Visualization method and device for fundus abnormality

Abstract

The invention provides a visualization method and device for fundus abnormality. One embodiment of the method comprises the following steps: firstly, carrying out identification processing on an abnormal region in a target fundus image to obtain an abnormal classification result; and secondly, on the basis of the anomaly classification result, performing differential display on all the anomaly regions in the target fundus image according to anomaly types, and generating a fundus anomaly type visualization result. Therefore, by automatically identifying all the fundus abnormalities, errors caused by human factors are reduced, and the accuracy of fundus abnormality diagnosis is improved; according to the method, the recognized anomaly classification result is visually displayed, a doctor can be helped to more visually explain the fundus anomaly condition to a patient, the patient can better understand the illness state of the patient, and the effectiveness of doctor-patient communication is enhanced.

Description

A visualization method and device for fundus abnormalities Technical Field

[0001] This invention belongs to the field of image processing technology, and in particular relates to a visualization method and device for fundus abnormalities. Background Technology

[0002] Fundus abnormalities refer to a wide range of abnormalities occurring in various structures of the eye. The main symptoms include decreased vision, distorted vision, visual field defects, floaters, pain, and photophobia.

[0003] Current technologies only display single-disease abnormalities in the fundus, lacking a method for labeling all fundus abnormalities. This is primarily because multiple fundus lesions may coexist and influence each other, potentially leading to unclear labeling. Even if current technologies can label various fundus abnormalities manually, the labeling results for the same fundus image can still vary significantly due to the influence of different doctors' experience.

[0004] Therefore, there is an urgent need to provide a visualization method for fundus abnormalities to solve the technical problem that the limitations of existing technology prevent patients or medical staff in other departments from having a clear and intuitive understanding of fundus abnormalities. Summary of the Invention

[0005] To address the aforementioned problems in existing technologies, embodiments of the present invention provide a method and apparatus for visualizing fundus abnormalities. This method can accurately identify all fundus abnormalities and provide an intuitive visualization of the identified abnormality classification results.

[0006] According to a first aspect of the present invention, a visualization method for fundus abnormalities is provided. The method includes: identifying abnormal regions in a target fundus image to obtain an abnormality classification result; and, based on the abnormality classification result, displaying all abnormal regions in the target fundus image differently according to the type of abnormality to generate a visualization result of fundus abnormality types.

[0007] Optionally, the step of identifying abnormal regions in the target fundus image to obtain anomaly classification results includes: performing fundus anomaly detection on the target fundus image and generating fundus anomaly detection results; if the fundus anomaly detection results indicate that there is an abnormal region in the target fundus image and the abnormal region is a fundus anomaly, then performing fundus anomaly classification processing on the abnormal region in the target fundus image and generating fundus anomaly classification results; if the fundus anomaly detection results indicate that there is an abnormal region in the target fundus image and the abnormal region is not a fundus anomaly, then performing non-fundus anomaly classification processing on the abnormal region in the target fundus image and generating non-fundus anomaly classification results; and determining the fundus anomaly classification results and the non-fundus anomaly classification results as the anomaly classification results for the abnormal region.

[0008] Optionally, the method further includes: if the fundus abnormality classification result indicates that the abnormal area is a lesion, then the abnormal area is subjected to lesion attribute classification processing to generate a lesion attribute classification result; if the fundus abnormality classification result indicates that the abnormal area is an area of ​​human intervention, then the abnormal area is subjected to human interference factor classification processing to generate a human intervention classification result; and the lesion attribute classification result and the human intervention classification result are determined as the abnormality cause classification result in the fundus abnormality classification result.

[0009] Optionally, the method further includes: if the lesion attribute classification result indicates that the lesion is a congenital lesion, then the abnormal area is subjected to congenital lesion classification processing to generate a congenital lesion classification result; if the lesion attribute classification result indicates that the lesion is a non-congenital lesion, then the abnormal area is subjected to non-congenital lesion classification processing to generate a non-congenital lesion classification result; and the congenital lesion classification result and the non-congenital lesion classification result are determined as the lesion classification result in the lesion attribute classification result.

