An ophthalmic disease visual video generation method and device, a terminal device, and a storage medium

Abstract

The application discloses an ophthalmic disease visual video generation method and device, a terminal equipment and a storage medium, which comprises the following steps: inputting an initial visual description text describing the vision of a patient into a preset language large model, so that the language large model combines a preset eye disease visual symptom knowledge graph to perform semantic analysis on the initial visual description text, and according to the semantic analysis result, the initial visual description text is optimized to generate a visual description text recording a dynamic visual scene of the patient; and inputting the visual description text into a preset text-to-video model, so that the text-to-video model outputs a subjective visual video consistent with the visual scene of the patient according to the visual description text. Therefore, the application can output a subjective visual video consistent with the visual scene of the patient, thereby effectively helping doctors understand the visual symptoms of the patient and improving the diagnosis efficiency of the doctors.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method, apparatus, terminal device, and storage medium for generating visual videos of ophthalmic diseases. Background Technology

[0002] Ophthalmic diseases are characterized by "what you see is what you have," making diagnosis heavily reliant on symptoms, signs, and imaging. Visual symptoms are among the most fundamental manifestations of ophthalmic diseases, with most patients experiencing them, such as blurred vision and distorted vision. In the past, textbooks and medical records simply categorized various visual symptoms using terms like "distorted vision" and "double vision," but this is a simplification of information and cannot replace the actual visual experience. For example, some patients' descriptions of "seeing a ball of yarn in bright light but not in darkness" or "seeing the road stretching to the sky while driving at night" cannot be directly categorized into existing symptom classifications. Furthermore, due to differences in patients' education levels and language abilities, their verbal descriptions of symptoms vary greatly and are highly subjective, potentially leading to misunderstandings by doctors.

[0003] Therefore, for ophthalmology, a discipline that relies heavily on visual symptoms for diagnosis and treatment evaluation, using only text and language to describe medical records reduces the efficiency of information transmission and affects doctors' understanding and judgment of patients' visual symptoms. Summary of the Invention

[0004] This invention provides a method, apparatus, terminal device, and storage medium for generating visual videos of ophthalmic diseases, which can effectively help doctors understand patients' visual symptoms and improve doctors' diagnostic efficiency.

[0005] An embodiment of the present invention provides a method for generating visual videos of ophthalmic diseases, comprising:

[0006] Obtain the initial visual description text of the patient's own vision;

[0007] The initial visual description text is input into a preset language model, so that the language model combines a preset knowledge graph of visual symptoms of eye diseases to perform semantic analysis on the initial visual description text, and optimizes the initial visual description text based on the semantic analysis results to generate a visual description text that records the patient's dynamic visual scene.

[0008] The visual description text is input into a preset text-based video model so that the text-based video model outputs a subjective visual video that matches what the patient sees, based on the visual description text.

[0009] Furthermore, the language big data model, combined with a pre-defined knowledge graph of visual symptoms of eye diseases, performs semantic analysis on the initial visual description text, and optimizes the initial visual description text based on the semantic analysis results to generate a visual description text, including:

[0010] Semantic analysis and dependency parsing are performed on the initial visual description text to extract symptom descriptions and dynamic scene descriptions from the initial visual description text.

[0011] Based on the symptom description, several symptom medical terms are determined. For each symptom medical term, a target symptom node is matched from the visual symptom knowledge graph of eye diseases, and typical dynamic features and associated scene features associated with the target symptom node are determined. The visual symptom knowledge graph of eye diseases is a ternary relationship network of symptom nodes, associated scene features, and typical dynamic features.

[0012] Based on the symptom medical terminology, the dynamic scene description, the typical dynamic features, and the associated scene features, the text is reconstructed to generate a visual description text that records the patient's dynamic visual scene.

