load of 2000Pa was applied to the inner surface of the eyeball model to simulate the intraocular pressure in physiological state.

[0110] III. Zero stress state model establishment

[0111] The clinical data is a corneal expansion model under intraocular pressure. To improve the simulation accuracy, Figure 4 The algorithm solves the eyeball model under zero stress. First, the feature points of the initial model are sampled to obtain By formula The error under the proportional factor is calculated, and the optimal proportional factor corresponding to the minimum error is obtained by iterating the proportional factor.

[0112] 4. Construction of corneal cross-linking surgery simulation and prediction system

[0113] I. Corneal cross-linking surgery simulation

[0114] The surgical parameters (such as UV radiation intensity, duration, and range of action) were set as shown in Table 2, and the stress distribution, deformation, and biomechanical property changes after corneal cross-linking were analyzed using the finite element method; the corneal regional displacement cloud map before and after corneal cross-linking is shown in Figure 5 and Figure 6 shown.

[0115] Table 2

[0116]

[0117] II. Prediction System Construction

[0118] Based on the personalized three-dimensional eyeball model and efficacy prediction model, a real-time interactive virtual surgery platform is developed; the platform supports dynamic adjustment of surgical parameters and real-time update of prediction results, providing doctors with a reference for preoperative planning and intraoperative adjustments.

[0119] In summary, the present invention provides a personalized prediction method for the efficacy of corneal cross-linking surgery. Based on real patient data, combined with numerical simulation and artificial intelligence technology, it can achieve personalized real-time prediction of surgical efficacy, assist in preoperative patient science education, doctor treatment decision-making and telemedicine, and significantly improve the accuracy and personalization of surgical planning.

[0120] An embodiment of the present invention further provides a storage medium for storing a computer program, which at least performs the above method when executed.

[0121] An embodiment of the present invention further provides a control device, comprising a processor and a storage medium for storing a computer program; wherein the processor is configured to execute at least the method described above when executing the computer program.

[0122] An embodiment of the present invention further provides a processor, wherein the processor executes a computer program and at least executes the method described above.

[0123] The storage medium can be implemented by any type of non-volatile storage device, or a combination thereof. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a ferromagnetic random access memory (Flash Memory), a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The storage medium described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0124] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0125] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0126] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0127] A person skilled in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, etc. Various media that can store program codes.

[0128] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0129] The methods disclosed in the several method embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0130] The features disclosed in several product embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new product embodiments.

[0131] The features disclosed in several method or device embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0132] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art of the present invention, several equivalent substitutions or obvious variations can be made without departing from the concept of the present invention, and the performance or use is the same, which should be regarded as belonging to the protection scope of the present invention.

Claims

1. A personalized prediction method for the efficacy of corneal cross-linking surgery, characterized in that: The following steps are involved: S1. Build a personalized 3D eyeball model: Obtain the patient's high-resolution OCT images and corneal topography, and build a personalized 3D eyeball model including the cornea and sclera through image segmentation, 3D reconstruction and data registration; S2. Establishing corneal biomechanics and cross-linking mechanism model: Based on the three-dimensional eyeball model constructed in step S1 and combined with clinical data, establish corneal biomechanics material model, keratoconus progression model and corneal cross-linking mechanism model; S3, constructing an eyeball finite element model: based on the three-dimensional eyeball model constructed in step S1 and the corneal biomechanical material model and keratoconus progression model established in step S2, meshing is performed and boundary conditions are determined to establish a finite element model of the eyeball; S4, performing surgery simulation and efficacy prediction: based on the three-dimensional eyeball model constructed in step S1, the cross-linking mechanism model established in step S2 and the finite element model established in step S3, the corneal cross-linking surgery is simulated using the finite element method, and a real-time prediction system is constructed based on a neural network to generate a postoperative efficacy prediction result; Step S2 specifically includes the following sub-steps: S21. Establishing corneal material model: Based on the Neo-Hookean hyperelastic material model, a biomechanical material model of the cornea is established, in which the volume strain energy is a quadratic function and the material property is incompressible; S22: Establish a keratoconus progression model: set a weak zone in the central area of ​​the cornea, and use a function that attenuates along the center to simulate the corneal intensity distribution of keratoconus patients to reflect the characteristics of the lesion area; S23: Establish a corneal cross-linking mechanism model: set the corneal vertex elastic modulus increase to the maximum, and the elastic modulus increment is linearly related to the radiation intensity; the radiation intensity decays exponentially with the central axis distance at a fixed depth, and decays exponentially with the depth at a fixed central axis distance, to simulate the effect of cross-linking surgery on the mechanical properties of the cornea.

