Postoperative lens capsular bag parameter prediction method, system, medium and equipment

Abstract

The invention discloses a postoperative lens capsular bag parameter prediction method, system, medium and equipment, and belongs to the field of biomedicines.The method comprises the steps that pre-operative anterior segment OCT images are obtained, and lens contour line parameters corresponding to multiple two-dimensional lens images with different axial angles in the anterior segment OCT images are extracted; according to each crystalline lens contour line parameter, constructing a crystalline lens three-dimensional model; performing three-dimensional Delaunay triangulation on the crystalline lens three-dimensional model to obtain each tetrahedral mesh; calculating the surface area of the crystalline lens three-dimensional model according to the area of the exposed triangular patch of each tetrahedral mesh; and taking half of the surface area as the posterior capsular membrane area of the postoperative crystalline lens, and calculating and outputting the simulated diameter of the capsular bag of the postoperative crystalline lens. Therefore, by implementing the method, the problem that the postoperative lens capsular bag diameter cannot be simply and effectively predicted and calculated in the prior art can be solved.

Description

Technical Field

[0001] This invention belongs to the field of biomedicine and relates to a method, system, medium and device for predicting parameters of the lens capsule after surgery. Background Technology

[0002] Measuring lens volume and capsular surface area plays a crucial role in the diagnosis and mechanistic study of eye diseases such as cataracts and glaucoma, as well as in the selection of intraocular implant sizes, including intraocular lenses (IOLs) and capsular tension rings. Early IOL size design primarily relied on anatomical data from ex vivo lenses combined with ocular bioparameters such as axial length for optimization. However, due to individual differences in lens and capsular size among patients, in vivo measurements are necessary for precise selection of personalized intraocular implant sizes. Traditional methods for in vivo measurement of lens and capsular size include three-dimensional magnetic resonance imaging (3D-MRI), capsular measurement rings, ultrasound biomicroscopy, and anterior segment optical coherence tomography (OCT). Among these, 3D-MRI is widely used for quantifying eyeball volume and shape. Capsular measurement rings require implantation of the capsular bag during cataract surgery, with the capsular diameter measured post-operatively under a slit lamp. Ultrasound biomicroscopy uses the principle of ultrasonic wave reflection to image superficial ocular tissues to observe the lens and its surrounding structures. Anterior segment OCT estimates lens volume using a simplified ellipsoidal model.

[0003] However, each of these methods has its limitations. 3D-MRI suffers from long detection times, high costs, unclear lens imaging, and is easily affected by surrounding tissues, leading to large measurement errors. Capsular bag measurement rings can only measure the diameter of the capsular bag after cataract surgery and require additional surgical implantation, increasing risks and costs. Ultrasonic biomicroscopy suffers from poor repeatability and penetration, unsatisfactory image quality, large errors in quantitative indicators, and the risk of eye infection due to contact operation. Anterior segment OCT suffers from poor accuracy in volume calculation because the ellipsoidal model used does not match the actual irregular biconvex or ellipsoidal shape of the lens, and there is a lack of effective methods for calculating the lens surface area and postoperative capsular bag diameter. Summary of the Invention

[0004] This application provides a method, system, medium, and device for predicting parameters of the postoperative lens capsule, which can solve the problem of the lack of a simple and effective method for predicting and calculating the diameter of the postoperative lens capsule in the prior art.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a method for predicting parameters of the postoperative lens capsule, comprising:

[0006] Obtain preoperative anterior segment OCT images and extract lens contour parameters corresponding to multiple two-dimensional lens images at different axial angles from each anterior segment OCT image;

[0007] Based on the lens contour parameters described above, a three-dimensional model of the lens is constructed.

[0008] The three-dimensional model of the lens is subjected to three-dimensional Delaunay triangulation to obtain each tetrahedral mesh;

[0009] The surface area of ​​the three-dimensional model of the lens is calculated based on the area of ​​the exposed triangular facets of each tetrahedral mesh.

[0010] Half of the surface area is taken as the posterior capsule area of ​​the lens after surgery, and the simulated diameter of the capsular bag of the lens after surgery is calculated and output.

[0011] Compared with existing technologies, the embodiments of this application have the following beneficial effects: By acquiring anterior segment OCT images and lens contour parameters from multiple axial angles, multi-dimensional morphological data is provided for three-dimensional modeling, avoiding the one-sidedness problem caused by single-axis or limited-axis measurements; a three-dimensional lens model is constructed based on the contour parameters, realizing a three-dimensional representation of the lens morphology, which is closer to the real anatomical structure than two-dimensional images; a three-dimensional Delaunay triangulation is performed on the three-dimensional model to obtain a tetrahedral mesh, discretizing the complex curved surface into computable geometric units, providing a foundation for the accurate calculation of the subsequent surface area; the total surface area of ​​the lens is calculated by accumulating the areas of exposed triangular facets, directly performing numerical summation based on the geometric units of the model surface, avoiding the problems caused by traditional approximate fitting or two-dimensional projection substitution. The error is considered; considering that the effective surface area of ​​the capsular bag after surgery is determined only by the preserved posterior capsule and anterior capsule remnants, but the role of the anterior capsule remnant is "covered" by the optical surface of the intraocular lens, the actual effective support area mainly depends on the posterior capsule. Half of the total surface area of ​​the lens is taken as the posterior capsule area. This design is based on the biconvex symmetry of the lens anatomical characteristics (the anterior and posterior surface areas are approximately equal), making the derivation of the posterior capsule area more consistent with the actual range of the membranous structure preserved after surgery, and solving the problem of the lack of a posterior capsule area calculation method in the existing technology; furthermore, the simulated diameter of the capsular bag is output by calculating the posterior capsule area, transforming the abstract area parameter into a linear index that can be directly referenced clinically, making up for the shortcomings of the existing technology that only relies on the transverse diameter of the cornea or the preoperative lens diameter under a regular elliptical model for size judgment. The overall solution integrates the entire process from multi-source data acquisition to 3D modeling, subdivision calculation, and parameter derivation. For the first time, it has achieved quantitative calculation of the posterior capsule area and lens surface area. It specifically addresses the lack of lens or posterior capsule area calculation methods and inaccurate capsular parameter prediction in existing technologies. It can provide clinicians with a more realistic simulated capsular diameter that closely matches the actual anatomical structure, guide the selection of implant size within the capsular bag, improve the matching degree between the artificial lens and the capsular bag, and reduce the risk of postoperative dislocation or refractive abnormalities.