[0010] Optionally, the method further includes: if the non-congenital lesion classification result indicates that the non-congenital lesion is a myopia disease, then by determining whether there are leopard spots and lesions outside the atrophic arc in the abnormal area, the abnormal area is processed for myopia disease classification, and a myopia disease classification result is generated; if the non-congenital lesion classification result indicates that the non-congenital lesion is a non-myopia disease, then the abnormal area is processed for non-myopia disease classification, and a non-myopia disease classification result is output; the myopia disease classification result and the non-myopia disease classification result are determined as the disease classification result in the non-congenital lesion classification result.

[0011] Optionally, the visualization results of fundus abnormality types include: a fundus abnormality display image and / or fundus abnormality descriptive text; the fundus abnormality display image is used to show the distribution location, shape, type, size, and distribution of the fundus abnormality; the fundus abnormality descriptive text includes at least one of the following: type, name, quantity information, area information, coordinate information, and orientation information of the fundus abnormality.

[0012] Optionally, the method further includes: classifying the abnormal regions corresponding to each abnormal classification result according to their severity into abnormal level categories, and outputting abnormal level classification results; and displaying different abnormal regions corresponding to the same abnormal classification result in a differentiated manner according to the abnormal level classification results to generate a visualization result of fundus abnormality levels.

[0013] Optionally, the abnormality classification results include fundus abnormality classification results and non-fundus abnormality classification results; the abnormal regions corresponding to the fundus abnormality classification results are classified into abnormality levels according to their severity, and fundus abnormality level classification results are output; wherein, the fundus abnormality level classification results include three cases, namely mild fundus abnormality classification results, moderate fundus abnormality classification results, and severe fundus abnormality classification results; based on the non-fundus abnormality classification results, the mild fundus abnormality classification results, the moderate fundus abnormality classification results, and the severe fundus abnormality classification results, all abnormal regions in the target fundus image are displayed differentially to generate fundus abnormality level visualization results.

[0014] Optionally, by assigning a differentiated identifier to the abnormal region corresponding to each of the abnormal classification results, all abnormal regions in the target fundus image are visualized to generate a fundus abnormality type visualization result; for any of the abnormal classification results: based on the differentiated identifier corresponding to the abnormal classification result, the abnormality level classification result is differentiated by the gradient change of the identifier attribute to generate a fundus abnormality level visualization result.

[0015] Optionally, the visualization results of the fundus abnormality level can be displayed through a fundus abnormality heat map.

[0016] According to a second aspect of the present invention, a visualization device for fundus abnormalities is also provided, comprising: an identification module for identifying abnormal regions in a target fundus image to obtain an abnormality classification result; and a classification differentiation display module for differentiating all abnormal regions in the target fundus image according to the abnormality type based on the abnormality classification result to generate a visualization result of fundus abnormality types.

[0017] According to a third aspect of the present invention, an electronic device is also provided, the electronic device comprising: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the method as described in the first aspect.

[0018] According to a fourth aspect of the present invention, a computer-readable medium is also provided, on which a computer program is stored, wherein the program, when executed by a processor, implements the method described in the first aspect.

[0019] This invention provides a visualization method and apparatus for fundus abnormalities. One embodiment of the method includes: first, identifying abnormal regions in a target fundus image to obtain an abnormality classification result; second, based on the abnormality classification result, differentially displaying all abnormal regions in the target fundus image according to the type of abnormality, generating a visualization result of fundus abnormality types. Thus, by automatically identifying all fundus abnormalities, errors caused by human factors are reduced, improving the accuracy of fundus abnormality diagnosis. Furthermore, by intuitively visualizing the identified abnormality classification results, doctors can more clearly explain the fundus abnormalities to patients, helping patients better understand their condition and enhancing the effectiveness of doctor-patient communication. Attached Figure Description

[0020] The following sections will describe some specific embodiments of the present invention in a detailed manner by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings indicate the same or similar parts or components. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 is a flowchart illustrating a visualization method for fundus abnormalities according to an embodiment of the present invention; Figure 2 is a flowchart illustrating the identification and processing of abnormal regions in a target fundus image according to an embodiment of the present invention; Figure 3 is a structural schematic diagram of a visualization device for fundus abnormalities according to an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0022] Figure 1 shows a flowchart of a visualization method for fundus abnormalities provided in an embodiment of the present invention.

