Eye tissue layering information determination method and device, equipment and storage medium

Abstract

The invention relates to an eye tissue hierarchical information determination method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring an OCT image of an examined eye, and determining an initial eye tissue probability graph and an initial eye tissue boundary curve corresponding to a plurality of eye tissues of the examined eye based on the OCT image; wherein the initial eye tissue probability graph is used for indicating the probability that the area where each pixel in the OCT image is located is the corresponding eye tissue; the initial eye tissue boundary curve is used for indicating the initial contour of the corresponding eye tissue in the OCT image; and constructing an energy function based on the OCT image, the initial eye tissue probability graph and the initial eye tissue boundary curve, and correcting the initial eye tissue boundary curve based on the energy function to obtain a target eye tissue boundary curve of the examined eye. By adopting the method, the accuracy of eye tissue layering information can be improved.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, computer device, computer-readable storage medium, and computer program product for determining eye tissue layering information. Background Technology

[0002] Optical coherence tomography (OCT) is a high-resolution imaging technique that uses the principle of low coherence interference to obtain cross-sectional structural information of eye tissues. Determining the layering information of eye tissues based on OCT images and conducting further analysis on this basis is an indispensable and important step in the analysis of eye abnormalities.

[0003] In existing technologies, most methods employ pre-trained deep learning models to determine the layering information of the eye tissues in the examined eye. However, this method of using deep learning models to determine the layering information of the eye tissues in the examined eye suffers from low accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide an accurate method, apparatus, computer device, computer-readable storage medium, and computer program product for determining ocular tissue layering information in response to the above-mentioned technical problems.

[0005] In a first aspect, this application provides a method for determining ocular tissue layering information, including:

[0006] Obtain the OCT image of the eye under examination, and determine the initial eye tissue probability map and initial eye tissue boundary curve corresponding to multiple eye tissues of the eye under examination based on the OCT image;

[0007] Among them, the initial eye tissue probability map is used to indicate the probability that the region where each pixel is located in the OCT image is the corresponding eye tissue; the initial eye tissue boundary curve is used to indicate the initial contour of the corresponding eye tissue in the OCT image.

[0008] An energy function is constructed based on OCT images, initial eye tissue probability maps, and initial eye tissue boundary curves. The initial eye tissue boundary curves are then corrected based on the energy function to obtain the target eye tissue boundary curves of the examined eye.

[0009] In one embodiment, an energy function is constructed based on an OCT image, an initial eye tissue probability map, and an initial eye tissue boundary curve. This includes: obtaining constraint information for each eye tissue, including layer sequence constraint information, thickness constraint information, and smoothness constraint information; performing correction processing on the initial eye tissue probability map based on the constraint information to obtain a target eye tissue probability map; and constructing an energy function based on the OCT image, the target eye tissue probability map, and the initial eye tissue boundary curve.

[0010] In one embodiment, the initial eye tissue probability map is corrected based on constraint information to obtain the target eye tissue probability map, including: determining outliers in the initial eye tissue probability map based on sequence constraint information, thickness constraint information and smoothness constraint information; and adjusting the outliers in the initial eye tissue probability map to obtain the target eye tissue probability map.

[0011] In one embodiment, determining outliers in the initial eye tissue probability map based on layer sequence constraint information, thickness constraint information, and smoothness constraint information includes: based on layer sequence constraint information, detecting whether there is a probability distribution in the initial eye tissue probability map that violates a preset physiological order; if so, determining the first pixel probability value corresponding to the probability distribution as a first outlier; based on thickness constraint information, detecting whether there is a second pixel probability value in the initial eye tissue probability map that exceeds a preset physiological thickness range; if so, determining the second pixel probability value as a second outlier; based on smoothness constraint information, detecting whether there is a third pixel probability value in the initial eye tissue probability map that has a sudden change in thickness of adjacent A-scan layers; if so, determining the third pixel probability value as a third outlier; and determining the first, second, and third outliers as outliers in the initial eye tissue probability map.

[0012] In one embodiment, an energy function is constructed based on an OCT image, a target eye tissue probability map, and an initial eye tissue boundary curve, including: determining an external energy term based on a vector field convolution algorithm and an OCT image; determining a smoothing term and a curvature term based on the initial eye tissue boundary curve, and determining a probability weight term based on the target eye tissue probability map; determining an internal energy term based on the smoothing term, the curvature term, and the probability weight term, and determining the energy function based on the external energy term and the internal energy term.

[0013] In one embodiment, the initial eye tissue boundary curve is corrected based on an energy function to obtain the target eye tissue boundary curve of the examined eye, including: using an active contour algorithm to correct the initial eye tissue boundary curve based on an energy function to obtain the target eye tissue boundary curve of the examined eye.

[0014] In one embodiment, determining an initial eye tissue probability map and an initial eye tissue boundary curve corresponding to multiple eye tissues of the examined eye based on an OCT image includes: performing image preprocessing on the OCT image to obtain a preprocessed OCT image; and inputting the preprocessed OCT image into a deep learning model to obtain the initial eye tissue probability map and the initial eye tissue boundary curve output by the deep learning model.

[0015] Secondly, this application also provides an apparatus for determining ocular tissue layering information, comprising:

[0016] The acquisition module is used to acquire the OCT image of the eye being examined, and to determine the initial eye tissue probability map and initial eye tissue boundary curve corresponding to multiple eye tissues of the eye being examined based on the OCT image.