[0012] In some embodiments of the first aspect of this application, constructing a three-dimensional model of the lens based on the lens contour parameters includes:

[0013] The parameters of each lens contour line are sampled to obtain the coordinates of each sampling point on the lens contour curve;

[0014] Based on the relative angles between the axial angles, the coordinates of the sampling points are integrated to construct three-dimensional point cloud data and output a three-dimensional model of the lens.

[0015] Compared with existing technologies, the above embodiments have the following beneficial effects: Sampling the contour parameters to obtain the coordinates of each sampling point on the lens contour curve, and discretizing the continuous contour curve into specific point coordinates, allows the contour shape to be presented in a calculable geometric data form, avoiding the problem of missing 3D modeling data caused by relying solely on the original contour parameters (such as continuous curve equations); Three-dimensional point cloud data is constructed by integrating the sampling point coordinates based on the relative angles between each axis, and the unified calibration of the spatial coordinate system ensures the positional correlation of contour points measured in different axes in three-dimensional space, solving the problem of point cloud misalignment caused by angle differences in multi-axis data, and providing a data foundation for the geometric consistency of subsequent 3D models.

[0016] In some embodiments of the first aspect of this application, the step of integrating the coordinates of each sampling point based on the relative angles between each of the axial angles, constructing three-dimensional point cloud data, and outputting a three-dimensional model of the lens includes:

[0017] Based on the relative angles between the axial angles, the coordinates of the sampling points are integrated to construct the original three-dimensional point cloud data.

[0018] In the original three-dimensional point cloud data, the irregular edge discrete points at the equator of the lens are fitted according to the Fourier series algorithm, and the sparse point cloud data of the lens is fitted according to the Zernike polynomial algorithm to output the three-dimensional model of the lens.

[0019] Compared with existing technologies, the above embodiments have the following beneficial effects: integrating the coordinates of sampling points at various axial angles to construct original three-dimensional point cloud data preserves the original information of multi-axial measurements; using the Fourier series algorithm to fit the discrete points of irregular edges at the lens equator corrects the noise error of the discrete points and improves the smoothness and continuity of the equatorial contour; and using the Zernike polynomial algorithm to fit the sparse point cloud data of the lens completes the geometric information of the missing point cloud areas, enhancing the integrity and geometric accuracy of the three-dimensional model.

[0020] In some embodiments of the first aspect of this application, the step of performing three-dimensional Delaunay triangulation on the three-dimensional model of the lens to obtain tetrahedral meshes includes:

[0021] Based on the Alpha shape algorithm, the three-dimensional model of the lens is subjected to three-dimensional Delaunay triangulation to generate each original tetrahedral mesh.

[0022] Select the original tetrahedral meshes whose circumscribed sphere radius is less than or equal to a preset threshold as the original tetrahedral meshes.

[0023] Compared with existing technologies, the above embodiments have the following advantages: the Alpha shape algorithm is used to perform three-dimensional Delaunay triangulation on the three-dimensional model of the lens, and the generated original tetrahedral mesh can better fit the topological structure of the lens surface, avoiding the over-segmentation or under-segmentation problems of traditional segmentation methods; the original tetrahedral mesh with a circumscribed sphere radius less than or equal to a preset threshold is selected as an effective mesh, eliminating redundant or abnormal meshes on the model surface, ensuring that only geometric units reflecting the true surface of the lens are retained, thus providing a guarantee for the accuracy of subsequent surface area calculation.

[0024] In some embodiments of the first aspect of this application, calculating the surface area of ​​the three-dimensional model of the lens based on the area of ​​the exposed triangular facets of each of the tetrahedral meshes includes:

[0025] The surface area of ​​the three-dimensional model of the lens is obtained by summing the areas of each of the exposed triangular facets.

[0026] Compared with the prior art, the above embodiments have the following beneficial effects: by summing the areas of each exposed triangular facet to calculate the surface area of ​​the three-dimensional model of the lens, the numerical summation is performed directly based on the geometric units of the model surface, avoiding the errors caused by indirect estimation or empirical formulas, and ensuring the authenticity and traceability of the surface area value.