[0023] A visualization method for fundus abnormalities, comprising at least the following steps: S101, identifying abnormal regions in a target fundus image to obtain an abnormality classification result; S102, based on the abnormality classification result, displaying all abnormal regions in the target fundus image differently according to the type of abnormality to generate a visualization result of fundus abnormality types.

[0024] In S101, "abnormal area" refers to an area formed by fundus abnormalities or non-fundus abnormalities; "fundus abnormality" refers to fundus abnormalities caused by congenital lesions, myopia, non-myopia, or human intervention. In other words, "abnormal area" refers to a fundus abnormality area or a non-fundus abnormality area; "fundus abnormality area" refers to a congenital lesion area, a myopia area, a non-myopia area, or a human intervention area.

[0025] Abnormal regions in the target fundus image are identified and processed based on a model or preset rules to obtain abnormal classification results. Here, abnormal classification results include, but are not limited to, classification results for congenital lesions, classification results for myopia, classification results for non-myopia, classification results for human intervention, and classification results for non-fundus abnormalities.

[0026] The classification results for non-fundus abnormalities include: vitreous opacities, silicone oil, and others; the classification results for artificial interventions include: laser spots and others; the classification results for congenital lesions include: myelinated nerve fibers, retinitis pigmentosa, and others; the classification results for myopia include: leopard spots and lesions other than atrophic arcs, and others; the classification results for non-myopia include: diabetes, hypertension, macular degeneration, optic disc disease, and others.

[0027] In S102, there are no restrictions on the method of differential display. Differential display can be based on color to differentiate the abnormal areas corresponding to different abnormal classification results, or it can be implemented based on other methods.

[0028] The visualization results of fundus abnormalities include: a fundus abnormality display image and / or fundus abnormality description text; the fundus abnormality display image is used to show the distribution location, shape, type, size, and distribution of the fundus abnormality; the fundus abnormality description text includes at least one of the following: type, name, quantity information, area information, coordinate information, and orientation information of the fundus abnormality.

[0029] This embodiment automatically identifies all fundus abnormalities, reducing errors caused by human factors and improving the accuracy of fundus abnormality diagnosis. By providing an intuitive visualization of the identified abnormality classification results, doctors can more clearly explain the fundus abnormalities to patients, helping patients better understand their condition and enhancing the effectiveness of doctor-patient communication.

[0030] In addition, the method in this embodiment is implemented through automated software, which greatly simplifies the diagnostic process, improves work efficiency, and reduces the workload of doctors.

[0031] In a preferred embodiment of this example, the abnormal regions corresponding to each abnormal classification result are classified according to their severity, and the abnormal classification results are output. Different abnormal regions corresponding to the same abnormal classification result are displayed differently according to the abnormal classification results to generate a visualization result of fundus abnormality level.

[0032] For example, the abnormality classification results include: non-fundus abnormality level, mild fundus abnormality level, moderate fundus abnormality level, and severe fundus abnormality level.

[0033] If the congenital lesion classification result is retinitis pigmentosa, the abnormal regions corresponding to retinitis pigmentosa include a first abnormal region, a second abnormal region, and a third abnormal region. The first abnormal region represents mild retinitis pigmentosa, the second abnormal region represents moderate retinitis pigmentosa, and the third abnormal region represents severe retinitis pigmentosa. Here, the first, second, and third abnormal regions can be different abnormal regions from the same target fundus image, or they can be abnormal regions from different target fundus images.