[0017] Among them, the initial eye tissue probability map is used to indicate the probability that the region where each pixel is located in the OCT image is the corresponding eye tissue; the initial eye tissue boundary curve is used to indicate the initial contour of the corresponding eye tissue in the OCT image.

[0018] The execution module is used to construct an energy function based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve, and to perform correction processing on the initial eye tissue boundary curve based on the energy function to obtain the target eye tissue boundary curve of the examined eye.

[0019] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any of the embodiments of the first aspect above.

[0020] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the embodiments of the first aspect above.

[0021] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the embodiments of the first aspect above.

[0022] The aforementioned method, apparatus, computer device, computer-readable storage medium, and computer program product for determining eye tissue layering information first acquire an OCT image of the eye being examined. Based on the OCT image, an initial eye tissue probability map and an initial eye tissue boundary curve corresponding to multiple eye tissues of the eye being examined are determined. The initial eye tissue probability map indicates the probability that the region where each pixel in the OCT image is located corresponds to the eye tissue, and the initial eye tissue boundary curve indicates the initial contour of the corresponding eye tissue in the OCT image. Then, an energy function is constructed based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve, and the initial eye tissue boundary curve is corrected based on the energy function to obtain the target eye tissue boundary curve of the eye being examined. The eye tissue layering information determination method provided in this application, after obtaining the initial eye tissue boundary curve, further corrects the initial eye tissue boundary curve using the OCT image and the initial eye tissue probability map to obtain the target eye tissue boundary curve, thereby effectively improving the accuracy of eye tissue layering information. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating a method for determining ocular tissue layering information in one embodiment;

[0025] Figure 2 This is a flowchart illustrating a method for determining an initial eye tissue probability map and an initial eye tissue boundary curve corresponding to multiple eye tissues of the examined eye based on OCT images in one embodiment.

[0026] Figure 3 This is a schematic diagram of the initial eye tissue probability map and the initial eye tissue boundary curve in one embodiment;

[0027] Figure 4 This is a flowchart illustrating a method for constructing an energy function in one embodiment;

[0028] Figure 5 This is a flowchart illustrating a method for obtaining a probability map of target eye tissue in one embodiment;

[0029] Figure 6 This is a flowchart illustrating a method for determining outliers in an initial eye tissue probability map in one embodiment.

[0030] Figure 7 This is a flowchart illustrating a method for constructing an energy function in another embodiment;

[0031] Figure 8 This is a flowchart illustrating a method for determining ocular tissue layering information in another embodiment;

[0032] Figure 9 This is a structural block diagram of an eye tissue layering information determination device in one embodiment;

[0033] Figure 10 This is an internal structural diagram of a computer device in one embodiment;

[0034] Figure 11 This is a diagram of the internal structure of a computer device in another embodiment. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0036] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0037] Optical coherence tomography (OCT) is a high-resolution imaging technique that uses the principle of low coherence interference to obtain cross-sectional structural information of eye tissues. Determining the layering information of eye tissues based on OCT images and conducting further analysis on this basis is an indispensable and important step in the analysis of eye abnormalities.

[0038] In one existing technology, the eye tissue layering information of the examined eye is determined by a combination of deep learning and shortest path methods. That is, a tissue probability map is first generated by a deep learning network, and then the eye tissue boundaries are determined from the tissue probability map using methods such as shortest path search, graph cut, and dynamic programming.

[0039] However, this method suffers from high computational complexity, which in turn makes it time-consuming to determine the layering information of eye tissues.

[0040] In another existing technique, a trained deep learning model is used to determine the layering information of the eye tissues in the examined eye. However, while this method of determining the layering information of the eye tissues using a deep learning model is fast, it suffers from low accuracy.

[0041] In view of this, this application provides a method for determining eye tissue layering information. First, an OCT image of the eye being examined is acquired. Based on the OCT image, an initial eye tissue probability map and an initial eye tissue boundary curve corresponding to multiple eye tissues in the examined eye are determined. The initial eye tissue probability map indicates the probability that the region where each pixel in the OCT image is located corresponds to the eye tissue, and the initial eye tissue boundary curve indicates the initial contour of the corresponding eye tissue in the OCT image. Then, an energy function is constructed based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve. The initial eye tissue boundary curve is then corrected based on the energy function to obtain the target eye tissue boundary curve of the examined eye. The eye tissue layering information determination method provided in this application, after obtaining the initial eye tissue boundary curve, further uses the OCT image and the initial eye tissue probability map to correct the initial eye tissue boundary curve to obtain the target eye tissue boundary curve, thereby effectively improving the accuracy of eye tissue layering information.

[0042] The method for determining ocular tissue layering information provided in this application can be executed by a computer device, which can be a terminal or a server.

[0043] In one exemplary embodiment, such as Figure 1 As shown, a method for determining ocular tissue layering information is provided, which includes the following steps:

[0044] Step 101: Obtain the OCT image of the eye being examined, and determine the initial eye tissue probability map and initial eye tissue boundary curve corresponding to multiple eye tissues of the eye being examined based on the OCT image.

[0045] Optionally, the OCT image can be a B-scan image. Multiple ocular tissues examined may include the cornea, lens, vitreous body, choroid, sclera, macula, optic nerve fiber layer, iris, ciliary body, corneal epithelium, corneal stroma, anterior lens capsule, and retina. The retina may include the retinal nerve fiber layer, ganglion cell layer, inner plexus layer, inner nuclear layer, outer nuclear layer, and retinal pigment epithelium.

