Method for visualizing myopic fundus changes, visualization device, storage medium, and electronic device

Abstract

SOLUTION: A visualization method of an overcorrection fundus change comprises: a step for, on the basis of a processing object fundus image, determining a fundus feature image including a tigroid feature region; and a step for determining a first visualization image corresponding to the processing object fundus image on the basis of, the fundus feature image, where the first visualization image is used for expressing a distribution state of the tigroid spots, the method of the disclosure determining the first visualization image corresponding to the processing object fundus image by the fundus feature image including a tigroid spot feature region.EFFECT: As an effect, a first visualization image can reflect a distribution state of a fundus tigroid spot, so that a doctor can determine a seriousness degree of a fundus tigroid spot, by the first visualization image. Therefore, the invention can support so that the doctor can precisely grasp a change and advance of a near-sightedness fundus, and the invention can contribute to improvement of recognition to changes of tigroid spot and near-sightedness fundus, and more preferably and generally grasp, the state and advance of near-sightedness.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present disclosure relates to the technical field of image processing, and more particularly to a method, a visualization device, a storage medium, and an electronic device for visualizing changes in myopic fundus. [Background technology]

[0002] Fundus tessellated is a typical fundus change in the ocular tissue of myopic patients, which is commonly seen in myopic patients and leads to retinal damage. If not treated in a timely manner, fundus tessellated may eventually develop into pathological changes in the fundus or induce macular degeneration, which may seriously affect the patient's vision.

[0003] The evaluation of fundus leopard spots based on traditional fundus images is mainly based on subjective experience, which cannot accurately reflect the severity of leopard spots, and is also disadvantageous to observation. Therefore, if a doctor lacks relevant medical knowledge, he or she cannot accurately diagnose the severity of fundus leopard spots. Since the severity of fundus leopard spots depends on the level of the doctor's relevant knowledge, the detailed evaluation of fundus leopard spots is limited, which is disadvantageous to accurately grasp the situation and progression of myopic fundus changes. This causes some patients to miss the optimal timing of intervention treatment, which causes irreversible damage to the patient's vision or even blindness. Summary of the Invention [Problem to be solved by the invention]

[0004] In view of this, the present disclosure provides a method, a visualization device, a storage medium and an electronic device for visualizing changes in myopic fundus, to assist doctors in precisely understanding the changes and progression of myopic fundus, contributing to improving awareness of leopard spots and changes in myopic fundus, achieving the goal of better and more comprehensively understanding the situation and progression of myopia, and contributing to the establishment of comprehensive and precise preventive control of myopia and a corresponding precise preventive control system. [Means for solving the problem]

[0005] According to a first aspect, a method for visualizing myopic fundus changes in one embodiment of the present disclosure includes a step of determining a fundus characteristic image including a leopard-spot characteristic region based on a fundus image to be processed, and a step of determining a first visualized image corresponding to the fundus image to be processed based on the fundus characteristic image, wherein the first visualized image is used to represent the distribution of leopard-spot spots.

[0006] Based on the first aspect, in some embodiments of the first aspect, the step of determining a first visualized image corresponding to the fundus image to be processed based on the fundus feature image includes a step of determining a digital image corresponding to the fundus image to be processed based on the fundus feature image, where the digital image represents a distribution status of leopard-striped spots in the form of numbers, and a step of determining a first visualized image corresponding to the fundus image to be processed based on the digital image.

[0007] Based on the first aspect, in some embodiments of the first aspect, the step of determining a digital image corresponding to the fundus image to be processed based on the fundus feature image includes the steps of determining a plurality of window areas in the fundus feature image using a predetermined window, where the window area includes at least one pixel, determining a leopard-spot feature of the window area for each of the plurality of window areas, determining a leopard-spot feature index using the area of ​​the window area or feature data of the area of ​​interest in the fundus feature image based on the leopard-spot feature of the window area, and determining a digital image corresponding to the fundus image to be processed based on the leopard-spot feature index.

[0008] Based on the first aspect, in some embodiments of the first aspect, the step of determining a leopard-print feature of the window region for each of the multiple window regions includes: performing edge processing on the region of interest for each of the multiple window regions when the window region includes pixels located at an edge of the region of interest and an area of ​​the window region located outside the region of interest is greater than a predetermined threshold, where the edge processing includes: performing a mirroring process on the pixels of the edge of the region of interest based on the edge of the region of interest to obtain a mirrored edge region, and determining a leopard-print feature of the window region based on the window region including at least a portion of the mirrored edge region; or the step of determining a leopard-print feature of the window region for each of the multiple window regions includes: performing a dilation process on the edge of the region of interest to determine a dilated edge region, where the window region includes pixels located at an edge of the region of interest, and determining a leopard-print feature of the window region based on the window region including at least a portion of the dilated edge region,

[0009] Based on the first aspect, in some embodiments of the first aspect, the first visualized image includes a two-dimensional first visualized image or a three-dimensional first visualized image, and the step of determining the first visualized image corresponding to the fundus image to be processed based on the digital image includes a step of mapping the digital image into a two-dimensional color space to determine the two-dimensional first visualized image, or a step of mapping the digital image into a three-dimensional space to determine the three-dimensional first visualized image, or a step of mapping the digital image into the two-dimensional color space and then mapping the two-dimensional color space into the three-dimensional space to determine the three-dimensional first visualized image.

[0010] Based on the first aspect, in some embodiments of the first aspect, the fundus feature image further includes an atrophy area, and after the step of determining the fundus feature image based on the fundus image to be processed, the method for visualizing myopic fundus changes further includes a step of determining an atrophy area in the fundus feature image based on the fundus feature image, and a step of determining a second visualized image corresponding to the fundus image to be processed based on the first visualized image and the atrophy area, or a step of determining the second visualized image based on the first visualized image, the atrophy area and the leopard-striped feature area.