[0027] In some embodiments of the first aspect of this application, the step of using half of the surface area as the posterior capsule area of ​​the lens after surgery and calculating and outputting the simulated diameter of the capsular bag of the lens after surgery includes:

[0028] The posterior capsule is considered to be roughly circular, and based on the geometric relationship between the area and diameter of the roughly circular shape, the diameter is calculated according to the area of ​​the posterior capsule, which is used as the simulated diameter of the capsule.

[0029] Compared with existing technologies, the above embodiments have the following beneficial effects: by considering the posterior capsule as a near-circular shape and calculating the simulated diameter of the capsule based on the geometric relationship between the area and diameter of the near-circular shape, the near-circular shape is used as a reasonable approximation of the expansion morphology of the posterior capsule (the posterior capsule approaches a circle after natural postoperative expansion). The abstract area parameter is transformed into a linear indicator (diameter) that can be directly referenced clinically, which simplifies the practicality of parameter output. In clinical practice, it is possible to more intuitively select the matching capsule implant size based on the capsule diameter, thereby improving the efficiency and accuracy of preoperative planning.

[0030] In a second aspect, the present invention also provides a parameter prediction system for postoperative lens capsule, comprising: a data acquisition module, a model construction module, a model segmentation module, an area calculation module, and a result output module;

[0031] The data acquisition module is used to acquire preoperative anterior segment OCT images and extract lens contour parameters corresponding to multiple two-dimensional lens images with different axial angles in each anterior segment OCT image.

[0032] The model building module is used to build a three-dimensional model of the lens based on the contour parameters of each lens.

[0033] The model subdivision module is used to perform three-dimensional Delaunay triangulation on the three-dimensional model of the lens to obtain each tetrahedral mesh.

[0034] The area calculation module is used to calculate the surface area of ​​the three-dimensional model of the lens based on the area of ​​the exposed triangular facets of each tetrahedral mesh.

[0035] The result output module is used to take half of the surface area as the posterior capsule area of ​​the lens after surgery, and to calculate and output the simulated diameter of the capsular bag of the lens after surgery.

[0036] Compared with existing technologies, the above embodiments of this application have the following beneficial effects: By acquiring anterior segment OCT images and lens contour parameters from multiple axial angles, multi-dimensional morphological data is provided for three-dimensional modeling, avoiding the one-sidedness problem caused by single-axis or limited-axis measurements; a three-dimensional lens model is constructed based on the contour parameters, realizing a three-dimensional representation of the lens morphology, which is closer to the real anatomical structure than two-dimensional images; a three-dimensional Delaunay triangulation is performed on the three-dimensional model to obtain a tetrahedral mesh, discretizing the complex curved surface into computable geometric units, providing a basis for the accurate calculation of the subsequent surface area; the total surface area of ​​the lens is calculated by accumulating the areas of exposed triangular facets, directly performing numerical summation based on the geometric units of the model surface, avoiding traditional approximate fitting or two-dimensional projection substitution. To mitigate errors, considering that the effective surface area of ​​the capsular bag after surgery is determined solely by the retained posterior and anterior capsular remnants, but the role of the anterior capsular remnant is "covered" by the optical surface of the intraocular lens, the actual effective support area mainly depends on the posterior capsular bag. Using half of the total lens surface area as the posterior capsular area is based on the biconvex symmetry of the lens (the anterior and posterior surface areas are approximately equal), making the derivation of the posterior capsular area more consistent with the actual range of the membranous structure retained after surgery. This solves the problem of the lack of a posterior capsular area calculation method in existing technologies. Furthermore, by calculating the posterior capsular area, the simulated diameter of the capsular bag is output, transforming the abstract area parameter into a linear indicator that can be directly referenced clinically. This compensates for the shortcomings of existing technologies that rely solely on the corneal transverse diameter or the preoperative lens diameter under a regular elliptical model for size judgment. The overall solution integrates the entire process from multi-source data acquisition to 3D modeling, subdivision calculation, and parameter derivation. For the first time, it has achieved quantitative calculation of the posterior capsule area and lens surface area. It specifically addresses the lack of lens or posterior capsule area calculation methods and inaccurate capsular parameter prediction in existing technologies. It can provide clinicians with a more realistic simulated capsular diameter that closely matches the actual anatomical structure, guide the selection of implant size within the capsular bag, improve the matching degree between the artificial lens and the capsular bag, and reduce the risk of postoperative dislocation or refractive abnormalities.

[0037] In some embodiments of the second aspect of this application, the model building module includes: a sampling unit and an integration unit;

[0038] The sampling unit is used to sample the parameters of each lens contour line to obtain the coordinates of each sampling point on the lens contour curve.

[0039] The integration unit is used to integrate the coordinates of each sampling point according to the relative angles between each axial angle, construct three-dimensional point cloud data, and output a three-dimensional model of the lens.

[0040] Compared with existing technologies, the above embodiments have the following beneficial effects: Sampling the contour parameters to obtain the coordinates of each sampling point on the lens contour curve, and discretizing the continuous contour curve into specific point coordinates, allows the contour shape to be presented in a calculable geometric data form, avoiding the problem of missing 3D modeling data caused by relying solely on the original contour parameters (such as continuous curve equations); Three-dimensional point cloud data is constructed by integrating the sampling point coordinates based on the relative angles between each axis, and the unified calibration of the spatial coordinate system ensures the positional correlation of contour points measured in different axes in three-dimensional space, solving the problem of point cloud misalignment caused by angle differences in multi-axis data, and providing a data foundation for the geometric consistency of subsequent 3D models.