[0046] Among them, the initial eye tissue probability map can be used to indicate the probability that the region where each pixel is located in the OCT image is the corresponding eye tissue, and the initial eye tissue boundary curve can be used to indicate the initial contour of the corresponding eye tissue in the OCT image.

[0047] In some exemplary embodiments, the computer device can acquire OCT images of the eye being examined. Specifically, the computer device can utilize an OCT device to acquire OCT images of the eye being examined. The OCT device can be a time-domain optical coherence tomography (TD-OCT), a spectral-domain optical coherence tomography (SD-OCT), or a swept-source optical coherence tomography (SS-OCT).

[0048] Furthermore, after acquiring the OCT image of the examined eye, the computer device can determine the initial eye tissue probability map and the initial eye tissue boundary curve corresponding to multiple eye tissues of the examined eye based on the OCT image.

[0049] Specifically, computer equipment can use threshold segmentation algorithms, K-means clustering algorithms, etc., to determine the initial eye tissue probability map and initial eye tissue boundary curve corresponding to multiple eye tissues of the examined eye based on OCT images.

[0050] Step 102: Construct an energy function based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve, and perform correction processing on the initial eye tissue boundary curve based on the energy function to obtain the target eye tissue boundary curve of the examined eye.

[0051] Optionally, the energy function is a core component of the active contouring algorithm. For example, the energy function may include an internal energy term and an external energy term.

[0052] In some exemplary embodiments, after determining the initial eye tissue probability map and initial eye tissue boundary curve corresponding to multiple eye tissues of the examined eye based on the OCT image, the computer device can construct an energy function based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve.

[0053] Specifically, computer equipment can determine the external energy term in the energy function based on OCT images, and determine the internal energy term in the energy function based on the initial eye tissue probability map and the initial eye tissue boundary curve, in order to construct the energy function.

[0054] Furthermore, after constructing an energy function based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve, the computer device can perform correction processing on the initial eye tissue boundary curve based on the energy function to obtain the target eye tissue boundary curve of the examined eye.

[0055] Specifically, the computer device can input the energy function and the initial eye tissue boundary curve into a pre-trained calibration model to obtain the target eye tissue boundary curve of the tested eye output by the calibration model.

[0056] The aforementioned method for determining eye tissue layering information first acquires an OCT image of the examined eye, and then determines an initial eye tissue probability map and an initial eye tissue boundary curve corresponding to multiple eye tissues in the examined eye based on the OCT image. The initial eye tissue probability map indicates the probability that the region where each pixel in the OCT image is located is the corresponding eye tissue, and the initial eye tissue boundary curve indicates the initial contour of the corresponding eye tissue in the OCT image. Then, an energy function is constructed based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve, and the initial eye tissue boundary curve is corrected based on the energy function to obtain the target eye tissue boundary curve of the examined eye. The eye tissue layering information determination method provided in this application, after obtaining the initial eye tissue boundary curve, further uses the OCT image and the initial eye tissue probability map to correct the initial eye tissue boundary curve to obtain the target eye tissue boundary curve, thereby effectively improving the accuracy of eye tissue layering information.

[0057] In one exemplary embodiment, such as Figure 2As shown, the initial eye tissue probability map and initial eye tissue boundary curve corresponding to multiple eye tissues of the examined eye are determined based on OCT images, including the following steps:

[0058] Step 201: Perform image preprocessing on the OCT image to obtain the preprocessed OCT image.

[0059] Optionally, image preprocessing may include one or more of the following: Gaussian filtering for noise reduction, adaptive histogram equalization, grayscale normalization, speckle artifact removal, motion artifact correction, region of interest extraction, and image smoothing.

[0060] In some exemplary embodiments, after acquiring an OCT image, the computer device can perform image preprocessing on the OCT image to obtain a preprocessed OCT image.

[0061] Specifically, the computer equipment can first employ an image resampling algorithm to scale and adjust the OCT image according to a preset physical size to unify the horizontal / vertical physical dimensions. This image resampling algorithm can be a linear interpolation algorithm, a bicubic interpolation algorithm, etc. The computer equipment can also flatten the OCT image to correct tilted eye tissue structures and reduce differences in slice position caused by natural tissue tilt or scanning angle deviations. Furthermore, the computer equipment can automatically crop the region of interest in the OCT image to remove invalid imaging areas and improve processing efficiency. Finally, the computer equipment can perform normalization and noise suppression processing on the OCT image.

[0062] Step 202: Input the preprocessed OCT image into the deep learning model to obtain the initial eye tissue probability map and initial eye tissue boundary curve output by the deep learning model.

[0063] Alternatively, the deep learning model can be an improved U-Net, U-Net++, Swin-Transformer-U-Net, etc.

[0064] In some exemplary embodiments, after obtaining the preprocessed OCT image, the computer device can input the preprocessed OCT image into a deep learning model to obtain the initial eye tissue probability map and the initial eye tissue boundary curve output by the deep learning model.

[0065] Specifically, after the computer device inputs the pre-processed OCT image into the deep learning model, the deep learning model can first identify the type of eye tissue in the pre-processed OCT image. For example, if the pre-processed OCT image is identified to contain eye tissues such as the retina, choroid, cornea, and retinal pigment epithelium, the output initial eye tissue probability map is a multi-channel probability map that corresponds one-to-one with each eye tissue. That is, it is a probability map of the retina, choroid, cornea, and retinal pigment epithelium. The pixel value in each probability map represents the probability that the corresponding pixel belongs to that eye tissue.