[0011] Based on the first aspect, in some embodiments of the first aspect, the fundus feature image further includes at least one of an atrophic arc region, a macular region, and an optic disc region, and after the step of determining the fundus feature image based on the fundus image to be processed, the method for visualizing myopic fundus changes further includes a step of determining an atrophic arc region in the fundus feature image based on the fundus feature image, a step of determining a landmark region of the fundus feature image based on at least one of the atrophic arc region, the macular region, and the optic disc region, wherein the landmark region represents an area that is presented to a user as requiring special attention, and a step of determining a third visualized image corresponding to the fundus image to be processed based on the first visualized image and the landmark region.

[0012] According to a second aspect, a visualization device for myopic fundus changes relating to one embodiment of the present disclosure includes a determination module configured to determine a fundus feature image including a leopard-spot feature region based on a fundus image to be processed, and a visualization module configured to determine a first visualized image corresponding to the fundus image to be processed based on the fundus feature image, wherein the first visualized image is used to represent the distribution status of the leopard-spot feature.

[0013] According to a third aspect, an electronic device according to one embodiment of the present disclosure includes a processor configured to execute the method for visualizing myopic fundus changes described in the first aspect above, and a memory configured to store instructions executable by the processor.

[0014] According to a fourth aspect, a computer-readable storage medium according to an embodiment of the present disclosure stores a computer program for executing the method for visualizing myopic fundus changes according to the first aspect. Effect of the Invention

[0015] In the embodiment of the present disclosure, a first visualized image expressing the distribution of leopard spots corresponding to the fundus image to be processed is determined by a fundus feature image including a leopard spot feature region. The first visualized image can intuitively reflect the distribution of leopard spots on the fundus, so that a doctor can determine the severity of the leopard spots on the fundus based on the first visualized image. Therefore, the method for visualizing changes in myopic fundus according to the embodiment of the present disclosure can assist a doctor to precisely grasp the changes and progression of myopic fundus, contribute to improving the awareness of leopard spots and changes in myopic fundus, achieve the purpose of better and more comprehensively grasping the situation and progression of myopia, and contribute to the comprehensive and precise prevention and control of myopia, and the establishment of a corresponding precise prevention and control system. [Brief description of the drawings]

[0016] The above and other objects, configurations and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail with reference to the drawings. The drawings are used to facilitate a better understanding of the embodiments of the present disclosure, constitute a part of the present specification, and are used to interpret the present disclosure together with the embodiments of the present disclosure, and are not intended to limit the present disclosure. [Figure 1] FIG. 1 is a schematic diagram of an application scene according to an embodiment of the present disclosure. [Diagram 2] 1 is a flowchart of a method for visualizing changes in myopic fundus according to an embodiment of the present disclosure. [Diagram 3] 11 is a flowchart for determining a first visualized image corresponding to a fundus image to be processed based on a fundus feature image according to an embodiment of the present disclosure. [Figure 4] 13 is a flowchart for determining a digital image corresponding to a target fundus image based on a fundus feature image according to an embodiment of the present disclosure. [Diagram 5] 6 is a flowchart for determining window region leopard print features for each of a plurality of window regions according to one embodiment of the present disclosure. [Figure 6] 11 is a flowchart of another method for visualizing changes in myopic fundus according to an embodiment of the present disclosure. [Figure 7] 11 is a flowchart of another method for visualizing changes in myopic fundus according to an embodiment of the present disclosure. [Figure 8] 1 is a structural schematic diagram of a visualization device for myopic fundus changes according to an embodiment of the present disclosure. [Figure 9] FIG. 1 is a structural schematic diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] The technical solutions in the embodiments of the present disclosure will be described below clearly and completely with reference to the drawings in the embodiments of the present disclosure. Needless to say, the described embodiments are only some embodiments of the present disclosure, but not all embodiments.

[0018] Fundus leopard spots are commonly seen in myopic patients, and are typical fundus changes in the ocular tissues of myopic patients, and the main damage they cause is retinal damage. The symptoms of retinal damage are various, such as peripheral retinal degeneration, retinal tears, or relatively close retinal separation centers, or the formation of neovascularization. If not treated in a timely manner, fundus leopard spots may eventually develop into pathological myopia, causing retinal detachment or inducing macular lesions, which will seriously affect the patient's vision.

[0019] Traditional evaluation of fundus leopard spots is mainly based on the subjective experience of experts, and is neither objective nor precise. Researchers or those with poor relevant medical knowledge cannot accurately determine the severity of fundus leopard spots through fundus images or fundus examination reports. Because the judgment of the severity of fundus leopard spots depends on the level of the doctor's relevant knowledge, the awareness of fundus leopard spots is limited, which is unfavorable to grasp the situation and progress of myopic fundus changes, and is unfavorable to the development of comprehensive and accurate myopia prevention and control and the establishment of a corresponding system. As a result, patients in some areas with underdeveloped medical care cannot receive timely intervention treatment, and their vision will eventually be irreversibly damaged.

[0020] Therefore, the present disclosure provides a method for visualizing changes in myopic fundus, assists physicians in precisely understanding the changes and progression of myopic fundus, contributes to improving awareness of leopard spots and changes in myopic fundus, achieves the goal of better and more comprehensively understanding the situation and progression of myopia, and is favorable for the establishment of comprehensive and precise preventive control of myopia and a corresponding precise preventive control system.

[0021] Below, a scene in which an embodiment of the present disclosure is applied will be briefly introduced with reference to FIG.

[0022] 1 is a schematic diagram of an application scene according to an embodiment of the present disclosure. As shown in FIG. 1, the scene is for processing an image. Specifically, the image processing scene includes a server 110 and a user terminal 120 communicatively connected to the server 110, and the server 110 is used to execute a method according to an embodiment of the present disclosure.

[0023] For example, in practical application, a user transmits an instruction to process an image via the user terminal 120. After receiving the instruction, the server 110 performs visualization processing on the fundus image to be processed to determine a first visualized image corresponding to the fundus image to be processed, and the first visualized image is used to represent the leopard-striped spot. Specifically, the specific process of visualization processing of the fundus image to be processed includes the steps of determining a fundus feature image including a leopard-striped spot feature region based on the fundus image to be processed, and determining a first visualized image corresponding to the fundus image to be processed based on the fundus feature image, where the first visualized image is used to represent the distribution of the leopard-striped spot.