[0041] Thirdly, the present invention also provides a parameter prediction device for postoperative lens capsule, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is loaded onto the processor, it implements the steps of any one of the postoperative lens capsule parameter prediction methods of the present invention.

[0042] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the postoperative lens capsule parameter prediction methods of the present invention. Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating a method for predicting parameters of the lens capsule after surgery, provided in some embodiments of the present invention.

[0044] Figure 2 This is a schematic diagram of the structure of a postoperative lens capsule parameter prediction system provided in some embodiments of the present invention.

[0045] Figure 3 : This is a structural diagram of a postoperative lens capsule parameter prediction device provided in some embodiments of the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and 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.

[0047] Example 1:

[0048] Please refer to Figure 1To address the lack of a simple and effective method for predicting and calculating the diameter of the lens capsule after surgery in existing technologies, an embodiment of the present invention provides a method for predicting parameters of the lens capsule after surgery, comprising steps S1 to S5:

[0049] Step S1: Obtain preoperative anterior segment OCT images and extract lens contour parameters corresponding to multiple two-dimensional lens images with different axial angles from each anterior segment OCT image.

[0050] In practice, the anterior segment OCT images are obtained by examining the subject using an OCT device. To ensure image quality, the subject's eye is first instilled with compound tropicamide three times, with an interval of 5-10 minutes between each instillation. Then, with the pupil fully dilated and fixation good (the pupil diameter should be at least greater than 6 mm), the anterior segment OCT examination is performed. After ensuring that the subject's eye does not have any special types of lens disease (such as anterior conic lens disease), the image detection quality is qualified, and the lens outline is clearly identified, the outline parameters of the two-dimensional lens image are extracted.

[0051] Furthermore, most commercially available anterior segment OCT devices include contour recognition and analysis functions for two-dimensional lens images, allowing direct export of lens contour parameters from various angles (i.e., different axial angles). For example, the Tomey anterior segment OCT (model Casia2) provides 16 two-dimensional lens images along different axes in a single scan. Since the two-dimensional lens image approximates a double-arc shape, Casia2 can automatically quantify the surface curvature of the anterior arc, the surface curvature of the posterior arc, lens thickness, equatorial diameter, and the values ​​of eccentricity and tilt relative to the reference axis (some instruments name these tilt / eccentricity angle and eccentricity distance).

[0052] Step S2: Construct a three-dimensional model of the lens based on the lens contour parameters described above.

[0053] Furthermore, step S2 can be implemented through the following preferred embodiments, including steps S21-S22, as follows:

[0054] S21: Sample the parameters of each lens contour line to obtain the coordinates of each sampling point on the lens contour curve;

[0055] S22: Based on the relative angles between the axial angles, integrate the coordinates of the sampling points to construct three-dimensional point cloud data and output a three-dimensional model of the lens.

[0056] In this preferred embodiment, the contour parameters are sampled to obtain the coordinates of each sampling point on the lens contour curve. By discretizing the continuous contour curve into specific point coordinates, the contour shape can be presented in the form of computable geometric data, avoiding the problem of missing three-dimensional modeling data caused by relying solely on the original contour parameters (such as the equation of a continuous curve). The coordinates of the sampling points are integrated according to the relative angles between each axis to construct three-dimensional point cloud data. Through the unified calibration of the spatial coordinate system, the positional correlation of contour points measured in different axes in three-dimensional space is ensured, solving the problem of point cloud misalignment caused by angle differences in multi-axis data, and providing a data foundation for the geometric consistency of the subsequent three-dimensional model.

[0057] Furthermore, step S22 can be implemented through the following preferred embodiments, including steps S221-S222, as follows:

[0058] S221: Based on the relative angles between the axial angles, integrate the coordinates of the sampling points to construct the original three-dimensional point cloud data;

[0059] S222: In the original three-dimensional point cloud data, the irregular edge discrete points at the equator of the lens are fitted according to the Fourier series algorithm, and the sparse point cloud data of the lens is fitted according to the Zernike polynomial algorithm to output the three-dimensional model of the lens.

[0060] In this preferred embodiment, the original three-dimensional point cloud data is constructed by integrating the coordinates of sampling points at various axial angles, preserving the original information of multi-axial measurements. The irregular edge discrete points at the lens equator are fitted using the Fourier series algorithm to correct noise errors in the discrete points and improve the smoothness and continuity of the equatorial contour. The sparse point cloud data of the lens is fitted using the Zernike polynomial algorithm to complete the geometric information of the missing point cloud areas, thereby enhancing the integrity and geometric accuracy of the three-dimensional model.

[0061] In practice, after acquiring the lens contour parameters, the coordinates of sampling points on the arcs are generated through uniform sampling based on the double-arc parameters corresponding to each image. Then, based on the relative angles of the 16 sets of data provided by the instrument, the coordinates of the corresponding sampling points are integrated to obtain preliminary 3D point cloud data. Next, the Fourier series fitting method is used to supplement the data at the lens equator, which can improve the fitting accuracy for irregular edges. Finally, the Zernike polynomial fitting method is used to further supplement the point cloud data, completing the 3D reconstruction of the lens, thus obtaining the 3D model of the lens.