[0066] Furthermore, deep learning models can employ soft parameterized maximization or differentiable maximization algorithms to perform differentiable peak localization on the probability distribution of each row and column in the multi-channel probability map, outputting discrete location points for each eye tissue boundary. Alternatively, through explicit regression network branches, control points of the parameterized curve of the eye tissue boundary can be regressed, and then a continuous initial eye tissue boundary curve can be generated based on these discrete location points or control points. The initial eye tissue probability map and the initial eye tissue boundary curve can be obtained as follows: Figure 3 As shown.

[0067] In one exemplary embodiment, such as Figure 4 As shown, the energy function is constructed based on OCT images, initial eye tissue probability maps, and initial eye tissue boundary curves, including the following steps:

[0068] Step 401: Obtain constraint information for each eye tissue.

[0069] Optionally, the constraint information may include layer sequence constraint information, thickness constraint information, and smoothness constraint information.

[0070] Layered constraint information can be used to define the relative positional relationships of different eye tissues. For example, in the case of the retina, the layered constraint information can be that the retina is located between the vitreous body and the retinal pigment epithelium.

[0071] Thickness constraint information can be used to limit the range of thickness variation for different eye tissues. For example, in the case where the eye tissue is the retinal nerve fiber layer, the thickness constraint information can be that the thickness of the retinal nerve fiber layer is between 80 and 120 μm.

[0072] Smoothing constraint information is used to limit the local continuity of different eye tissues. For example, when the eye tissue is corneal epithelium, the smoothing constraint information can be that the curvature value of the boundary curve of the corneal epithelium is not greater than a preset threshold, so as to ensure that the boundary curve has no abrupt changes and no jagged fluctuations, which conforms to the physiological morphological continuity characteristics of the eye tissue.

[0073] Step 402: Based on the constraint information, the initial eye tissue probability map is corrected to obtain the target eye tissue probability map.

[0074] In some exemplary embodiments, after acquiring constraint information for each eye tissue, the computer device can perform correction processing on the initial eye tissue probability map based on the constraint information to obtain the target eye tissue probability map.

[0075] Specifically, computer equipment can perform correction processing on the initial eye tissue probability map based on hierarchical constraint information, thickness constraint information, and smoothness constraint information to obtain an eye tissue probability map carrying physiological prior constraints, which is the target eye tissue probability map. By incorporating physiological prior knowledge (i.e., constraint information) into the pixel-level distribution of the target eye tissue probability map, not only are abnormal distributions that may violate physiological laws in the initial probability map effectively corrected, but the target eye tissue probability map also strictly follows the anatomical structure of the eye tissue, possessing structural continuity and distribution rationality that conform to physiological characteristics.

[0076] Step 403: Construct an energy function based on the OCT image, the probability map of the target eye tissue, and the initial eye tissue boundary curve.

[0077] In some exemplary embodiments, after the computer device corrects the initial eye tissue probability map based on constraint information to obtain the target eye tissue probability map, it can construct an energy function based on the OCT image, the target eye tissue probability map, and the initial eye tissue boundary curve.

[0078] Specifically, computer equipment can determine the external energy term based on OCT images, determine the internal energy term based on the target eye tissue probability map and the initial eye tissue boundary curve, and construct an energy function based on the external and internal energy terms.

[0079] In one exemplary embodiment, such as Figure 5 As shown, based on constraint information, the initial eye tissue probability map is corrected to obtain the target eye tissue probability map, including the following steps:

[0080] Step 501: Determine outliers in the initial eye tissue probability map based on sequence constraint information, thickness constraint information, and smoothness constraint information.

[0081] In some exemplary embodiments, after obtaining constraint information, the computer can determine outliers in the initial eye tissue probability map based on layer sequence constraint information, thickness constraint information, and smoothness constraint information.

[0082] Specifically, computer equipment can input sequence constraint information, thickness constraint information, smoothness constraint information, and initial eye tissue probability map into the image analysis model, so that the image analysis model can determine outliers in the initial eye tissue probability map that do not meet the sequence constraint information, thickness constraint information, and smoothness constraint information based on the sequence constraint information, thickness constraint information, and smoothness constraint information.

[0083] Step 502: Adjust the outliers in the initial eye tissue probability map to obtain the target eye tissue probability map.

[0084] In some exemplary embodiments, after determining outliers in the initial eye tissue probability map based on hierarchical constraint information, thickness constraint information, and smoothness constraint information, the computer device can adjust the outliers in the initial eye tissue probability map to obtain the target eye tissue probability map.

[0085] Specifically, outliers can be divided into two categories: one is determined based on sequence constraint information, and the other is determined based on thickness constraint information and smoothness constraint information. For outliers determined based on sequence constraint information, redistribution processing can be performed, while for outliers determined based on thickness constraint information and smoothness constraint information, suppression processing can be performed.

[0086] In one exemplary embodiment, such as Figure 6 As shown, the outliers in the initial eye tissue probability map are determined based on sequence constraint information, thickness constraint information, and smoothness constraint information, including the following steps:

[0087] Step 601: Based on the hierarchical constraint information, detect whether there is a probability distribution in the initial eye tissue probability map that violates the preset physiological order. If so, determine the first pixel probability value corresponding to the probability distribution as the first outlier.

[0088] In some exemplary embodiments, a computer device can detect, based on hierarchical constraint information, whether there is a probability distribution in the initial eye tissue probability map that violates a preset physiological order.