[0024] Exemplarily, the above-mentioned fundus image to be processed includes, but is not limited to, all image data stored in the medical institution during the treatment period of the same or different patients, related fundus image data of one or more patients input by the user, and fundus image data taken by one or more patients upon request during a consultation. The fundus image to be processed may be a 45° fundus image, a 60° fundus image, a wide-angle fundus image, or a fundus image of another viewing angle, or even a fundus image of another modality. The fundus image may be taken centered on the optic disc, centered on the macula, or an image of another eye position. Exemplarily, the server 110 may receive the fundus image to be processed directly or may retrieve it from a data storage device.

[0025] By way of example, the above-mentioned user terminal 120 includes, but is not limited to, computer terminals such as desktop computers and notebook computers, and mobile terminals such as tablet computers and mobile phones.

[0026] 2 is a flowchart of a method for visualizing changes in myopic fundus according to an embodiment of the present disclosure. As shown in FIG. 2, the method for visualizing changes in myopic fundus according to an embodiment of the present disclosure includes the following steps.

[0027] In step S210, a fundus feature image is determined based on the fundus image to be processed.

[0028] The fundus feature image includes a leopard print feature region.

[0029] For example, a fundus characteristic image is determined based on the fundus image to be processed according to the feature data of the leopard print, and the fundus characteristic image includes a leopard print characteristic region corresponding to the leopard print feature data in the fundus image to be processed.

[0030] For example, fundus features of the fundus image to be processed are segmented and extracted, for example, leopard-spot spots are segmented and extracted, and feature data of the leopard-spot spots is determined based on the segmentation result.

[0031] In some embodiments, an initial processing target fundus image is acquired, and a pre-processing is performed on the initial processing target fundus image to determine the processing target fundus image. The pre-processing on the initial processing target fundus image includes, but is not limited to, extracting a region of interest (ROI) from the initial processing target fundus image, and performing image enhancement processing, such as sharpening, filtering, smoothing, etc., on the initial processing target fundus image to determine the processing target fundus image. Alternatively, after extracting an ROI from the initial processing target fundus image, an image enhancement processing may be performed on the extracted ROI region to determine the processing target fundus image. The embodiments of the present disclosure do not further limit the manner in which the initial processing target fundus image is pre-processed.

[0032] In step S220, a first visualized image corresponding to the fundus image to be processed is determined based on the fundus feature image.

[0033] The first visualized image is used to represent the distribution of leopard spots.

[0034] For example, the first visualized image corresponding to the fundus image to be processed is determined based on the fundus feature image by the density data of the leopard-striped spots in the feature data of the fundus leopard-striped spots (i.e., the ratio of the area of ​​the leopard-striped spots in the calculation range to the area of ​​the calculation range), the proportion data of the leopard-striped spots (i.e., the ratio of the area of ​​the leopard-striped spots in the calculation range to the area of ​​the leopard-striped spots in the fundus image to be processed), the fractal dimension of the leopard-striped spots, or the morphological data of the leopard-striped spots. The first visualized image can intuitively show the distribution of the leopard-striped spots. The first visualized image may be a two-dimensional heat map or a three-dimensional contour topographic map. Therefore, the doctor can intuitively observe the distribution of the leopard-striped spots through the first visualized image, and determine the severity of the leopard-striped spots according to the distribution of the leopard-striped spots.

[0035] In some embodiments, the specific implementation of step S220 is shown in FIG. 3, and the description is omitted here.

[0036] According to the visualization method of myopic fundus changes according to the embodiment of the present disclosure, a first visualized image corresponding to a processing target fundus image can be determined based on a fundus characteristic image including a leopard-spot feature region. Since the first visualized image can intuitively reflect the distribution of the leopard-spot feature, a doctor can determine the severity of the leopard-spot feature by the first visualized image. Therefore, the visualization method of myopic fundus changes according to the present disclosure can assist a doctor in diagnosing a leopard-spot feature and achieve the purpose of popularizing the diagnosis of a leopard-spot feature. In addition, the embodiment of the present disclosure determines a processing target fundus image by performing preprocessing on an initial processing target fundus image, thereby contributing to improving the accuracy of the subsequent processing process and the robustness of the method. This makes it possible to more accurately show the distribution of the leopard-spot feature shown in the first visualized image, improving the accuracy with which a doctor determines the severity of the leopard-spot feature by the first visualized image. This will enable doctors to better understand changes in the fundus of myopic patients, accurately monitor its progression, and better monitor the progress of myopia and provide preventive control.

[0037] 3 is a flowchart of determining a first visualized image corresponding to a fundus image to be processed based on a fundus feature image according to an embodiment of the present disclosure. As shown in FIG. 3, the step of determining a first visualized image corresponding to a fundus image to be processed based on a fundus feature image according to an embodiment of the present disclosure (i.e., step S220 according to the embodiment shown in FIG. 2) may specifically include the following steps.

[0038] In step S310, a digital image corresponding to the fundus image to be processed is determined based on the fundus feature image.

[0039] The digital image shows the distribution of the leopard spots in the form of numbers.

[0040] For example, a calculation range is selected based on the fundus feature image, and a digital image corresponding to the fundus image to be processed is determined according to the area of ​​the leopard-striped spots in the calculation range and the density or percentage of the leopard-striped spots. For example, the numbers in the digital image correspond to the density of the leopard-striped spots, and in this case, different numbers represent different distribution conditions of the leopard-striped spots. Note that the numbers in the digital image may represent the distribution conditions of the leopard-striped spots by corresponding to other characteristic parameters of the leopard-striped spots, such as the percentage of the leopard-striped spots, the fractal dimension of the leopard-striped spots, and the morphological data of the leopard-striped spots. The density of the leopard-striped spots is just one of the indices used for visualization, and visualization may be performed based on other indices.

[0041] In some embodiments, a specific implementation of step S310 is shown in FIG. 4, and the description is omitted here.

[0042] In step S320, a first visualized image corresponding to the fundus image to be processed is determined based on the digital image.