[0013] Furthermore, the step of reconstructing the text based on the symptom medical terminology, the dynamic scene description, the typical dynamic features, and the associated scene features to generate visual description text recording the patient's dynamic visual scene includes:

[0014] Perform dynamic scene analysis on the dynamic scene description and generate analysis results;

[0015] When the analysis results indicate that the dynamic scene description lacks specific scene elements, the associated scene features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0016] When the analysis results indicate that the dynamic scene description lacks dynamic temporal elements, the typical dynamic features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0017] When the analysis results indicate that the dynamic scene description lacks specific scene elements and dynamic temporal elements, the associated scene features and the typical dynamic features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0018] Furthermore, the text-based video model outputs a subjective visual video that matches the patient's perception, based on the visual description text, including:

[0019] Generate a dynamic scene video based on the dynamic scene description in the visual description text;

[0020] Based on the symptom medical terms in the visual description text, extract the corresponding symptom materials from a preset symptom video material library;

[0021] The symptom footage is integrated with the dynamic scene video to generate the subjective visual video.

[0022] Another embodiment of the present invention provides a visual video generation device for ophthalmic diseases, comprising:

[0023] The text acquisition module is used to acquire the initial visual description text of the patient describing their own vision;

[0024] The text reconstruction module is used to input the initial visual description text into a preset language model, so that the language model combines with a preset knowledge graph of visual symptoms of eye diseases to perform semantic analysis on the initial visual description text, and optimize the initial visual description text based on the semantic analysis results to generate visual description text that records the patient's dynamic visual scene.

[0025] The video editing module is used to input the visual description text into a preset text-based video model, so that the text-based video model outputs a subjective visual video that matches the patient's view based on the visual description text.

[0026] Furthermore, the language big data model, combined with a pre-defined knowledge graph of visual symptoms of eye diseases, performs semantic analysis on the initial visual description text, and optimizes the initial visual description text based on the semantic analysis results to generate a visual description text, including:

[0027] Semantic analysis and dependency parsing are performed on the initial visual description text to extract symptom descriptions and dynamic scene descriptions from the initial visual description text.

[0028] Based on the symptom description, several symptom medical terms are determined. For each symptom medical term, a target symptom node is matched from the visual symptom knowledge graph of eye diseases, and typical dynamic features and associated scene features associated with the target symptom node are determined. The visual symptom knowledge graph of eye diseases is a ternary relationship network of symptom nodes, associated scene features, and typical dynamic features.

[0029] Based on the symptom medical terminology, the dynamic scene description, the typical dynamic features, and the associated scene features, the text is reconstructed to generate a visual description text that records the patient's dynamic visual scene.

[0030] Furthermore, the step of reconstructing the text based on the symptom medical terminology, the dynamic scene description, the typical dynamic features, and the associated scene features to generate visual description text recording the patient's dynamic visual scene includes:

[0031] Perform dynamic scene analysis on the dynamic scene description and generate analysis results;

[0032] When the analysis results indicate that the dynamic scene description lacks specific scene elements, the associated scene features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0033] When the analysis results indicate that the dynamic scene description lacks dynamic temporal elements, the typical dynamic features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0034] When the analysis results indicate that the dynamic scene description lacks specific scene elements and dynamic temporal elements, the associated scene features and the typical dynamic features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0035] Furthermore, the text-based video model outputs a subjective visual video that matches the patient's perception, based on the visual description text, including:

[0036] Generate a dynamic scene video based on the dynamic scene description in the visual description text;

[0037] Based on the symptom medical terms in the visual description text, extract the corresponding symptom materials from a preset symptom video material library;

[0038] The symptom footage is integrated with the dynamic scene video to generate the subjective visual video.

[0039] Another embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement a visual video generation method for an ophthalmic disease as described in any of the embodiments.

[0040] Another embodiment of the present invention provides a storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the storage medium is located to perform a visual video generation method for an ophthalmic disease as described in any of the above embodiments.