[0062] Furthermore, when constructing the three-dimensional model, this application combines Fourier series and Zernike polynomials for data fitting. The resulting three-dimensional model is closer to the real lens than the ellipsoidal model obtained by the traditional method. After applying the model to approximately 100 eyes, the root mean square of the fitting residual of the Zernike model is approximately 0.03 mm, while that of the ellipsoidal model is approximately 0.06 mm.

[0063] Step S3: Perform three-dimensional Delaunay triangulation on the three-dimensional model of the lens to obtain each tetrahedral mesh.

[0064] Furthermore, step S3 can be implemented through the following preferred embodiments, including steps S31-S32, as follows:

[0065] S31: According to the Alpha shape algorithm, the three-dimensional model of the lens is subjected to three-dimensional Delaunay triangulation to generate each original tetrahedral mesh;

[0066] S32: Select the original tetrahedral meshes whose circumscribed sphere radius is less than or equal to a preset threshold as the original tetrahedral meshes.

[0067] In this preferred embodiment, the Alpha shape algorithm is used to perform three-dimensional Delaunay triangulation on the three-dimensional model of the lens. The generated original tetrahedral mesh can better fit the topological structure of the lens surface, avoiding the over-segmentation or under-segmentation problems of traditional segmentation methods. The original tetrahedral mesh with a circumscribed sphere radius less than or equal to a preset threshold is selected as the effective mesh, eliminating redundant or abnormal meshes on the model surface, ensuring that only geometric units reflecting the true surface of the lens are retained, thus providing a guarantee for the accuracy of subsequent surface area calculation.

[0068] Step S4: Calculate the surface area of ​​the three-dimensional model of the lens based on the area of ​​the exposed triangular facets of each tetrahedral mesh.

[0069] Furthermore, step S4 can be implemented through the following preferred embodiments, as detailed below:

[0070] The surface area of ​​the three-dimensional model of the lens is obtained by summing the areas of each of the exposed triangular facets.

[0071] In this preferred embodiment, the surface area of ​​the three-dimensional lens model is calculated by summing the areas of each exposed triangular facet. The numerical summation is performed directly based on the geometric units of the model surface, avoiding errors caused by indirect estimation or empirical formulas, and ensuring the authenticity and traceability of the surface area value.

[0072] Step S5: Take half of the surface area as the posterior capsule area of ​​the lens after surgery, and calculate and output the simulated diameter of the capsular bag of the lens after surgery.

[0073] Furthermore, step S5 can be implemented through the following preferred embodiments, as follows:

[0074] The posterior capsule is considered to be roughly circular, and based on the geometric relationship between the area and diameter of the roughly circular shape, the diameter is calculated according to the area of ​​the posterior capsule, which is used as the simulated diameter of the capsule.

[0075] In this preferred embodiment, the posterior capsule is considered to be roughly circular, and the simulated diameter of the capsule is calculated based on the geometric relationship between the area and diameter of the roughly circular shape. This utilizes the reasonable approximation of the expansion morphology of the posterior capsule to a roughly circular shape (the posterior capsule naturally expands after surgery), transforming the abstract area parameter into a linear indicator (diameter) that can be directly referenced clinically. This simplifies the practicality of the parameter output and allows for a more intuitive selection of the matching capsule implant size based on the capsule diameter in clinical practice, improving the efficiency and accuracy of preoperative planning.

[0076] In practical implementation, after obtaining the 3D model, it is necessary to quantify its parameters. This solution uses the Alphashape algorithm to calculate the volume and surface area of ​​the 3D model: First, the model is triangulated in 3D using Delaunay to generate a tetrahedral mesh; then, for each tetrahedron, its circumscribed sphere radius is checked to see if it is less than or equal to a preset threshold, and only if the condition is met is the tetrahedron retained; after filtering, the volume of the lens 3D model is the sum of the volumes of all the filtered and retained tetrahedrons, while the surface area is the sum of the areas of all exposed triangular facets.

[0077] Furthermore, in actual surgery, after the intraocular lens (IOL) is implanted, the capsular bag (excluding the optical surface) adheres to each other, and the capsulotomy opening is only slightly smaller than the optical surface, whose thickness is usually less than 1 mm. Therefore, the surface area of ​​the lens calculated before surgery can be considered twice the postoperative posterior capsule area, especially in cases involving capsular tension rings where the capsular shape is nearly circular and remains expanded. In such cases, the postoperative capsular diameter can be predicted and calculated using the lens surface area, as described in step S5 above. As a supplement, the simulated diameter can be corrected based on the model and power of the IOL used in the specific surgery. By calculating the preoperative surface area of ​​the lens to estimate the capsular size after IOL implantation, this parameter can be used clinically to guide the selection of implant size within the capsular bag. Moreover, in practical applications, only anterior segment OCT equipment is needed, and the corresponding parameters can be directly obtained by incorporating the algorithm of this scheme, demonstrating significant clinical translational value.