[0043] Exemplarily, a first visualized image corresponding to the fundus image to be processed is determined based on the digital image through a mapping process. Alternatively, a first visualized image corresponding to the fundus image to be processed is determined based on the digital image through processing with a visualization tool. Note that, according to needs, the first visualized image can be expressed in two dimensions or three dimensions, and a two-dimensional image or a three-dimensional image can be obtained by mapping the digital image or processing with a visualization tool accordingly.

[0044] Exemplarily, the numbers in the digital image can be mapped as greyscale values ​​into a colour space to obtain a two-dimensional first visualization image.

[0045] The embodiment of the present disclosure determines a first visualized image corresponding to the fundus image to be processed by a digital image, and the digital image shows the distribution of leopard-spot spots in the form of numbers. The result of the first visualized image obtained by the digital image is more accurate. In addition, the embodiment of the present disclosure can further select the expression format of the first visualized image according to different needs, so as to adapt to the needs of different scenes, and increase the use range of the first visualized image. Therefore, the embodiment of the present disclosure can achieve the purpose of assisting doctors in diagnosing leopard-spot spots in different scenes, thereby further promoting the popularization of the diagnosis of leopard-spot spots.

[0046] 4 is a flowchart of determining a digital image corresponding to a fundus image to be processed based on a fundus feature image according to an embodiment of the present disclosure. As shown in FIG. 4, the step of determining a digital image corresponding to a fundus image to be processed based on a fundus feature image according to an embodiment of the present disclosure (i.e., step S310 according to the embodiment shown in FIG. 3) may specifically include the following steps:

[0047] In step S410, a plurality of window regions in the fundus feature image are determined using a predetermined window.

[0048] The window region includes at least one pixel.

[0049] For example, a predetermined window that meets different observation ranges is set according to clinical needs, that is, a calculation range is selected. The shape and size of the predetermined window can be set according to clinical needs, for example, the predetermined window may be a square of 3 mm x 3 mm, a circle with a radius of 5 mm, or a polygon of other sizes selected according to needs. When the size of the predetermined window is set based on the diameter of the optic disc (PD), for example, when the window size is set between 1 PD and 2 PD, the processing effect is better.

[0050] For example, a predetermined window is used to slide to select window regions at different positions of the fundus feature image, thereby determining a plurality of window regions in the fundus feature image; or a predetermined window is used to select a plurality of regions from the fundus feature image that are the same size as the predetermined window according to needs based on the size of the predetermined window, thereby determining a plurality of window regions in the fundus feature image.

[0051] In step S420, a window region leopard print feature is determined for each of the plurality of window regions.

[0052] Illustratively, for each of the plurality of window regions, a leopard print characteristic of the window region is determined based on the leopard print in the respective window region.

[0053] In step S430, based on the leopard-spot feature of the window region, a leopard-spot index is determined by using the area of ​​the window region or the feature data of the region of interest in the fundus feature image.

[0054] Exemplarily, the leopard-spot index includes at least one of the density of the leopard-spot, the proportion of the leopard-spot, and the morphological parameters of the leopard-spot. The feature data of the region of interest in the fundus feature image includes at least one of the diameter of the region of interest, the area of ​​the region of interest, and the boundary data of the region of interest. Exemplarily, based on the leopard-spot feature of the window region, the area of ​​the leopard-spot in the leopard-spot feature of each window region is calculated to determine the leopard-spot area of ​​the window region. Based on the leopard-spot area of ​​the window region, the density of the leopard-spot or the proportion of the leopard-spot, i.e., the leopard-spot index is determined using the area of ​​the window region or the area of ​​the region of interest in the fundus feature image. Alternatively, the leopard-spot index is determined based on at least one of the diameter of the region of interest in the fundus feature image or the boundary data of the region of interest. The morphological parameters of the leopard pattern are parameters determined by the shape of the leopard pattern in the image, and the shape of the leopard pattern in the image includes at least one of the shape, curvature, area, density, width, fractal dimension, compactness, degree of intersection, and degree of branching of the leopard pattern. In some embodiments, the morphological parameters of the leopard pattern are directly proportional to the shape, curvature, area, density, width, fractal dimension, compactness, degree of intersection, and degree of branching of the leopard pattern, that is, the higher the shape, curvature, area, density, width, fractal dimension, compactness, degree of intersection, and degree of branching of the leopard pattern, the higher the morphological parameters of the leopard pattern. Here, the compactness of the leopard pattern is determined based on the length of the contour line of the region and the actual area within the region. Those skilled in the art can flexibly adjust the reference characteristics of the morphological parameters of the leopard pattern according to the actual application. The leopard spot index in the fundus image quantified by calculation can be mapped to biological parameters of the eye, such as eye axis, refractive power, etc. The leopard spot index can also reflect the change status of ocular structures, such as choroid and retina, and provide a reference standard for subsequent clinical diagnosis by doctors, surgery, eye examination by optometrists, etc.

[0055] In step S440, a digital image corresponding to the fundus image to be processed is determined based on the leopard mark index.

[0056] For example, the digital image corresponding to the target fundus image is determined based on the density or percentage of leopard-spot spots, and numbers in the digital image indicate the density or percentage of leopard-spot spots at the corresponding positions in the target fundus image.

[0057] The embodiment of the present disclosure determines a digital image corresponding to the fundus image to be processed by the leopard-striped spot index, and expresses the density or the proportion of the leopard-striped spot in the fundus image to be processed in a numerical value to represent the distribution of the leopard-striped spot. The embodiment of the present disclosure expresses the distribution of the leopard-striped spot with a digital image, thereby making it possible to more clearly show the distribution of the leopard-striped spot, and further improving the accuracy of the first visualized image.

[0058] Exemplarily, the leopard-spot index and the digital image corresponding to the fundus image to be processed may be used for image classification, i.e., to express whether the fundus image is a leopard-spot fundus or not. Optionally, a leopard-spot index threshold may be set, and if the leopard-spot index is less than the leopard-spot index threshold, the fundus image is expressed as a non-leopard-spot fundus, and if the leopard-spot index is greater than the leopard-spot index threshold, the fundus image is determined as a leopard-spot fundus. Furthermore, the severity of the leopard-spot fundus may be determined based on the interval in which the leopard-spot index is located. Exemplarily, the severity of the leopard-spot fundus includes mild, moderate and severe.