[0041] The following benefits can be obtained by implementing the present invention:

[0042] This invention discloses a method, apparatus, terminal device, and storage medium for generating visual videos of ophthalmic diseases. The method involves inputting an initial visual description text of the patient's vision into a preset language model. This language model, combined with a preset knowledge graph of visual symptoms of ophthalmic diseases, performs semantic analysis on the initial visual description text. Based on the semantic analysis results, the initial visual description text is optimized to generate visual description text recording the patient's dynamic visual scene. This visual description text is then input into a preset text-based video model, which outputs a subjective visual video that matches the patient's perception. Therefore, this invention can output a subjective visual video that matches the patient's perception, effectively helping doctors understand the patient's visual symptoms and improving diagnostic efficiency. Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating a method for generating visual videos of ophthalmic diseases according to an embodiment of the present invention.

[0044] Figure 2 This is a schematic diagram of the structure of a visual video generation device for ophthalmic diseases provided in an embodiment of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0047] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0048] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0049] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0050] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0051] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0052] See Figure 1 This is a flowchart illustrating a method for generating visual videos of ophthalmic diseases according to an embodiment of the present invention, comprising:

[0053] S1. Obtain the initial visual description text of the patient's own vision;

[0054] In a preferred embodiment of the present invention, the visual description text can be speech-to-text of a patient's dictation after watching a video, speech-to-text of a patient's dictation of their own vision, or a questionnaire filled out by the patient. It should be noted that, in order to avoid the patient's limited expressive ability, the visual description text can also be the text of a question-and-answer recording generated from a question-and-answer dialogue between a doctor and a patient.

[0055] S2. Input the initial visual description text into a preset language model so that the language model combines with a preset knowledge graph of visual symptoms of eye diseases to perform semantic analysis on the initial visual description text, and optimize the initial visual description text based on the semantic analysis results to generate a visual description text that records the patient's dynamic visual scene.

[0056] Preferably, the language big data model, combined with a pre-defined knowledge graph of visual symptoms of eye diseases, performs semantic analysis on the initial visual description text, and optimizes the initial visual description text based on the semantic analysis results to generate visual description text, including:

[0057] S21. Perform semantic analysis and dependency parsing on the initial visual description text to extract symptom descriptions and dynamic scene descriptions from the initial visual description text;

[0058] S22. Based on the symptom description, determine several symptom medical terms. Based on each symptom medical term, match the target symptom node from the eye disease visual symptom knowledge graph, and determine the typical dynamic features and associated scene features associated with the target symptom node; wherein, the eye disease visual symptom knowledge graph is a ternary relationship network of symptom nodes, associated scene features, and typical dynamic features.

[0059] S23. Based on the symptom medical terminology, the dynamic scene description, the typical dynamic features, and the associated scene features, perform text reconstruction to generate visual description text that records the patient's dynamic visual scene.

[0060] Preferably, the step of reconstructing the text based on the symptom medical terminology, the dynamic scene description, the typical dynamic features, and the associated scene features to generate visual description text recording the patient's dynamic visual scene includes:

[0061] S231. Perform dynamic scene analysis on the dynamic scene description and generate analysis results;

[0062] S232. When the analysis results indicate that the dynamic scene description lacks specific scene elements, the associated scene features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0063] S233. When the analysis results indicate that the dynamic scene description lacks dynamic temporal elements, the typical dynamic features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0064] S234. When the analysis results indicate that the dynamic scene description lacks specific scene elements and dynamic temporal elements, the associated scene features and the typical dynamic features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0065] In a preferred embodiment of the present invention, Large Language Model (LLM) is an advanced model in the field of artificial intelligence. Its core lies in mastering language rules and generating natural and fluent text through training on massive amounts of text data. The structure of the knowledge graph of visual symptoms of eye diseases is a symptom-scene-dynamic ternary relation network structure. For example, "floaters" is associated with the scene "more obvious on a white background", and the dynamic attribute "drifts with eye movement" is defined.