[0034] To provide a more intuitive visualization of fundus abnormality types and levels, in a preferred embodiment of this invention, a differentiated identifier is assigned to each abnormal region corresponding to the abnormality classification result to visualize all abnormal regions in the target fundus image, generating a fundus abnormality type visualization result. For any abnormality classification result: based on the differentiated identifier corresponding to the abnormality classification result, the abnormality level classification result is differentiated by gradient changes in the identifier attributes, generating a fundus abnormality level visualization result. For example, the differentiated identifier includes, but is not limited to, color and / or height.

[0035] For example, visualization results of fundus abnormality types and fundus abnormality grades can be displayed simultaneously on a single image or on different images. When a target fundus image contains a first abnormality region corresponding to mild retinitis pigmentosa and a second abnormality region corresponding to severe laser spots, since the labeling color for retinitis pigmentosa is blue and the labeling color for severe laser spots is red, the first abnormality region corresponding to mild retinitis pigmentosa will be displayed in light blue, and the second abnormality region corresponding to severe laser spots will be displayed in dark red.

[0036] This embodiment uses different colors and shades to mark fundus abnormalities, making the location and severity of abnormalities immediately apparent, greatly improving the intuitiveness and comprehensibility of the diagnostic results.

[0037] To enable patients or medical personnel to have a clear understanding of the severity of the target fundus image, in a preferred embodiment of this invention, the abnormality classification result includes fundus abnormality classification result and non-fundus abnormality classification result. The abnormal regions corresponding to the fundus abnormality classification result are classified according to their severity, and a fundus abnormality level classification result is output. The fundus abnormality level classification result includes three scenarios: mild fundus abnormality classification result, moderate fundus abnormality classification result, and severe fundus abnormality classification result. Based on the non-fundus abnormality classification result, the mild fundus abnormality classification result, the moderate fundus abnormality classification result, and the severe fundus abnormality classification result, all abnormal regions in the target fundus image are displayed differentially, generating a fundus abnormality level visualization result.

[0038] For example, based on different abnormality levels, the first abnormal region corresponding to the non-fundus abnormality classification result, the second abnormal region corresponding to the mild fundus abnormality classification result, the third abnormal region corresponding to the moderate fundus abnormality classification result, and the fourth abnormal region corresponding to the severe fundus abnormality classification result are marked with different colors for differentiated display, generating a fundus abnormality level visualization result. Alternatively, based on different abnormality levels, the first abnormal region corresponding to the non-fundus abnormality classification result, the second abnormal region corresponding to the mild fundus abnormality classification result, the third abnormal region corresponding to the moderate fundus abnormality classification result, and the fourth abnormal region corresponding to the severe fundus abnormality classification result are displayed differently using the same marking color, generating a fundus abnormality level visualization result.

[0039] This embodiment employs a unified color coding standard, which helps improve diagnostic consistency among different doctors or medical institutions and ensures the reliability of diagnostic results. The varying shades of color visually indicate the severity of abnormalities, helping doctors develop more precise treatment plans and providing strong support for patient care.

[0040] The visualization results of fundus abnormality levels can be two-dimensional or three-dimensional graphs; two-dimensional graphs include: heatmaps, bar charts, scatter plots, area charts, radar charts, bubble charts, pie charts, etc. In order to clearly display the visualization results of fundus abnormality levels, in a preferred embodiment of this example, the visualization results of fundus abnormality levels are displayed using a fundus abnormality heatmap.

[0041] For example: determine the abnormal gray value of the fundus corresponding to each abnormal region in the target fundus image, map the abnormal gray value of the fundus to the color space, and generate a fundus abnormality heat map corresponding to the target fundus image.