[0059] For example, when the leopard-striped spot index is the density of leopard-striped spots, the leopard-striped spot index threshold may be between 0 and 20%, and may be specifically 2%, 5%, 7.3%, 8%, 8.5%, 9%, etc. If the leopard-striped spot density is greater than the leopard-striped spot index threshold, the fundus image to be processed is determined as a leopard-striped fundus. If the leopard-striped spot density is less than the leopard-striped spot index threshold, the current fundus image is determined not to be a leopard-striped spot. Similarly, those skilled in the art can flexibly adjust the leopard-striped spot index threshold to adapt to the proportion of leopard-striped spots, the morphological parameters of leopard-striped spots, etc. according to the actual situation.

[0060] FIG. 5 is a flowchart of determining window region leopard print characteristics for each of a plurality of window regions according to one embodiment of the present disclosure.

[0061] In an embodiment of the present disclosure, the leopard-spot feature of the window area includes a leopard-spot area of ​​the window area, and as shown in FIG. 5, the step of determining the leopard-spot feature of the window area for each of the multiple window areas in the embodiment of the present disclosure (i.e., step S420 in the embodiment shown in FIG. 4) may specifically include the following steps:

[0062] In step S510, for each of the plurality of window regions, it is determined whether the area of ​​the window region that includes pixels located at the edge of the region of interest and that is located outside the region of interest is greater than a predetermined threshold.

[0063] Exemplarily, the plurality of window regions are determined based on the fundus image, so that the plurality of windows include at least one window region. For each of the plurality of window regions, at least one window region including pixels at an edge of the region of interest is determined.

[0064] For example, if the window region includes the pixel of the edge region of the region of interest, execute step S520 and / or step S540. Step S520 and step S540 can be selectively executed or executed simultaneously according to needs. If the window region does not include the pixel of the edge region of the region of interest, do not process the pixel in the window region. Note that, if the window region does not include the pixel of the edge region of the region of interest, the leopard-spot feature of the window region can be directly determined by calculation, for example, the leopard-spot feature of the window region can be determined by calculating the area of ​​the leopard-spot feature in the window region.

[0065] In step S520, a mirroring process is performed on the pixels on the edge of the region of interest based on the edge of the region of interest to obtain a mirrored edge region.

[0066] For example, since the fundus image to be processed is a circular region, it is easy to make an error in determining the value of the density of leopard spots or the percentage of leopard spots contained in the edge region, so the edge region of the region of interest is optimized.

[0067] For example, a mirroring process is performed on the pixels at the edge of the region of interest based on the edge of the region of interest, that is, a mirroring edge region is obtained by performing a mirroring process on the pixels at the edge of the region of interest along the edge of the region of interest.

[0068] In step S530, a mottling area of ​​the window region is determined based on the window region including at least a portion of the mirroring edge region.

[0069] Exemplarily, a more accurate window region mottling area is obtained by calculating a window region mottling area based on the window region including at least a portion of the mirroring edge region.

[0070] In step S540, a dilation process is performed on the edge of the region of interest to determine a dilated edge region.

[0071] For example, by performing a dilation process on the edge of the region of interest, the value of the leopard-spot density or the value of the leopard-spot percentage in the edge region can be prevented from becoming small, and thus the deviation can be reduced.

[0072] In step S550, a leopard print area of ​​the window region is determined based on the window region including at least a portion of the enlarged edge region.

[0073] For example, a more accurate window region mottling area is obtained by calculating a window region mottling area based on the window region including at least a portion of the enlarged edge region.

[0074] The embodiment of the present disclosure can reduce the deviation caused by the density of leopard spots or the proportion of leopard spots in the edge area being too small by performing optimization processing on the window area including the pixels of the edge area of ​​the region of interest. Therefore, the embodiment of the present disclosure can obtain more accurate leopard spots characteristics of the window area, which can improve the accuracy of subsequent processing, improve the accuracy of the visualized image, and provide a reference to help doctors judge the severity of fundus leopard spots and the severity of myopic fundus changes.

[0075] In some embodiments, the first visualized image includes a two-dimensional first visualized image or a three-dimensional first visualized image, where the two-dimensional first visualized image includes a heat map and the three-dimensional first visualized image includes a contour topographic map. Determining the first visualized image corresponding to the fundus image to be processed based on the digital image includes mapping the digital image into a two-dimensional color space to determine the two-dimensional first visualized image, or mapping the digital image into a three-dimensional space to determine the three-dimensional first visualized image, or mapping the digital image into the two-dimensional color space and then mapping the two-dimensional color space into the three-dimensional space to determine the three-dimensional first visualized image.

[0076] Exemplarily, the digital image is mapped to a two-dimensional color space with a color mapping algorithm (e.g., a JET color mapping algorithm) to determine a two-dimensional first visualization (e.g., a heat map), or the digital image is processed with a mapping tool to map to a two-dimensional color space to determine a two-dimensional first visualization, or the numbers in the digital image are mapped as grayscale values ​​to the two-dimensional color space to determine a two-dimensional visualization.

[0077] Illustratively, the digital image is mapped into a three-dimensional space to determine a three-dimensional first visualization (eg, a contour topographical map).

[0078] Fig. 6 is a flowchart of another method for visualizing changes in myopic fundus according to an embodiment of the present disclosure. The embodiment shown in Fig. 6 is obtained by expanding the embodiment shown in Fig. 2, and the following description will focus on the differences between the embodiment shown in Fig. 6 and the embodiment shown in Fig. 2, and will omit a description of the similarities.

[0079] In another embodiment of the present disclosure, the fundus feature image further includes an atrophic spot area, and as shown in FIG. 6, another method for visualizing myopic fundus changes according to another embodiment of the present disclosure further includes the following steps after the step of determining a fundus feature image based on the fundus image to be processed (i.e., step S210 according to the embodiment shown in FIG. 2).

[0080] In step S610, an atrophic spot area in the fundus characteristic image is determined based on the fundus characteristic image.