[0066] It should be further explained that NER technology is used to extract symptom keywords (such as "dark shadows" and "distorted vision") from patient texts and annotate them with standard medical terminology. The time sequence (such as "appears in the morning, worsens in the afternoon"), spatial relationships (such as "outer left eye"), and triggering conditions (such as "suddenly occurs when turning the head") in the description are analyzed as dynamic scene descriptions. Furthermore, the identified symptom entities are aligned with knowledge graph nodes to retrieve their associated scenes and typical dynamic features. In this embodiment, potential missing information is also discovered through graph traversal; for example, if a patient describes "ripples in front of their eyes" but does not mention the duration, the system automatically associates it with the possible progressive deterioration characteristic of "retinal detachment." Furthermore, when the analysis results indicate that the dynamic scene description lacks specific scene elements, scene prompts recommended by the knowledge graph are inserted into the patient's original description, such as optimizing "seeing things as distorted" to "noticing straight lines being distorted when reading a book." When the analysis results indicate that the dynamic scene description lacks dynamic temporal elements, the static description is transformed into a dynamic process based on the typical dynamic features of the symptoms, for example, adding "when looking to the left, vision gradually blurs," etc. Finally, this embodiment also eliminates contradictory descriptions (such as the conflict between "persistent flashes of light" and "transient migraine visual aura") by using taboo relationships in the atlas.

[0067] Furthermore, the construction process of the knowledge graph of visual symptoms of eye diseases is as follows:

[0068] First, detailed descriptions of various eye diseases and their symptoms from experts are obtained, standardized descriptions from authoritative ophthalmology guidelines are compiled, and subjective descriptions of symptoms from actual patients are collected. The collected text data is then cleaned and standardized to remove redundant information. The text data is then labeled, and the labeling and feature extraction process is detailed as follows: Based on the patients' final eye disease diagnoses in the existing data, their subjective visual effect descriptions are categorized and labeled, such as: visual effects of glaucoma, visual effects of fundus diseases, visual effects of strabismus, etc. Referring to guidelines and textbooks for the visual effects of each labeled disease, standard subjective descriptive features for each labeled disease are extracted, such as: visual effects of glaucoma—halos around lights, visual field defects; visual effects of fundus diseases—distorted vision, fixed black shadows in front of the eyes; visual effects of strabismus—double vision, etc. Finally, typical characteristics, associated scenarios (such as under strong light, during movement), and dynamic change patterns (such as over time, triggered by head movements) of visual symptoms (such as visual field defects, flashes of light) are compiled.

[0069] S3. Input the visual description text into a preset text-based video model so that the text-based video model outputs a subjective visual video that matches the patient's view based on the visual description text.

[0070] Preferably, the text-based video model outputs a subjective visual video that matches the patient's perception, based on the visual description text, including:

[0071] S31. Generate a dynamic scene video based on the dynamic scene description in the visual description text;

[0072] S32. Based on the symptom medical terms in the visual description text, extract the corresponding symptom materials from the preset symptom video material library;

[0073] S33. Integrate the symptom materials with the dynamic scene video to generate the subjective visual video.

[0074] In a preferred embodiment of the present invention, the text-based video model first parses the input dynamic scene description using natural language processing techniques (such as word semantic analysis and semantic understanding), extracts scene elements, and combines related datasets of text, images, and videos to map key elements into visual elements. Then, it extracts the dynamic description, constructs a video script, and builds a dynamic scene video based on the visual elements. Finally, based on the extracted symptom materials, the dynamic scene video is processed and edited.

[0075] Furthermore, the training steps for the video editing model are as follows:

[0076] The model is fine-tuned using a CogVi deo-based training framework. Standardized text descriptions and corresponding video inputs based on subjective vision are used to adjust the output video. This reduces the discrepancy between the generated video and the patient's selected video, making it easier for the model to achieve the desired effect and improving the efficiency of patient video selection. The generated video is compared with patient feedback to evaluate the model's generation effect and its matching degree with real patients. Based on patient feedback, the "text description-video" data pairs are optimized to construct a high-quality secondary training dataset. By introducing new optimized data pairs, the model is further fine-tuned to improve the matching accuracy between the generated video and the patient's description. Continuous evaluation continues until the model's results achieve high satisfaction in clinical applications. Furthermore, this invention can also be applied to the learning scenarios of ophthalmology medical students to help them understand ophthalmic diseases.