[0089] Specifically, taking the initial eye tissue probability map corresponding to the retinal nerve fiber layer in the eye tissue as an example, the layer order constraint information indicates that the retinal nerve fiber layer is located in the deep layer of the vitreous body and the surface layer of the ganglion cell layer. If, in the initial eye tissue probability map corresponding to the retinal nerve fiber layer, the high probability pixel region of the retinal nerve fiber layer appears in the region that does not meet the indication of the layer order constraint information, then it can be determined that there is a probability distribution that violates the preset physiological order.

[0090] Furthermore, if the computer device determines that there is a probability distribution in the initial eye tissue probability map that violates the preset physiological order, then the first pixel probability value corresponding to the probability distribution can be determined as the first outlier.

[0091] Step 602: Based on the thickness constraint information, detect whether there is a second pixel probability value in the initial eye tissue probability map that exceeds the preset physiological thickness range. If it exists, the second pixel probability value is determined as the second outlier.

[0092] In some exemplary embodiments, the computer device can detect whether there is a second pixel probability value in the initial eye tissue probability map that exceeds a preset physiological thickness range based on thickness constraint information. If it does exist, the second pixel probability value is determined as a second outlier.

[0093] Specifically, taking the initial eye tissue probability map corresponding to the retinal nerve fiber layer in the eye tissue as an example, the thickness constraint information indicates the preset physiological thickness range of the retinal nerve fiber layer. If the predicted thickness calculated based on the high probability pixel region of the tissue in the initial eye tissue probability map corresponding to the retinal nerve fiber layer exceeds the preset physiological thickness range indicated by the thickness constraint information, then it can be determined that there is a second pixel probability value that exceeds the preset physiological thickness range.

[0094] Furthermore, if the computer device determines that there is a second pixel probability value in the initial eye tissue probability map that exceeds the preset physiological thickness range, then the pixel probability value in the high probability pixel region corresponding to the predicted thickness that exceeds the range can be identified as the second outlier.

[0095] Step 603: Based on the smoothing constraint information, detect whether there is a third pixel probability value in the initial eye tissue probability map where the thickness of adjacent A-scan layers changes abruptly. If so, the third pixel probability value is determined as the third outlier.

[0096] In some exemplary embodiments, the computer device can detect, based on smoothing constraint information, whether there is a third pixel probability value in the initial eye tissue probability map where the thickness of adjacent A-scan layers abruptly changes.

[0097] Specifically, taking the initial eye tissue probability map corresponding to the retinal nerve fiber layer in the eye tissue as an example, the smoothing constraint information indicates that the thickness of the retinal nerve fiber layer should maintain physiological continuity between adjacent A-scans without significant abrupt changes. If, in the initial eye tissue probability map corresponding to the retinal nerve fiber layer, after extracting the thickness of the high-probability pixel region of the eye tissue on each A-scan, the difference in thickness between adjacent A-scans exceeds the range of continuous change allowed by the smoothing constraint information, then the probability value of the third pixel with a sudden change in thickness between adjacent A-scans can be determined.

[0098] Furthermore, if the computer device determines that there is a third pixel probability value in the initial eye tissue probability map where there is a change in the thickness of an adjacent A-scan layer, then the pixel probability value in the high-probability pixel region of the retinal nerve fiber layer on the A-scan corresponding to the change in layer thickness can be identified as the third outlier.

[0099] Step 604: Identify the first outlier, the second outlier, and the third outlier as outliers in the initial eye tissue probability map.

[0100] In some exemplary embodiments, after determining the first outlier, the second outlier, and the third outlier, the computer device may identify the first outlier, the second outlier, and the third outlier as outliers in the initial eye tissue probability map.

[0101] In one exemplary embodiment, such as Figure 7 As shown, the energy function is constructed based on the OCT image, the probability map of the target eye tissue, and the initial eye tissue boundary curve, including the following steps:

[0102] Step 701: Determine the external energy term based on the vector field convolution algorithm and OCT image.

[0103] Optionally, the external energy term can be used to provide convergence guidance for the initial eye tissue boundary curve, driving the initial eye tissue boundary curve to move towards the true edge of the eye tissue in the OCT image during the iteration process.

[0104] In some exemplary embodiments, a computer device can determine the external energy term based on a vector field convolution algorithm and an OCT image.

[0105] Specifically, the computer equipment can first use Gaussian filtering or a speckle removal algorithm to suppress speckle noise in the OCT image and retain the gray-level gradient features of the eye tissue edge. Then, it can calculate the gray-level gradient magnitude and gradient direction of the OCT image. Based on the gray-level gradient magnitude and gradient direction, an initial guiding vector field is constructed for each pixel of the OCT image, pointing towards the tissue edge and with intensity positively correlated with the gradient magnitude. Then, a preset convolution kernel is used to perform convolution operation on the initial guiding vector field to strengthen the continuous guiding signal of weak edges and broken edges and suppress noise interference. Finally, the convolution-enhanced vector field is mapped to an energy value distribution, which is the external energy term used to attract the initial eye tissue boundary curve to converge to the true position of the target eye tissue.

[0106] Step 702: Determine the smoothing term and curvature term based on the initial eye tissue boundary curve, and determine the probability weight term based on the target eye tissue probability map.

[0107] Optionally, the smoothing term, also known as the first-order derivative constraint, can suppress sawtooth fluctuations or local abrupt changes in the initial eye tissue boundary curve, ensuring that the constraint curve remains continuous and smooth.