[0081] For example, according to the principle of fundus disease, generally, leopard-spotted spots appear first, and with the progression of the disease, the leopard-spotted spots gradually disappear after becoming severe to a certain extent, and atrophic spots begin to appear instead of the leopard-spotted spots, that is, the atrophic spots cover the leopard-spotted spots after appearing, and the severity of the atrophic spots is higher than that of the leopard-spotted spots.

[0082] Illustratively, segment and extract features based on the fundus feature image to determine the atrophic spot area in the fundus feature image. After performing step S610, perform step S620 and / or step S630, that is, step S620 and step S630 can be selectively performed or simultaneously performed according to needs.

[0083] In step S620, a second visualized image corresponding to the fundus image to be processed is determined based on the first visualized image and the atrophic spot area.

[0084] For example, a digital image corresponding to the fundus image to be processed is determined based on the atrophic spot area in the fundus feature image. The numbers in the areas corresponding to the atrophic spots in the digital image indicate the distribution of the atrophic spots. The atrophic spot area is represented in a color different from the leopard-striped spots, and the atrophic spot area is superimposed on the first visualized image to determine the second visualized image. The second visualized image is used to represent the distribution of the leopard-striped spots and the atrophic spots.

[0085] In some embodiments, a digital image corresponding to the fundus image to be processed is determined according to the atrophic spot area in the fundus feature image, and a second visualized image including only the atrophic spot area is generated. The second visualized image can be generated directly based on the atrophic spot area. The second visualized image is used to represent the distribution of the atrophic spot.

[0086] In step S630, a second visualized image is determined based on the first visualized image, the atrophic spot region, and the leopard-spot characteristic region.

[0087] For example, a two-dimensional image including the atrophic spot region and the leopard-striped spot characteristic region is generated by a mapping process based on digital images corresponding to the atrophic spot region and the leopard-striped spot characteristic region. A second visualized image is generated by superimposing the two-dimensional image and the first visualized image. The second visualized image can represent the distribution of the leopard-striped spots and / or the atrophic spots.

[0088] According to the principle of fundus lesions, leopard spots appear first, and after the leopard spots become severe to a certain extent, they are covered by atrophic spots. That is, when the number of leopard spots in the fundus image to be processed decreases, the severity becomes severe. In order to suppress errors in the image processing results, the embodiment of the present disclosure generates a second visualized image by combining the first visualized image and the atrophic spots. The second visualized image can represent the distribution of the leopard spots and the atrophic spots. Therefore, a doctor can diagnose the leopard spots and the atrophic spots by the second visualized image, and the diagnosis and dissemination of the fundus leopard spots and the atrophic spots can be further promoted.

[0089] Fig. 7 is a flowchart of another method for visualizing changes in myopic fundus according to an embodiment of the present disclosure. Since the embodiment shown in Fig. 7 is obtained by expanding the embodiment shown in Fig. 2, the following description will focus on the differences between the embodiment shown in Fig. 7 and the embodiment shown in Fig. 2, and will omit a description of the similarities.

[0090] In an embodiment of the present disclosure, the fundus feature image further includes at least one of an atrophic arc region, a macular region, and an optic disc region. As shown in Fig. 7, another method for visualizing myopic fundus changes according to another embodiment of the present disclosure further includes the following steps after the step of determining a fundus feature image based on a fundus image to be processed (i.e., step S210 according to the embodiment shown in Fig. 2).

[0091] In step S710, atrophic arc regions in the fundus feature images are determined based on the fundus feature images.

[0092] For example, based on the optic disc in the fundus characteristic image, an atrophic arc is extracted, and an atrophic arc region in the fundus characteristic image is determined. Specifically, the optic disc region is fitted to a circle, and the enclosed angle range of the atrophic arc relative to the optic disc is calculated. Here, the enclosed angle is the included angle between the connecting lines between the two end points of the atrophic arc and the center of the fitted circular optic disc region.

[0093] In step S720, a landmark region of the fundus feature image is determined based on at least one of an atrophic arc region, a macular region, and an optic disc region.

[0094] The indicator area is used to present to the user an area on which he or she should focus his or her attention.

[0095] For example, when determining the marking area of ​​the fundus feature image based on the atrophy arc area, the marking area is obtained by marking with the extension line of the enclosing corner, which is the extension line obtained by extending two sides of the enclosing corner.

[0096] For example, when determining the marking region of the fundus feature image based on the macular region, the marking region is obtained by marking the macular region in the fundus feature image.

[0097] For example, when the marking area of ​​the fundus feature image is determined based on the optic disc area, the marking area is obtained by marking the optic disc area in the fundus feature image. When the marking area is determined based on a combination of any two or three of the atrophic arc area, the macular area, and the optic disc area, the marking lines of different areas are made different colors to distinguish different marking areas. The doctor can pay attention to different attention areas by different marking areas.

[0098] In step S730, a third visualized image corresponding to the fundus image to be processed is determined based on the first visualized image and the marked region.

[0099] For example, the mark area is superimposed on the first visualized image to determine a third visualized image corresponding to the fundus image to be processed, the third visualized image including the distribution of the leopard-spot spots in the mark area.

[0100] In some embodiments, after generating a second visualized image based on the first visualized image and the atrophic spot region, a marking region of the fundus feature image is determined based on at least one of the atrophic arc region, the macular region, and the optic disc region. The obtained marking region is superimposed on the second visualized image to determine a third visualized image corresponding to the fundus image to be processed. The third visualized image includes the distribution status of the leopard-striped spots and / or the atrophic spots in the marking region.

[0101] The embodiment of the present disclosure uses the mark area to remind the user to pay attention to the disease status in the mark area. The disease degree in the mark area has a greater impact on the patient than other areas. Therefore, the mark area can promptly inform the user of the focus, reduce the risk of misdiagnosis, and provide reference for early intervention, which is favorable to the establishment of a comprehensive and precise prevention and control of myopia and a corresponding precise prevention and control system.

[0102] 8 is a structural schematic diagram of a visualization device for myopic fundus changes according to an embodiment of the present disclosure. As shown in FIG. 8, a visualization device 800 for myopic fundus changes according to an embodiment of the present disclosure includes a determination module 801 and a visualization module 802. Specifically, the determination module 801 is used to determine a fundus feature image including a leopard-striped feature region based on a fundus image to be processed. The visualization module 802 is used to determine a first visualized image corresponding to the fundus image to be processed based on the fundus feature image, and the first visualized image is used to represent the distribution state of the leopard-striped spot.