[0042] It should be noted that, depending on the actual needs of patients or medical staff, the different output methods mentioned above can be selected for the visualization results of fundus abnormality levels. This embodiment's method is not only applicable to clinical diagnosis but can also serve as a teaching tool, helping ophthalmologists and students better understand and master the characteristics of fundus abnormalities. This method facilitates telemedicine services, enabling remote doctors to quickly and accurately assess a patient's fundus condition by remotely transmitting color-coded fundus images.

[0043] Figure 2 shows a flowchart of the process for identifying abnormal regions in a target fundus image according to an embodiment of the present invention.

[0044] The method involves identifying and processing abnormal regions in a target fundus image, including at least the following steps: S201, performing fundus abnormality detection on the target fundus image and generating fundus abnormality detection results; if the fundus abnormality detection results indicate the presence of an abnormal region in the target fundus image and the abnormal region is a fundus abnormality, then performing fundus abnormality classification processing on the abnormal region in the target fundus image and generating fundus abnormality classification results; if the fundus abnormality detection results indicate the presence of an abnormal region in the target fundus image and the abnormal region is not a fundus abnormality, then performing non-fundus abnormality classification processing on the abnormal region in the target fundus image and generating non-fundus abnormality classification results; S202, if the fundus abnormality classification results indicate that the abnormal region is a region of human intervention, then performing human interference factor classification processing on the abnormal region and generating human intervention classification results; if the fundus abnormality classification results indicate that the abnormal region is a lesion, then performing lesion attribute classification processing on the abnormal region and generating lesion attribute classification results. S203, If the lesion attribute classification result indicates that the lesion is a congenital lesion, then the abnormal area is processed for congenital lesion classification, and a congenital lesion classification result is generated; if the lesion attribute classification result indicates that the lesion is a non-congenital lesion, then the abnormal area is processed for non-congenital lesion classification, and a non-congenital lesion classification result is generated; S204, If the non-congenital lesion classification result indicates that the non-congenital lesion is a myopia disease, then the abnormal area is processed for myopia disease classification by determining whether there are leopard spots and lesions outside the atrophy arc, and a myopia disease classification result is generated; if the non-congenital lesion classification result indicates that the non-congenital lesion is a non-myopia disease, then the abnormal area is processed for non-myopia disease classification, and a non-myopia disease classification result is output; S205, The non-fundus abnormality classification result, the artificial intervention classification result, the congenital lesion classification result, the myopia disease classification result, and the non-myopia disease classification result are determined as the abnormal classification result of the abnormal area.

[0045] For example: Fundus anomaly detection is performed on the target fundus image based on an existing model or preset rules. If the fundus anomaly detection result indicates the presence of an abnormal region in the target fundus image, and the abnormal region is not a fundus anomaly, then the abnormal region in the target fundus image is classified as a non-fundus anomaly based on the existing model or preset rules, generating a non-fundus anomaly classification result. The non-fundus anomaly classification result may be vitreous opacity, silicone oil abnormality, or other abnormalities. If the fundus anomaly detection result indicates the presence of an abnormal region in the target fundus image, and the abnormal region is a fundus anomaly, then the abnormal region in the target fundus image is classified as a fundus anomaly based on the existing model or preset rules, generating a fundus anomaly classification result.

[0046] If the fundus abnormality classification result indicates that the abnormal area is an area of ​​artificial intervention, then the abnormal area is processed for artificial interference factors based on existing models or preset rules to generate an artificial intervention classification result; wherein, the artificial intervention classification result is a laser spot or other. If the fundus abnormality classification result indicates that the abnormal area is a lesion, then the abnormal area is processed for lesion attribute classification based on existing models or preset rules to generate a lesion attribute classification result.

[0047] If the lesion attribute classification result indicates that the lesion is a congenital lesion, then the abnormal area is classified as a congenital lesion based on the existing model or preset rules to generate a congenital lesion classification result; wherein, the congenital lesion classification result is myelinated nerve fiber, retinitis pigmentosa, or other; if the lesion attribute classification result indicates that the lesion is a non-congenital lesion, then the abnormal area is classified as a non-congenital lesion based on the existing model or preset rules to generate a non-congenital lesion classification result.