[0108] The curvature term, also known as the second-order derivative constraint, can guide the initial eye tissue boundary curve to converge towards a natural shape with low curvature, avoiding deviation from the true contour of the tissue due to noise or local feature interference.

[0109] The probability weight term can bind the position of the initial eye tissue boundary curve to the high-probability region of the probability map, driving the initial eye tissue boundary curve to fit the real tissue region indicated by the target eye tissue probability map, thereby improving segmentation accuracy.

[0110] In some exemplary embodiments, the computer device may determine the smoothing term and curvature term based on the initial eye tissue boundary curve.

[0111] Specifically, the computer device can determine a smoothing term based on the discrete point distribution characteristics of the initial eye tissue boundary curve, and a curvature term based on the degree of curvature of the initial eye tissue boundary curve.

[0112] Furthermore, the computer device can also determine the probability weights based on the target eye tissue probability map.

[0113] Specifically, computer devices can determine probability weights based on the pixel probability distribution of the target eye tissue probability map.

[0114] Step 703: Determine the internal energy term based on the smoothing term, curvature term, and probability weight term, and determine the energy function based on the external energy term and the internal energy term.

[0115] In some exemplary embodiments, after determining the smoothing term, curvature term, and probability weight term, the computer device can determine the internal energy term based on the smoothing term, curvature term, and probability weight term.

[0116] Specifically, computer devices can identify the smoothing term, curvature term, and probability weight term as internal energy terms.

[0117] Furthermore, after determining the internal energy term based on the smoothing term, curvature term, and probability weight term, the computer device can determine the energy function based on the external energy term and the internal energy term.

[0118] Specifically, computer devices can define external energy terms and internal energy terms as energy functions.

[0119] In an exemplary embodiment, the initial eye tissue boundary curve is corrected based on an energy function to obtain the target eye tissue boundary curve of the examined eye, including: using an active contour algorithm to correct the initial eye tissue boundary curve based on an energy function to obtain the target eye tissue boundary curve of the examined eye.

[0120] In some exemplary embodiments, the computer device may use an active contouring algorithm to correct the initial eye tissue boundary curve based on an energy function to obtain the target eye tissue boundary curve of the examined eye.

[0121] Specifically, the computer device can utilize an active contour algorithm to iteratively optimize the initial eye tissue boundary curve based on an energy function to determine a new initial eye tissue boundary curve. Based on this new boundary curve, it determines new smoothing and curvature terms, and then, based on these terms and probability weights, determines a new internal energy term. Finally, it determines a new energy function based on these internal and external energy terms. The device then executes the iterative optimization of the new initial eye tissue boundary curve using the active contour algorithm and the new energy function, until the energy function converges to a preset threshold, the positional change of the boundary curve is less than a set threshold, or the number of iterations reaches a preset upper limit. The new initial eye tissue boundary curve is then identified as the target eye tissue boundary curve. The target eye tissue boundary curve is characterized by continuous position, smooth curvature, and consistent physiological order.

[0122] In an optional embodiment of this application, after obtaining the target eye tissue boundary curve, the computer device can determine the corneal thickness distribution map, anterior chamber depth, lens morphology imaging, thickness of each retinal layer, volume map of each retinal layer, and three-dimensional reconstruction model based on the target eye tissue boundary curve.

[0123] In one exemplary embodiment, such as Figure 8 As shown, another method for determining ocular tissue layering information is provided, which includes the following steps:

[0124] Step 801: Acquire the OCT image of the eye being examined, perform image preprocessing on the OCT image to obtain the preprocessed OCT image; input the preprocessed OCT image into the deep learning model to obtain the initial eye tissue probability map and initial eye tissue boundary curve output by the deep learning model; the initial eye tissue probability map is used to indicate the probability that the region where each pixel in the OCT image is located is the corresponding eye tissue; the initial eye tissue boundary curve is used to indicate the initial contour of the corresponding eye tissue in the OCT image;

[0125] Step 802: Obtain constraint information for each eye tissue, including layer sequence constraint information, thickness constraint information, and smoothness constraint information; based on the layer sequence constraint information, detect whether there is a probability distribution in the initial eye tissue probability map that violates the preset physiological order. If so, determine the first pixel probability value corresponding to the probability distribution as the first outlier; based on the thickness constraint information, detect whether there is a second pixel probability value in the initial eye tissue probability map that exceeds the preset physiological thickness range. If so, determine the second pixel probability value as the second outlier; based on the smoothness constraint information, detect whether there is a third pixel probability value in the initial eye tissue probability map that has a sudden change in thickness of adjacent A-scan layers. If so, determine the third pixel probability value as the third outlier; determine the first outlier, the second outlier, and the third outlier as outliers in the initial eye tissue probability map.

[0126] Step 803: Adjust the outliers in the initial eye tissue probability map to obtain the target eye tissue probability map; determine the external energy term based on the vector field convolution algorithm and OCT image; determine the smoothing term and curvature term based on the initial eye tissue boundary curve, and determine the probability weight term based on the target eye tissue probability map; determine the internal energy term based on the smoothing term, curvature term and probability weight term, and determine the energy function based on the external energy term and internal energy term; use the active contour algorithm to correct the initial eye tissue boundary curve based on the energy function to obtain the target eye tissue boundary curve of the examined eye.

[0127] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0128] Based on the same inventive concept, this application also provides an eye tissue layering information determination device for implementing the above-described eye tissue layering information determination method. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more eye tissue layering information determination device embodiments provided below can be found in the limitations of the eye tissue layering information determination method described above, and will not be repeated here.