[0078] In summary, compared with the prior art, the above embodiments of this application have the following beneficial effects: by acquiring anterior segment OCT images and lens contour parameters at multiple axial angles, multi-dimensional morphological data is provided for three-dimensional modeling, avoiding the one-sidedness problem caused by single-axis or limited-axis measurements; a three-dimensional lens model is constructed based on the contour parameters, realizing a three-dimensional representation of the lens morphology, which is closer to the real anatomical structure than two-dimensional images; a three-dimensional Delaunay triangulation is performed on the three-dimensional model to obtain a tetrahedral mesh, discretizing the complex curved surface into computable geometric units, providing a basis for the accurate calculation of the subsequent surface area; the total surface area of ​​the lens is calculated by accumulating the areas of exposed triangular facets, directly performing numerical summation based on the geometric units of the model surface, avoiding traditional approximate fitting or two-dimensional projection substitution. This approach mitigates errors introduced by the previous method. Considering that the effective surface area of ​​the capsular bag after surgery is determined solely by the retained posterior and anterior capsular remnants, but the role of the anterior capsular remnant is "covered" by the optical surface of the intraocular lens, the actual effective support area primarily depends on the posterior capsular. Using half of the total lens surface area as the posterior capsular area is based on the lens's biconvex symmetry (approximately equal anterior and posterior surface areas), making the derivation of the posterior capsular area more consistent with the actual range of the retained membranous structure after surgery. This solves the problem of the lack of a posterior capsular area calculation method in existing technologies. Furthermore, by calculating the posterior capsular area, the simulated diameter of the capsular bag is output, transforming the abstract area parameter into a linear indicator that can be directly referenced clinically. This compensates for the shortcomings of existing technologies that rely solely on the corneal transverse diameter or the preoperative lens diameter under a regular elliptical model for size judgment. The overall solution integrates the entire process from multi-source data acquisition to 3D modeling, subdivision calculation, and parameter derivation. For the first time, it has achieved quantitative calculation of the posterior capsule area and lens surface area. It specifically addresses the lack of lens or posterior capsule area calculation methods and inaccurate capsular parameter prediction in existing technologies. It can provide clinicians with a more realistic simulated capsular diameter that closely matches the actual anatomical structure, guide the selection of implant size within the capsular bag, improve the matching degree between the artificial lens and the capsular bag, and reduce the risk of postoperative dislocation or refractive abnormalities.

[0079] Example 2:

[0080] Please refer to Figure 2 Based on the same inventive concept, the present invention discloses a parameter prediction system for postoperative lens capsule, comprising: a data acquisition module M1, a model construction module M2, a model segmentation module M3, an area calculation module M4, and a result output module M5.

[0081] The data acquisition module M1 is used to acquire preoperative anterior segment OCT images and extract lens contour parameters corresponding to multiple two-dimensional lens images with different axial angles in each anterior segment OCT image.

[0082] The model building module M2 is used to build a three-dimensional model of the lens based on the contour parameters of each lens.

[0083] Furthermore, the model building module M2 includes: a sampling unit and an integration unit;

[0084] The sampling unit is used to sample the parameters of each lens contour line to obtain the coordinates of each sampling point on the lens contour curve.

[0085] The integration unit is used to integrate the coordinates of each sampling point according to the relative angles between each axial angle, construct three-dimensional point cloud data, and output a three-dimensional model of the lens.

[0086] In this preferred embodiment, the contour parameters are sampled to obtain the coordinates of each sampling point on the lens contour curve. By discretizing the continuous contour curve into specific point coordinates, the contour shape can be presented in the form of computable geometric data, avoiding the problem of missing three-dimensional modeling data caused by relying solely on the original contour parameters (such as the equation of a continuous curve). The coordinates of the sampling points are integrated according to the relative angles between each axis to construct three-dimensional point cloud data. Through the unified calibration of the spatial coordinate system, the positional correlation of contour points measured in different axes in three-dimensional space is ensured, solving the problem of point cloud misalignment caused by angle differences in multi-axis data, and providing a data foundation for the geometric consistency of the subsequent three-dimensional model.

[0087] Furthermore, the integration unit includes: an original integration subunit and an adjustment subunit;

[0088] The original integration subunit is used to integrate the coordinates of each sampling point according to the relative angle between each of the axial angles to construct the original three-dimensional point cloud data.

[0089] The adjustment subunit is used to fit the irregular edge discrete points at the equator of the lens to the original three-dimensional point cloud data according to the Fourier series algorithm, and to fit the sparse point cloud data of the lens according to the Zernike polynomial algorithm, and output the three-dimensional model of the lens.

[0090] In this preferred embodiment, the original three-dimensional point cloud data is constructed by integrating the coordinates of sampling points at various axial angles, preserving the original information of multi-axial measurements. The irregular edge discrete points at the lens equator are fitted using the Fourier series algorithm to correct noise errors in the discrete points and improve the smoothness and continuity of the equatorial contour. The sparse point cloud data of the lens is fitted using the Zernike polynomial algorithm to complete the geometric information of the missing point cloud areas, thereby enhancing the integrity and geometric accuracy of the three-dimensional model.

[0091] The model subdivision module M3 is used to perform three-dimensional Delaunay triangulation on the three-dimensional model of the lens to obtain each tetrahedral mesh.

[0092] Furthermore, the model segmentation module M3 includes: a segmentation unit and a filtering unit;

[0093] The subdivision unit is used to perform three-dimensional Delaunay triangulation on the three-dimensional model of the lens according to the Alpha shape algorithm to generate each original tetrahedral mesh.

[0094] The filtering unit is used to filter each of the original tetrahedral meshes, and the original tetrahedral meshes whose circumscribed sphere radius is less than or equal to a preset threshold are used as the original tetrahedral meshes.