[0048] If the classification result of the non-congenital lesion indicates that the non-congenital lesion is a myopia condition, then it is determined whether there are leopard spots and lesions outside the atrophic arc in the abnormal area; if so, the leopard spots and lesions outside the atrophic arc are classified as myopia conditions; if the classification result of the non-congenital lesion indicates that the non-congenital lesion is a non-myopia condition, then the abnormal area is classified as a non-myopia condition based on the existing model and preset rules, and the non-myopia condition classification result is output; wherein, the non-myopia condition classification result is diabetes, hypertension, macular degeneration, optic disc disease, or others.

[0049] The logical judgment algorithm in this embodiment can accurately identify the type and severity of abnormalities in the target fundus image. This not only reduces errors caused by human factors and improves the accuracy of fundus diagnosis, but also helps doctors explain fundus abnormalities to patients more intuitively, which helps patients better understand their condition and enhances the effectiveness of doctor-patient communication.

[0050] It should be understood that, in the various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0051] Figure 3 shows a schematic diagram of the structure of a multimodal fundus image classification device provided in an embodiment of the present invention.

[0052] A visualization device for fundus abnormalities, the device 300 includes: a recognition module 301, used to identify abnormal regions in a target fundus image to obtain an abnormality classification result; and a classification differentiation display module 302, used to differentiate all abnormal regions in the target fundus image according to the abnormality type based on the abnormality classification result, and generate a visualization result of fundus abnormality types.

[0053] In a preferred embodiment of this example, the identification module includes: a detection unit, configured to perform fundus anomaly detection on a target fundus image and generate a fundus anomaly detection result; a first generation unit, configured to perform fundus anomaly classification processing on the abnormal region in the target fundus image and generate a fundus anomaly classification result if the fundus anomaly detection result indicates that there is an abnormal region in the target fundus image and the abnormal region is a fundus anomaly; a second generation unit, configured to perform non-fundus anomaly classification processing on the abnormal region in the target fundus image and generate a non-fundus anomaly classification result if the fundus anomaly detection result indicates that there is an abnormal region in the target fundus image and the abnormal region is not a fundus anomaly; and a first determination unit, configured to determine the fundus anomaly classification result and the non-fundus anomaly classification result as the anomaly classification result of the abnormal region.

[0054] In a preferred embodiment of this example, the first generation unit includes: a lesion attribute classification processing unit, used to perform lesion attribute classification processing on the abnormal area and generate a lesion attribute classification result if the fundus abnormality classification result indicates that the abnormal area is a lesion; a human interference factor classification processing unit, used to perform human interference factor classification processing on the abnormal area and generate a human intervention classification result if the fundus abnormality classification result indicates that the abnormal area is a human intervention area; and a determination subunit, used to determine the lesion attribute classification result and the human intervention classification result as the abnormal cause classification result in the fundus abnormality classification result.

[0055] In a preferred embodiment of this example, the lesion attribute classification processing unit includes: a congenital lesion classification processing unit, used to perform congenital lesion classification processing on the abnormal area and generate a congenital lesion classification result if the lesion attribute classification result indicates that the lesion is a congenital lesion; a non-congenital lesion classification processing unit, used to perform non-congenital lesion classification processing on the abnormal area and generate a non-congenital lesion classification result if the lesion attribute classification result indicates that the lesion is a non-congenital lesion; and a determination unit, used to determine the congenital lesion classification result and the non-congenital lesion classification result as the lesion classification result in the lesion attribute classification result.

[0056] In a preferred embodiment of this example, the visualization result of fundus abnormality types includes: a fundus abnormality display image and / or fundus abnormality descriptive text; the fundus abnormality display image is used to show the distribution location, shape, type, size, and distribution of the fundus abnormality; the fundus abnormality descriptive text includes at least one of the following: type, name, quantity information, area information, coordinate information, and orientation information of the fundus abnormality.