[0103] In some embodiments, the visualization module 802 is further configured to determine a digital image corresponding to the fundus image to be processed based on the fundus feature image, where the digital image represents the distribution of the leopard print in digital form, and to determine a first visualization image corresponding to the fundus image to be processed based on the digital image.

[0104] In some embodiments, the visualization module 802 is further configured to: determine a plurality of window regions in the fundus feature image using a predetermined window, where the window region includes at least one pixel; determine a leopard-spot feature of the window region for each of the plurality of window regions; determine a leopard-spot index based on the leopard-spot feature of the window region using an area of ​​the window region or feature data of a region of interest in the fundus feature image; and determine a digital image corresponding to the fundus image to be processed based on the leopard-spot index.

[0105] In some embodiments, the leopard-print feature of the window region includes a leopard-print area of ​​the window region. The visualization module 802 is further configured to perform edge processing on each of the plurality of window regions when the window region includes pixels located at the edge of the region of interest and the area of ​​the window region located outside the region of interest is greater than a predetermined threshold, the edge processing being configured to perform a mirroring process on the pixels of the edge of the region of interest based on the edge of the region of interest to obtain a mirrored edge area, and determine a leopard-print area of ​​the window region based on the window region including at least a part of the mirrored edge area; or to perform an enlargement process on the edge of the region of interest to determine an enlarged edge area, and determine a leopard-print area of ​​the window region based on the window region including at least a part of the enlarged edge area, when the window region includes pixels located at the edge of the region of interest. The predetermined threshold may be 0% to 50% of the area of ​​the window region, and more specifically, may be 10% of the area of ​​the window region, 20% of the area of ​​the window region, 5% of the area of ​​the window region, and the like.

[0106] In some embodiments, the first visualized image includes a two-dimensional first visualized image or a three-dimensional first visualized image, and the visualization module 802 is further configured to map the digital image into a two-dimensional color space to determine the two-dimensional first visualized image, or to map the digital image into a three-dimensional space to determine the three-dimensional first visualized image, or to map the digital image into the two-dimensional color space and then map the two-dimensional color space into the three-dimensional space to determine the three-dimensional first visualized image.

[0107] In some embodiments, the fundus feature image further includes an atrophic spot region, and the visualization module 802 is further configured to, after determining the fundus feature image based on the fundus image to be processed, determine an atrophic spot region in the fundus feature image based on the fundus feature image, and determine a second visualization image corresponding to the fundus image to be processed based on the first visualization image and the atrophic spot region, or determine the second visualization image based on the first visualization image, the atrophic spot region, and the second leopard-striped spot feature region.

[0108] In some embodiments, the fundus feature image further includes at least one of an atrophic arc region, a macular region, and an optic disc region, and the visualization module 802 is further configured to, after determining the fundus feature image based on the fundus image to be processed, determine an atrophic arc region in the fundus feature image based on the fundus feature image, determine a marked region of the fundus feature image based on at least one of the atrophic arc region, the macular region, and the optic disc region, wherein the marked region represents an area that is presented to a user as requiring special attention, and determine a third visualized image corresponding to the fundus image to be processed based on the first visualized image and the marked region.

[0109] 9 is a structural schematic diagram of an electronic device according to an embodiment of the present disclosure. The electronic device 900 (which may specifically be a computer device) includes a memory 901, a processor 902, a communication interface 903, and a bus 904. The memory 901, the processor 902, and the communication interface 903 communicate with each other via the bus 904.

[0110] The memory 901 may be a read only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 901 can store a program, and the processor 902 and the communication interface 903 execute each step in the method for visualizing myopic fundus changes according to the embodiment of the present disclosure when the program stored in the memory 901 is executed by the processor 902.

[0111] The processor 902 may be a general-purpose central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), a graphics processing unit (GPU), or one or more integrated circuits, and is configured to execute such programs to realize the functions handled by each unit of the myopic fundus change visualization device of the embodiment of the present disclosure.

[0112] The processor 902 may be an integrated circuit chip having signal processing capabilities. In the process of implementation, each step of the visualization method of myopic fundus change of the present disclosure can be completed by an integrated logic circuit of hardware in the processor 902 or an instruction in the form of software. The processor 902 may be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, and can realize or execute each method, step and logic block diagram according to the embodiment of the present disclosure. The general-purpose processor may be a microprocessor, any general processor, etc. The steps of the method described in relation to the embodiment of the present disclosure may be executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in memory 901, and the processor 902 reads the information in the memory 901 and cooperates with the hardware to complete the functions handled by the units included in the myopic fundus change visualization device according to the embodiment of the present disclosure, or executes the myopic fundus change visualization method according to the embodiment of the present disclosure.

[0113] The communication interface 903 uses a transceiver device, such as, but not limited to, a transceiver, to realize communication between the electronic device 900 and other devices or a communication network. For example, a fundus image to be processed can be acquired through the communication interface 903.

[0114] The bus 904 may include a path for transmitting information between each component of the electronic device 900 (eg, the memory 901, the processor 902, and the communication interface 903).

[0115] In addition, in the electronic device 900 shown in FIG. 9, only a memory, a processor, and a communication interface are shown, but in a specific implementation process, the electronic device 900 further includes other devices necessary for realizing normal operation, which should be understood by those skilled in the art. In addition, according to specific needs, the electronic device 900 may further include hardware devices for realizing other additional functions, which should be understood by those skilled in the art. In addition, the electronic device 900 may include only devices necessary for realizing the embodiments of the present disclosure, and it is not necessary for the electronic device 900 to include all the devices shown in FIG. 9, which should be understood by those skilled in the art.

[0116] Those skilled in the art may realize that the units and algorithm steps according to each example described in connection with the embodiments described herein can be realized in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to realize the described functions for each specific application, and such realization should not be considered as going beyond the scope of the present disclosure.