[0129] In one exemplary embodiment, such as Figure 9 As shown, an eye tissue layering information determination device 900 is provided, including: an acquisition module 901 and an execution module 902, wherein:

[0130] The acquisition module 901 is used to acquire the OCT image of the eye under examination, and determine the initial eye tissue probability map and initial eye tissue boundary curve corresponding to multiple eye tissues of the eye under examination based on the OCT image.

[0131] Among them, the initial eye tissue probability map is used to indicate the probability that the region where each pixel is located in the OCT image is the corresponding eye tissue; the initial eye tissue boundary curve is used to indicate the initial contour of the corresponding eye tissue in the OCT image.

[0132] The execution module 902 is used to construct an energy function based on the OCT image, the initial eye tissue probability map and the initial eye tissue boundary curve, and to perform correction processing on the initial eye tissue boundary curve based on the energy function to obtain the target eye tissue boundary curve of the examined eye.

[0133] In one embodiment, the execution module 902 is specifically used to acquire constraint information of each eye tissue, including layer sequence constraint information, thickness constraint information and smoothness constraint information; based on the constraint information, the initial eye tissue probability map is corrected to obtain the target eye tissue probability map; and an energy function is constructed based on the OCT image, the target eye tissue probability map and the initial eye tissue boundary curve.

[0134] In one embodiment, the execution module 902 is specifically used to determine outliers in the initial eye tissue probability map based on the sequence constraint information, thickness constraint information, and smoothness constraint information; and to adjust the outliers in the initial eye tissue probability map to obtain the target eye tissue probability map.

[0135] In one embodiment, the execution module 902 is specifically configured to: detect, based on layer order constraint information, whether there is a probability distribution in the initial eye tissue probability map that violates a preset physiological order; if so, determine the first pixel probability value corresponding to the probability distribution as a first outlier; detect, based on thickness constraint information, whether there is a second pixel probability value in the initial eye tissue probability map that exceeds a preset physiological thickness range; if so, determine the second pixel probability value as a second outlier; detect, based on smoothness constraint information, whether there is a third pixel probability value in the initial eye tissue probability map that has a sudden change in thickness of adjacent A-scan layers; if so, determine the third pixel probability value as a third outlier; and determine the first outlier, the second outlier, and the third outlier as outliers in the initial eye tissue probability map.

[0136] In one embodiment, the execution module 902 is specifically used to determine the external energy term based on the vector field convolution algorithm and the OCT image; determine the smoothing term and curvature term based on the initial eye tissue boundary curve, and determine the probability weight term based on the target eye tissue probability map; determine the internal energy term based on the smoothing term, curvature term and probability weight term, and determine the energy function based on the external energy term and the internal energy term.

[0137] In one embodiment, the execution module 902 is specifically used to use an active contour algorithm to correct the initial eye tissue boundary curve based on an energy function in order to obtain the target eye tissue boundary curve of the examined eye.

[0138] In one embodiment, the acquisition module 901 is specifically used to perform image preprocessing on the OCT image to obtain the preprocessed OCT image; and input the preprocessed OCT image into the deep learning model to obtain the initial eye tissue probability map and the initial eye tissue boundary curve output by the deep learning model.

[0139] The modules in the aforementioned eye tissue layering information determination device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0140] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 10As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a method for determining eye tissue layering information.

[0141] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 11 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a method for determining eye tissue layering information. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0142] Those skilled in the art will understand that Figure 10 and Figure 11The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0143] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0144] Obtain the OCT image of the eye under examination, and determine the initial eye tissue probability map and initial eye tissue boundary curve corresponding to multiple eye tissues of the eye under examination based on the OCT image;

[0145] Among them, the initial eye tissue probability map is used to indicate the probability that the region where each pixel is located in the OCT image is the corresponding eye tissue; the initial eye tissue boundary curve is used to indicate the initial contour of the corresponding eye tissue in the OCT image.

[0146] An energy function is constructed based on OCT images, initial eye tissue probability maps, and initial eye tissue boundary curves. The initial eye tissue boundary curves are then corrected based on the energy function to obtain the target eye tissue boundary curves of the examined eye.

[0147] In one embodiment, when the processor executes the computer program, it further performs the following steps: acquiring constraint information for each eye tissue, including layer sequence constraint information, thickness constraint information, and smoothness constraint information; performing correction processing on the initial eye tissue probability map based on the constraint information to obtain the target eye tissue probability map; and constructing an energy function based on the OCT image, the target eye tissue probability map, and the initial eye tissue boundary curve.

[0148] In one embodiment, when the processor executes the computer program, it further performs the following steps: determining outliers in the initial eye tissue probability map based on hierarchical constraint information, thickness constraint information, and smoothness constraint information; and adjusting the outliers in the initial eye tissue probability map to obtain the target eye tissue probability map.

[0149] In one embodiment, when the processor executes the computer program, it further implements the following steps: based on the layer order constraint information, detecting whether there is a probability distribution in the initial eye tissue probability map that violates the preset physiological order; if so, determining the first pixel probability value corresponding to the probability distribution as a first outlier; based on the thickness constraint information, detecting whether there is a second pixel probability value in the initial eye tissue probability map that exceeds the preset physiological thickness range; if so, determining the second pixel probability value as a second outlier; based on the smoothness constraint information, detecting whether there is a third pixel probability value in the initial eye tissue probability map that has a sudden change in thickness of adjacent A-scan layers; if so, determining the third pixel probability value as a third outlier; and determining the first outlier, the second outlier, and the third outlier as outliers in the initial eye tissue probability map.