2. The personalized prediction method for the efficacy of corneal cross-linking surgery according to claim 1, characterized in that: Step S1 specifically includes the following sub-steps: S11: Obtain high-resolution OCT images of the patient's eyeball, remove speckle noise in the background, and unify the image background through mask processing; build a Resnet-18 neural network to identify the key points of the chamber angle in the OCT image, and use the central point of the cornea as the displacement reference point for registration; S12: Use deep learning algorithms to segment the cornea, iris, and lens, extract geometric feature points on the front and back surfaces of the cornea, construct a spatial coordinate change matrix, and use point cloud reconstruction technology to generate a three-dimensional corneal model; based on corneal topography data, use an iterative optimization algorithm to finely align the OCT-reconstructed corneal data to obtain calibrated corneal point cloud data; S13: Based on the sclera geometric parameters, a three-dimensional model of the patient's sclera is constructed, and the cornea and sclera are assembled to generate a complete three-dimensional eyeball geometric model.

3. The personalized prediction method for the efficacy of corneal cross-linking surgery according to any one of claims 1 to 2, characterized in that: Step S3 specifically includes the following sub-steps: S31: The eyeball model is divided using a free tetrahedral mesh, where different maximum and minimum unit sizes are set for the sclera and cornea regions; S32: Apply a fixed constraint to the optic nerve at the bottom of the sclera, apply a fixed constraint in the z direction to the axial boundary of the equatorial part of the sclera, and apply a uniform pressure load to the inner surface of the eyeball model to simulate the intraocular pressure under physiological conditions; S33: Based on clinical data, the optimal proportional factor is iteratively solved through feature point sampling and error calculation to establish a zero-stress state model of the eyeball to improve the accuracy of the simulation.

4. The personalized prediction method for the efficacy of corneal cross-linking surgery according to claim 3, characterized in that: In step S31, the maximum unit size of the sclera is 815 , the maximum corneal cell size is 150 .

5. The personalized prediction method for the efficacy of corneal cross-linking surgery according to claim 3, characterized in that: In step S33, establishing the zero stress state model specifically includes the following sub-steps: S331: sampling feature points of the initial eyeball model to obtain spatial position information of key points as a basis for subsequent error calculation; S332: calculating the error between the feature points of the deformed model and the initial model by adjusting the scale factor of the model scaling; iteratively adjusting the scale factor to minimize the error and optimize the scale factor of the model; S333: According to the error minimization principle, determine the optimal proportional factor, and establish a zero stress state model of the eyeball based on the proportional factor.

6. The personalized prediction method for the efficacy of corneal cross-linking surgery according to any one of claims 1 to 2, characterized in that: In step S4, the stress distribution, deformation and biomechanical property changes after corneal cross-linking are analyzed using the finite element method; a displacement cloud map of the corneal area before and after corneal cross-linking is generated to intuitively demonstrate the impact of the surgery on the mechanical properties of the cornea.

7. The personalized prediction method for the efficacy of corneal cross-linking surgery according to any one of claims 1 to 2, characterized in that: Also includes: Based on the personalized three-dimensional eyeball model and the efficacy prediction model, a real-time interactive virtual surgery platform is established. The virtual surgery platform supports dynamic adjustment of surgical parameters and real-time update of prediction results, providing doctors with a reference for preoperative planning and intraoperative adjustment.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for personalized prediction of the efficacy of corneal cross-linking surgery as described in any one of claims 1 to 7 is implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for personalized prediction of the efficacy of corneal cross-linking surgery as described in any one of claims 1 to 7 is implemented.

Images