[0057] In a preferred embodiment of this example, the non-congenital lesion classification processing unit includes: a myopia classification processing unit, used to classify the abnormal area as myopia by determining whether there are leopard spots and lesions other than the atrophic arc in the abnormal area if the non-congenital lesion classification result indicates that the non-congenital lesion is a myopia disease, and generate a myopia classification result; a non-myopia classification processing unit, used to classify the abnormal area as a non-myopia disease if the non-congenital lesion classification result indicates that the non-congenital lesion is a non-myopia disease, and output a non-myopia classification result; and a determination subunit, used to determine the myopia classification result and the non-myopia classification result as the disease classification result in the non-congenital lesion classification result.

[0058] In a preferred embodiment of this invention, the device further includes: an abnormality level classification processing module, used to classify the abnormal regions corresponding to each abnormality classification result according to their severity and output the abnormality level classification result; and a level differentiation display module, used to differentiate the display of different abnormal regions corresponding to the same abnormality classification result according to the abnormality level classification result, and generate a fundus abnormality level visualization result.

[0059] In a preferred embodiment of this invention, the abnormality classification result includes fundus abnormality classification result and non-fundus abnormality classification result; the device further includes: an abnormality level classification processing module, used to classify the abnormal regions corresponding to the fundus abnormality classification result according to their severity, and output the fundus abnormality level classification result; wherein, the fundus abnormality level classification result includes three cases, namely mild fundus abnormality classification result, moderate fundus abnormality classification result, and severe fundus abnormality classification result; a level differentiation display module, used to differentiate and display all abnormal regions in the target fundus image based on the non-fundus abnormality classification result, the mild fundus abnormality classification result, the moderate fundus abnormality classification result, and the severe fundus abnormality classification result, and generate a fundus abnormality level visualization result.

[0060] In a preferred embodiment of this example, the classification differentiation display module is further configured to assign a differentiation identifier to the abnormal region corresponding to each of the abnormal classification results, so as to visualize all abnormal regions in the target fundus image and generate a fundus abnormality type visualization result; For any of the abnormal classification results: based on the differentiation identifier corresponding to the abnormal classification result, the abnormality level classification result is differentiated by the gradient change of the identifier attribute to generate a fundus abnormality level visualization result.

[0061] In a preferred embodiment of this example, the grade differentiation display module is further used to visualize the results of the fundus abnormality grade visualization through a fundus abnormality heatmap.

[0062] The above-described apparatus can execute the visualization method for fundus abnormalities provided in an embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the visualization method for fundus abnormalities. Technical details not described in detail in this embodiment can be found in the visualization method for fundus abnormalities provided in an embodiment of the present invention.

[0063] The present invention also provides an electronic device, comprising: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the visualization method for fundus abnormalities described in the present invention.

[0064] In addition to the methods and apparatus described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0065] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0066] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the methods according to the following embodiments of this application described in the "Exemplary Methods" section above.

[0067] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0068] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0069] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0070] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0071] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0072] The above description has been given for illustrative and descriptive purposes. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

[0073] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0074] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0075] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A visualization method for fundus abnormalities, characterized in that, include: Abnormal regions in the target fundus image are identified and processed to obtain abnormal classification results; Based on the anomaly classification results, all abnormal regions in the target fundus image are displayed differently according to the type of anomaly, generating a visualization result of fundus anomaly types.

2. The method according to claim 1, characterized in that, The step of identifying and processing abnormal regions in the target fundus image to obtain anomaly classification results includes: performing fundus anomaly detection on the target fundus image and generating fundus anomaly detection results; if the fundus anomaly detection results indicate that there is an abnormal region in the target fundus image and the abnormal region is a fundus anomaly, then performing fundus anomaly classification processing on the abnormal region in the target fundus image and generating fundus anomaly classification results; if the fundus anomaly detection results indicate that there is an abnormal region in the target fundus image and the abnormal region is not a fundus anomaly, then performing non-fundus anomaly classification processing on the abnormal region in the target fundus image and generating non-fundus anomaly classification results; and determining the fundus anomaly classification results and the non-fundus anomaly classification results as the anomaly classification results for the abnormal region.