[0150] In one embodiment, when the processor executes the computer program, it further performs the following steps: determining an external energy term based on a vector field convolution algorithm and an OCT image; determining a smoothing term and a curvature term based on an initial eye tissue boundary curve, and determining a probability weight term based on a target eye tissue probability map; determining an internal energy term based on the smoothing term, curvature term, and probability weight term, and determining an energy function based on the external energy term and the internal energy term.

[0151] In one embodiment, when the processor executes the computer program, it further performs the following steps: using an active contour algorithm to correct the initial eye tissue boundary curve based on an energy function to obtain the target eye tissue boundary curve of the examined eye.

[0152] In one embodiment, when the processor executes the computer program, it further performs the following steps: preprocessing the OCT image to obtain a preprocessed OCT image; and inputting the preprocessed OCT image into a deep learning model to obtain an initial eye tissue probability map and an initial eye tissue boundary curve output by the deep learning model.

[0153] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the above embodiments.

[0154] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described in any of the above embodiments.

[0155] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0156] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0157] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0158] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for determining ocular tissue layering information, characterized in that, The method includes: Obtain an OCT image of the eye being examined, and determine an initial eye tissue probability map and an initial eye tissue boundary curve corresponding to multiple eye tissues of the eye being examined based on the OCT image; The initial eye tissue probability map is used to indicate the probability that the region where each pixel in the OCT image is located is the corresponding eye tissue; the initial eye tissue boundary curve is used to indicate the initial contour of the corresponding eye tissue in the OCT image. An energy function is constructed based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve. The initial eye tissue boundary curve is then corrected based on the energy function to obtain the target eye tissue boundary curve of the examined eye.

2. The method according to claim 1, characterized in that, The construction of the energy function based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve includes: Obtain constraint information for each of the aforementioned eye tissues, the constraint information including layer sequence constraint information, thickness constraint information, and smoothness constraint information; Based on the constraint information, the initial eye tissue probability map is corrected to obtain the target eye tissue probability map; An energy function is constructed based on the OCT image, the probability map of the target eye tissue, and the initial eye tissue boundary curve.

3. The method according to claim 2, characterized in that, The step of correcting the initial eye tissue probability map based on the constraint information to obtain the target eye tissue probability map includes: The outliers in the initial eye tissue probability map are determined based on the layer sequence constraint information, the thickness constraint information, and the smoothness constraint information. Outliers in the initial eye tissue probability map are adjusted to obtain the target eye tissue probability map.

4. The method according to claim 3, characterized in that, The step of determining outliers in the initial eye tissue probability map based on the layer sequence constraint information, the thickness constraint information, and the smoothness constraint information includes: Based on the layer order constraint information, it is detected whether there is a probability distribution that violates the preset physiological order in the initial eye tissue probability map. If there is, the first pixel probability value corresponding to the probability distribution is determined as the first outlier. Based on the thickness constraint information, it is detected whether there is a second pixel probability value in the initial eye tissue probability map that exceeds the preset physiological thickness range. If it exists, the second pixel probability value is determined as a second outlier. Based on the smoothing constraint information, it is detected whether there is a third pixel probability value in the initial eye tissue probability map where the thickness of adjacent A-scan layers changes abruptly. If it exists, the third pixel probability value is determined as the third outlier. The first outlier, the second outlier, and the third outlier are identified as outliers in the initial eye tissue probability map.

5. The method according to claim 2, characterized in that, The construction of the energy function based on the OCT image, the target eye tissue probability map, and the initial eye tissue boundary curve includes: The external energy term is determined based on the vector field convolution algorithm and the OCT image. Based on the initial eye tissue boundary curve, a smoothing term and a curvature term are determined, and based on the target eye tissue probability map, a probability weight term is determined. The internal energy term is determined based on the smoothing term, curvature term, and probability weight term, and the energy function is determined based on the external energy term and the internal energy term.

6. The method according to claim 2, characterized in that, The step of correcting the initial eye tissue boundary curve based on the energy function to obtain the target eye tissue boundary curve of the examined eye includes: Using an active contouring algorithm, the initial eye tissue boundary curve is corrected based on the energy function to obtain the target eye tissue boundary curve of the examined eye.

7. The method according to claim 1, characterized in that, The step of determining the initial eye tissue probability map and initial eye tissue boundary curve corresponding to multiple eye tissues of the examined eye based on the OCT image includes: The OCT image is preprocessed to obtain a preprocessed OCT image; The preprocessed OCT image is input into a deep learning model to obtain the initial eye tissue probability map and the initial eye tissue boundary curve output by the deep learning model.

8. A device for determining ocular tissue layering information, characterized in that, The device includes: The acquisition module is used to acquire the OCT image of the eye being examined, and to determine the initial eye tissue probability map and the initial eye tissue boundary curve corresponding to multiple eye tissues of the eye being examined based on the OCT image. The initial eye tissue probability map is used to indicate the probability that the region where each pixel in the OCT image is located is the corresponding eye tissue; the initial eye tissue boundary curve is used to indicate the initial contour of the corresponding eye tissue in the OCT image. The execution module is used to construct an energy function based on the OCT image, the initial eye tissue probability map, and the initial eye tissue boundary curve, and to perform correction processing on the initial eye tissue boundary curve based on the energy function to obtain the target eye tissue boundary curve of the examined eye.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images