[0095] In this preferred embodiment, the Alpha shape algorithm is used to perform three-dimensional Delaunay triangulation on the three-dimensional model of the lens. The generated original tetrahedral mesh can better fit the topological structure of the lens surface, avoiding the over-segmentation or under-segmentation problems of traditional segmentation methods. The original tetrahedral mesh with a circumscribed sphere radius less than or equal to a preset threshold is selected as the effective mesh, eliminating redundant or abnormal meshes on the model surface, ensuring that only geometric units reflecting the true surface of the lens are retained, thus providing a guarantee for the accuracy of subsequent surface area calculation.

[0096] The area calculation module M4 is used to calculate the surface area of ​​the three-dimensional model of the lens based on the area of ​​the exposed triangular facets of each tetrahedral mesh.

[0097] Furthermore, the area calculation module M4 can be implemented through the following preferred embodiments, as detailed below:

[0098] The surface area of ​​the three-dimensional model of the lens is obtained by summing the areas of each of the exposed triangular facets.

[0099] In this preferred embodiment, the surface area of ​​the three-dimensional lens model is calculated by summing the areas of each exposed triangular facet. The numerical summation is performed directly based on the geometric units of the model surface, avoiding errors caused by indirect estimation or empirical formulas, and ensuring the authenticity and traceability of the surface area value.

[0100] The result output module M5 is used to take half of the surface area as the posterior capsule area of ​​the lens after surgery, and calculate and output the simulated diameter of the capsular bag of the lens after surgery.

[0101] Furthermore, the result output module M5 can be implemented through the following preferred embodiments, as detailed below:

[0102] The posterior capsule is considered to be roughly circular, and based on the geometric relationship between the area and diameter of the roughly circular shape, the diameter is calculated according to the area of ​​the posterior capsule, which is used as the simulated diameter of the capsule.

[0103] In this preferred embodiment, the posterior capsule is considered to be roughly circular, and the simulated diameter of the capsule is calculated based on the geometric relationship between the area and diameter of the roughly circular shape. This utilizes the reasonable approximation of the expansion morphology of the posterior capsule to a roughly circular shape (the posterior capsule naturally expands after surgery), transforming the abstract area parameter into a linear indicator (diameter) that can be directly referenced clinically. This simplifies the practicality of the parameter output and allows for a more intuitive selection of the matching capsule implant size based on the capsule diameter in clinical practice, improving the efficiency and accuracy of preoperative planning.

[0104] In summary, compared with the prior art, the embodiments of this application have the following beneficial effects: by acquiring anterior segment OCT images and lens contour parameters from multiple axial angles, multi-dimensional morphological data is provided for three-dimensional modeling, avoiding the one-sidedness problem caused by single-axis or limited-axis measurements; a three-dimensional lens model is constructed based on the contour parameters, realizing a three-dimensional representation of the lens morphology, which is closer to the real anatomical structure than two-dimensional images; a three-dimensional Delaunay triangulation is performed on the three-dimensional model to obtain a tetrahedral mesh, discretizing the complex curved surface into computable geometric units, providing a basis for the accurate calculation of the subsequent surface area; the total surface area of ​​the lens is calculated by accumulating the areas of exposed triangular facets, directly performing numerical summation based on the geometric units of the model surface, avoiding traditional approximate fitting or two-dimensional projection substitution. The resulting error is addressed by considering that the effective surface area of ​​the capsular bag after surgery is determined solely by the retained posterior and anterior capsular remnants. However, the role of the anterior capsular remnant is "covered" by the optical surface of the intraocular lens. Therefore, the actual effective support area mainly depends on the posterior capsular bag. Using half of the total lens surface area as the posterior capsular area is based on the biconvex symmetry of the lens (the anterior and posterior surface areas are approximately equal), making the derivation of the posterior capsular area more consistent with the actual range of the membranous structure retained after surgery. This solves the problem of the lack of a posterior capsular area calculation method in existing technologies. Furthermore, by calculating the posterior capsular area, the simulated diameter of the capsular bag is output, transforming the abstract area parameter into a linear indicator that can be directly referenced clinically. This compensates for the shortcomings of existing technologies that rely solely on the corneal transverse diameter or the preoperative lens diameter under a regular elliptical model for size judgment. The overall solution integrates the entire process from multi-source data acquisition to 3D modeling, subdivision calculation, and parameter derivation. For the first time, it has achieved quantitative calculation of the posterior capsule area and lens surface area. It specifically addresses the lack of lens or posterior capsule area calculation methods and inaccurate capsular parameter prediction in existing technologies. It can provide clinicians with a more realistic simulated capsular diameter that closely matches the actual anatomical structure, guide the selection of implant size within the capsular bag, improve the matching degree between the artificial lens and the capsular bag, and reduce the risk of postoperative dislocation or refractive abnormalities.

[0105] Example 3:

[0106] Figure 3 A structural diagram of a postoperative lens capsule parameter prediction device according to this application is presented. (See diagram below.) Figure 3 As shown, the postoperative lens capsule parameter prediction device may include: processor N1, memory N2, data interface N3, and communication bus N4.

[0107] Wherein: processor N1, memory N2, and data interface N3 communicate with each other through communication bus N4; data interface N3 is used for data communication with other devices such as input devices or output devices; processor N1 is used to execute program N5, which can specifically execute the relevant steps in any of the above embodiments of the postoperative lens capsule parameter prediction method.

[0108] Specifically, program N5 may include program code, which includes computer-executable instructions.