3. The method according to claim 2, characterized in that, Also includes: If the fundus abnormality classification result indicates that the abnormal area is a lesion, then the abnormal area is subjected to lesion attribute classification processing to generate lesion attribute classification result; If the fundus abnormality classification result indicates that the abnormal area is a human intervention area, then the abnormal area is subjected to human interference factor classification processing to generate human intervention classification result; The lesion attribute classification result and the human intervention classification result are determined as the abnormal cause classification result in the fundus abnormality classification result.

4. The method according to claim 3, characterized in that, Also includes: If the lesion attribute classification result indicates that the lesion is a congenital lesion, then the abnormal area is subjected to congenital lesion classification processing to generate a congenital lesion classification result; If the lesion attribute classification result indicates that the lesion is a non-congenital lesion, then the abnormal area is subjected to non-congenital lesion classification processing to generate a non-congenital lesion classification result; The classification results of congenital lesions and the classification results of non-congenital lesions are determined as the lesion classification results in the lesion attribute classification results.

5. The method according to claim 4, characterized in that, Also includes: If the classification result of the non-congenital lesion indicates that the non-congenital lesion is a myopia disease, then the abnormal area is classified as a myopia disease by determining whether there are leopard spots and lesions outside the atrophic arc in the abnormal area, and a myopia disease classification result is generated. If the classification result of the non-congenital lesion indicates that the non-congenital lesion is a non-myopia disease, then the abnormal area is processed for non-myopia disease classification, and the non-myopia disease classification result is output. The myopia disease classification results and the non-myopia disease classification results are determined as the disease classification results in the non-congenital lesion classification results.

6. The method according to claim 1, characterized in that, The visualization results of fundus abnormalities include: a fundus abnormality display image and / or fundus abnormality description text; the fundus abnormality display image is used to show the distribution location, shape, type, size, and distribution of the fundus abnormality; the fundus abnormality description text includes at least one of the following: type, name, quantity information, area information, coordinate information, and orientation information of the fundus abnormality.

7. The method according to claim 1, characterized in that, Also includes: For each anomaly classification result, the anomaly region is classified according to its severity, and the anomaly classification result is output. Different abnormal regions corresponding to the same abnormality classification result are displayed differently according to the abnormality level classification result, generating a visualization result of fundus abnormality level.

8. The method according to claim 1, characterized in that, The abnormality classification results include fundus abnormality classification results and non-fundus abnormality classification results. The abnormal regions corresponding to the fundus abnormality classification results are classified according to severity, and fundus abnormality level classification results are output. The fundus abnormality level classification results include three scenarios: mild fundus abnormality classification results, moderate fundus abnormality classification results, and severe fundus abnormality classification results. Based on the non-fundus abnormality classification results, the mild fundus abnormality classification results, the moderate fundus abnormality classification results, and the severe fundus abnormality classification results, all abnormal regions in the target fundus image are displayed differentially to generate a fundus abnormality level visualization result.

9. The method according to claim 7, characterized in that, By assigning differentiated identifiers to the abnormal regions corresponding to each of the abnormal classification results, all abnormal regions in the target fundus image are visualized to generate a visualization result of fundus abnormality types; for any of the abnormal classification results: based on the differentiated identifiers corresponding to the abnormal classification results, the abnormality level classification results are differentiated by gradient changes in the identifier attributes to generate a visualization result of fundus abnormality levels.

10. A visualization device for fundus abnormalities, characterized in that, include: The identification module is used to identify abnormal regions in the target fundus image and obtain abnormal classification results; The classification differentiation display module is used to differentiate all abnormal regions in the target fundus image according to the abnormality type based on the abnormality classification result, and generate a visualization result of fundus abnormality types.

Images