[0077] This embodiment provides a method for generating visual videos of ophthalmic diseases. The method involves acquiring visual description text after a patient watches an initial video; inputting this initial visual description text into a preset language model, which then extracts and outputs several visual keywords; and inputting these visual keywords along with the initial video into a preset video editing model, which edits the initial video based on the visual keywords to output a subjective visual video that matches the patient's perception. Therefore, this invention can effectively help doctors understand patients' visual symptoms and improve diagnostic efficiency by editing the initial video to output a subjective visual video that matches the patient's perception.

[0078] See Figure 2 This is a schematic diagram of the structure of a visual video generation device for ophthalmic diseases provided in an embodiment of the present invention, comprising:

[0079] The text acquisition module is used to acquire the initial visual description text of the patient describing their own vision;

[0080] The text reconstruction module is used to input the initial visual description text into a preset language model, so that the language model combines with a preset knowledge graph of visual symptoms of eye diseases to perform semantic analysis on the initial visual description text, and optimize the initial visual description text based on the semantic analysis results to generate visual description text that records the patient's dynamic visual scene.

[0081] The video editing module is used to input the visual description text into a preset text-based video model, so that the text-based video model outputs a subjective visual video that matches the patient's view based on the visual description text.

[0082] Preferably, the language big data model, combined with a pre-defined knowledge graph of visual symptoms of eye diseases, performs semantic analysis on the initial visual description text, and optimizes the initial visual description text based on the semantic analysis results to generate visual description text, including:

[0083] Semantic analysis and dependency parsing are performed on the initial visual description text to extract symptom descriptions and dynamic scene descriptions from the initial visual description text.

[0084] Based on the symptom description, several symptom medical terms are determined. For each symptom medical term, a target symptom node is matched from the visual symptom knowledge graph of eye diseases, and typical dynamic features and associated scene features associated with the target symptom node are determined. The visual symptom knowledge graph of eye diseases is a ternary relationship network of symptom nodes, associated scene features, and typical dynamic features.

[0085] Based on the symptom medical terminology, the dynamic scene description, the typical dynamic features, and the associated scene features, the text is reconstructed to generate a visual description text that records the patient's dynamic visual scene.

[0086] Preferably, the step of reconstructing the text based on the symptom medical terminology, the dynamic scene description, the typical dynamic features, and the associated scene features to generate visual description text recording the patient's dynamic visual scene includes:

[0087] Perform dynamic scene analysis on the dynamic scene description and generate analysis results;

[0088] When the analysis results indicate that the dynamic scene description lacks specific scene elements, the associated scene features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0089] When the analysis results indicate that the dynamic scene description lacks dynamic temporal elements, the typical dynamic features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0090] When the analysis results indicate that the dynamic scene description lacks specific scene elements and dynamic temporal elements, the associated scene features and the typical dynamic features are used to complete the dynamic scene description, and a visual description text is generated based on the completed dynamic scene description and the symptom medical terminology.

[0091] Preferably, the text-based video model outputs a subjective visual video that matches the patient's perception, based on the visual description text, including:

[0092] Generate a dynamic scene video based on the dynamic scene description in the visual description text;

[0093] Based on the symptom medical terms in the visual description text, extract the corresponding symptom materials from a preset symptom video material library;

[0094] The symptom footage is integrated with the dynamic scene video to generate the subjective visual video.

[0095] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0096] Those skilled in the art will clearly understand that, for convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0097] Another preferred embodiment of the present invention provides a terminal device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a visual video generation method for an ophthalmic disease as described in any of the above embodiments.