[0109] The processor N1 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. The postoperative lens capsule parameter prediction device includes one or more processors, which may be processors of the same type, such as one or more CPUs, or processors of different types, such as one or more CPUs and one or more ASICs.

[0110] Memory N2 is used to store program N5. Memory N2 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage.

[0111] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Furthermore, the embodiments in this application are not directed to any particular programming language.

[0112] Example 4:

[0113] This invention also provides a computer-readable storage medium storing at least one executable instruction that, when executed on a postoperative lens capsule parameter prediction device / system, causes the postoperative lens capsule parameter prediction device / system to perform a postoperative lens capsule parameter prediction method from any of the above method embodiments.

[0114] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. Similarly, for the purpose of simplification and aiding understanding of one or more aspects of the invention, in the above description of exemplary embodiments of this application, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of this application.

[0115] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

Claims

1. A method for predicting parameters of the lens capsule after surgery, characterized in that, The method comprises the following steps: obtaining preoperative anterior segment OCT images, and extracting lens profile line parameters corresponding to two-dimensional lens images at different axial angles in each of the anterior segment OCT images; constructing a three-dimensional lens model according to the lens profile line parameters; performing three-dimensional Delaunay triangulation on the three-dimensional lens model to obtain each tetrahedral mesh; calculating the surface area of the three-dimensional lens model according to the area of the exposed triangular face of each tetrahedral mesh; taking half of the surface area as the posterior capsule area of the postoperative lens, and calculating and outputting the simulated diameter of the capsular bag of the postoperative lens.

2. A method of predicting parameters of a post-surgical lens capsular bag as claimed in claim 1, wherein, The method of constructing a three-dimensional lens model according to the lens profile line parameters comprises the following steps: sampling each lens profile line parameter to obtain each sampling point coordinate on the lens profile curve; integrating each sampling point coordinate according to the relative angle between each axial angle, constructing three-dimensional point cloud data, and outputting a three-dimensional lens model.

3. A method of predicting parameters of a post-surgical lens capsular bag according to claim 2, wherein, The method of integrating each sampling point coordinate according to the relative angle between each axial angle, constructing three-dimensional point cloud data, and outputting a three-dimensional lens model comprises the following steps: integrating each sampling point coordinate according to the relative angle between each axial angle to construct original three-dimensional point cloud data; in the original three-dimensional point cloud data, performing fitting processing on the irregular edge discrete points at the equator of the lens according to the Fourier series algorithm, and performing fitting processing on the sparse point cloud data of the lens according to the Zernike polynomial algorithm, and outputting the three-dimensional lens model.

4. The method for predicting parameters of the postoperative lens capsule as described in claim 1, characterized in that, The method of performing three-dimensional Delaunay triangulation on the three-dimensional lens model to obtain each tetrahedral mesh comprises the following steps: performing three-dimensional Delaunay triangulation on the three-dimensional lens model according to the Alpha shape algorithm to generate each original tetrahedral mesh; screening each original tetrahedral mesh with an inscribed sphere radius less than or equal to a preset threshold as each tetrahedral mesh.

5. A method of predicting parameters of a post-surgical lens capsular bag according to claim 4, wherein, The method of calculating the surface area of the three-dimensional lens model according to the area of the exposed triangular face of each tetrahedral mesh comprises the following steps: accumulating and summing the area of each exposed triangular face to obtain the surface area of the three-dimensional lens model.

6. A method of predicting parameters of a post-surgical lens capsular bag according to claim 5, wherein, The method of taking half of the surface area as the posterior capsule area of the postoperative lens, and calculating and outputting the simulated diameter of the capsular bag of the postoperative lens comprises the following steps: regarding the posterior capsule as a circle, and calculating the diameter according to the area of the posterior capsule based on the geometric relationship between the area and the diameter of the circle, as the simulated diameter of the capsular bag.

7. A system for post-operative intraocular lens capsular bag parameter prediction, comprising: The method comprises the following steps: data acquisition module, model construction module, model division module, area calculation module and result output module; the data acquisition module is used for obtaining preoperative anterior segment OCT images, and extracting lens profile line parameters corresponding to two-dimensional lens images at different axial angles in each of the anterior segment OCT images; the model construction module is used for constructing a three-dimensional lens model according to the lens profile line parameters; The model splitting module is configured to perform three-dimensional Delaunay triangulation on the three-dimensional lens model to obtain each tetrahedral mesh; The area calculation module is configured to calculate the surface area of the three-dimensional lens model according to the area of the exposed triangular patch of each tetrahedral mesh; The result output module is configured to output half of the surface area as the posterior capsular area of the postoperative lens, and calculate and output the simulated diameter of the capsular bag of the postoperative lens.

8. A system for predicting parameters of a post-surgical lens capsular bag as claimed in claim 7, wherein, The model construction module comprises a sampling unit and an integration unit. The sampling unit is configured to sample each lens profile line parameter to obtain each sampling point coordinate on the lens profile curve. The integration unit is configured to integrate each sampling point coordinate according to the relative angle between each axial angle, construct three-dimensional point cloud data, and output a three-dimensional lens model.

9. An apparatus for post-surgical lens capsule bag parameter prediction, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The computer program, when loaded into the processor, implements the steps of the parameter prediction method of the postoperative lens capsular bag according to any one of claims 1-6.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program, when executed by the processor, implements the steps of the parameter prediction method of the postoperative lens capsular bag according to any one of claims 1-6.

Images