[0117] It should be noted that, for convenience and conciseness of description, the specific operation processes of the systems, devices and units described above may refer to the corresponding processes in the method embodiments, and will not be described again here.

[0118] However, in some embodiments according to the present disclosure, the system, device, and method may be realized in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is merely a division of logical functions, and other division methods may be used in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some configurations may be ignored or not implemented. In addition, the couplings, direct couplings, or communication connections between the described or discussed mutually may be indirect couplings or communication connections via some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0119] The units described above as separate components may or may not be physically separated, and the components described as units may or may not be physical units, i.e., they may be located in one place or distributed among multiple network units. According to actual needs, some or all of the units can be selected to achieve the purpose of the solution of the present embodiment.

[0120] Furthermore, each functional unit in each embodiment of the present disclosure may be integrated into a single processing unit, each unit may exist physically alone, or two or more units may be integrated into a single unit.

[0121] An embodiment of the present disclosure may be a computer-readable storage medium on which computer program instructions are stored that, when executed by a processor, cause the processor to execute steps in the methods according to various embodiments of the present disclosure described herein. The functions may be realized as software functional units and stored in a computer-readable storage medium when sold or used as an independent product. Based on this view, the technical solution of the present disclosure may be substantially embodied in the form of a software product, or a part of the contribution to the prior art or the technical solution may be embodied in the form of a software product. The computer software product is stored in one storage medium and includes some instructions for causing one computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or some steps of the methods described in each embodiment of the present disclosure. The storage medium includes various media that can store program code, such as a USB disk, a portable hard disk, a read-only memory, a random access memory, a magnetic disk, or an optical disk. The computer-readable storage medium may utilize any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof.

[0122] Although only specific embodiments of the present disclosure have been described above, the scope of protection of the present disclosure is not limited thereto. Any modifications or replacements that a person skilled in the art can easily think of within the technical scope described in the present disclosure should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure is subject to the scope of protection of the claims.

Claims

1. A method for visualizing changes in myopic fundus, comprising: determining a fundus characteristic image including a leopard-striped feature region based on the fundus image to be processed; determining a first visualized image corresponding to the fundus image to be processed based on the fundus feature image, the first visualized image being used to represent a distribution state of leopard-spot spots; A method for visualizing changes in myopic fundus comprising:

2. The step of determining a first visualized image corresponding to the target fundus image based on the fundus feature image includes: determining a digital image corresponding to the target fundus image based on the fundus feature image, the digital image representing a distribution state of the leopard-spot spots in a number format; determining a first visualized image corresponding to the target fundus image based on the digital image; The method for visualizing changes in myopic fundus according to claim 1 .

3. The step of determining a digital image corresponding to a target fundus image based on the fundus feature image includes: determining a plurality of window regions in the fundus feature image using a predetermined window, the window regions including at least one pixel; determining, for each of the plurality of window regions, a leopard print characteristic of the window region; determining a leopard-spot index based on the leopard-spot feature of the window region by using an area of ​​the window region or feature data of a region of interest in the fundus feature image; determining a digital image corresponding to the target fundus image based on the leopard marking index; The method for visualizing changes in myopic fundus according to claim 2 .

4. The step of determining, for each of the plurality of window regions, a leopard print characteristic of the window region includes: performing edge processing on each of the plurality of window regions when the window region includes pixels located at an edge of the region of interest and an area of ​​the window region located outside the region of interest is greater than a predetermined threshold, the edge processing including performing a mirroring process on pixels at the edge of the region of interest based on the edge of the region of interest to obtain a mirrored edge region; determining a leopard print characteristic of the window area based on the window area including at least a portion of the mirrored edge area; or or, for each of the plurality of window regions, if the window region includes a pixel located at an edge of the region of interest, performing a dilation process on the edge of the region of interest to determine a dilated edge region; determining a leopard print characteristic of the window area based on the window area including at least a portion of the enlarged edge area; The method for visualizing changes in myopic fundus according to claim 3 .

5. The first visualized image includes a two-dimensional first visualized image or a three-dimensional first visualized image; The step of determining a first visualized image corresponding to the target fundus image based on the digital image includes: Mapping the digital image into a two-dimensional color space to determine the two-dimensional first visualization; or Mapping the digital image into a three-dimensional space to determine the three-dimensional first visualization; or mapping the digital image into a two-dimensional color space and then mapping the two-dimensional color space into a three-dimensional space to determine the three-dimensional first visualization; The method for visualizing changes in myopic fundus according to claim 2 .

6. The fundus characteristic image further includes an atrophic spot region, After the step of determining a fundus feature image based on the target fundus image, determining an atrophic spot region in the fundus characteristic image based on the fundus characteristic image; determining a second visualized image corresponding to the fundus image to be processed based on the first visualized image and the atrophic spot region, or determining the second visualized image based on the first visualized image, the atrophic spot region, and the leopard-striped spot characteristic region. The method for visualizing changes in myopic fundus according to any one of claims 1 to 5.

7. The fundus feature image further includes at least one of an atrophic arc region, a macular region, and an optic disc region; After the step of determining a fundus feature image based on the target fundus image, determining an atrophic arc region in the fundus characteristic image based on the fundus characteristic image; determining a mark area of ​​the fundus feature image based on at least one of the atrophy arc area, the macular area, and the optic disc area, the mark area representing an area that is presented to a user as a region that should be focused on; determining a third visualized image corresponding to the processing target fundus image based on the first visualized image and the marking region. The method for visualizing changes in myopic fundus according to any one of claims 1 to 5.

8. A visualization device for myopic fundus changes, comprising: A determination module configured to determine a fundus feature image including a leopard print feature region based on the target fundus image; a visualization module for determining a first visualized image corresponding to the fundus image to be processed based on the fundus feature image, the first visualized image being configured to represent a distribution state of leopard-spot spots; A visualization device for myopic fundus changes.

9. An electronic device, A processor configured to execute the method for visualizing myopic fundus changes according to any one of claims 1 to 5; and a memory configured to store instructions executable by the processor.

1. An electronic device comprising:

10. A computer program for executing the method for visualizing changes in myopic fundus according to any one of claims 1 to 5 is stored. A computer-readable storage medium comprising:

Images