[0098] The terminal device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0099] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0100] The memory can be used to store the computer program. The processor implements various functions of the terminal device by running or executing the computer program stored in the memory and calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart memory card (SMC), secure digital card (SD) card, flash memory card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0101] Another preferred embodiment of the present invention provides a storage medium, which is a computer-readable storage medium, in which a computer program is stored. When executed by a processor, the computer program can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium can include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0102] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for generating visual videos of ophthalmic diseases, characterized in that, include: Obtain the initial visual description text of the patient's own vision; The initial visual description text is input into a preset language model, which performs semantic and dependency parsing analysis to extract symptom descriptions and dynamic scene descriptions. Based on the symptom descriptions, several symptom medical terms are determined. For each symptom medical term, a target symptom node is matched from an ophthalmic visual symptom knowledge graph, and typical dynamic features and associated scene features associated with the target symptom node are determined. The ophthalmic visual symptom knowledge graph is a ternary relation network of symptom nodes, associated scene features, and typical dynamic features. Dynamic scene analysis is performed on the dynamic scene description to generate analysis results. The analysis results characterize the deficiencies in the dynamic scene description. When specific scene elements are lacking, the associated scene features are used to complete the dynamic scene description, and visual description text is generated based on the completed dynamic scene description and the symptom medical terminology; when the analysis results indicate that the dynamic scene description lacks dynamic temporal elements, the typical dynamic features are used to complete the dynamic scene description, and visual description text is generated based on the completed dynamic scene description and the symptom medical terminology; when the analysis results indicate that the dynamic scene description lacks both specific scene elements and dynamic temporal elements, the associated scene features and the typical dynamic features are used to complete the dynamic scene description, and visual description text is generated based on the completed dynamic scene description and the symptom medical terminology. The visual description text is input into a preset text-based video model so that the text-based video model outputs a subjective visual video that matches what the patient sees, based on the visual description text.

2. The method for generating visual videos of ophthalmic diseases as described in claim 1, characterized in that, The text-based video model outputs a subjective visual video that matches the patient's perception, based on the visual description text, including: Generate a dynamic scene video based on the dynamic scene description in the visual description text; Based on the symptom medical terms in the visual description text, extract the corresponding symptom materials from a preset symptom video material library; The symptom footage is integrated with the dynamic scene video to generate the subjective visual video.

3. A visual video generation device for ophthalmic diseases, characterized in that, include: The text acquisition module is used to acquire the initial visual description text of the patient describing their own vision; The text reconstruction module is used to input the initial visual description text into a preset language model, so that the language model performs semantic analysis and dependency syntax analysis on the initial visual description text to extract symptom descriptions and dynamic scene descriptions from the initial visual description text; based on the symptom descriptions, several symptom medical terms are determined; based on each symptom medical term, a target symptom node is matched from the ophthalmic visual symptom knowledge graph, and typical dynamic features and associated scene features associated with the target symptom node are determined; wherein, the ophthalmic visual symptom knowledge graph is a ternary relation network of symptom nodes, associated scene features, and typical dynamic features; dynamic scene analysis is performed on the dynamic scene description to generate analysis results; the dynamic scene is characterized by the analysis results. When the scene description lacks specific scene elements, the associated scene features are used to complete the dynamic scene description, and visual description text is generated based on the completed dynamic scene description and the symptom medical terminology; when the analysis result indicates that the dynamic scene description lacks dynamic temporal elements, the typical dynamic features are used to complete the dynamic scene description, and visual description text is generated based on the completed dynamic scene description and the symptom medical terminology; when the analysis result indicates that the dynamic scene description lacks both specific scene elements and dynamic temporal elements, the associated scene features and the typical dynamic features are used to complete the dynamic scene description, and visual description text is generated based on the completed dynamic scene description and the symptom medical terminology. The video editing module is used to input the visual description text into a preset text-based video model, so that the text-based video model outputs a subjective visual video that matches the patient's view based on the visual description text.

4. The visual video generation device for ophthalmic diseases as described in claim 3, characterized in that, The text-based video model outputs a subjective visual video that matches the patient's perception, based on the visual description text, including: Generate a dynamic scene video based on the dynamic scene description in the visual description text; Based on the symptom medical terms in the visual description text, extract the corresponding symptom materials from a preset symptom video material library; The symptom footage is integrated with the dynamic scene video to generate the subjective visual video.

5. A terminal device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements a visual video generation method for an ophthalmic disease as described in any one of claims 1 to 2.

6. A storage medium, characterized in that, The storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the storage medium is located to perform a visual video generation method for an ophthalmic disease as described in any one of claims 1 to 2.

Images