Health analysis report generation method based on large model

Abstract

The invention discloses a health analysis report generation method based on a large model. The report generation method comprises the following steps: inputting a physical examination report; preprocessing the physical examination report to obtain a preprocessing result; performing visual language model analysis according to the preprocessing result; carrying out structured processing; compressing the cue word; generating an HTML (Hypertext Markup Language): generating webpage contents by utilizing And rendering and exporting. By introducing a large model with natural language understanding, data reasoning and code generation capabilities, an intelligent report generation mechanism is constructed, while the system development workload is reduced and the development period is shortened, specialized, precise and personalized output of health reports is realized, and the scientificity, readability and practical value of report contents are improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method for generating health analysis reports based on large models. Background Technology

[0002] With the rapid development of artificial intelligence technology, large-scale models have significantly improved their capabilities in data analysis and code generation. It has become a reality to use large-scale models to intelligently analyze data and automatically generate web page content including visual charts. In the healthcare field, especially in the interpretation and analysis of medical examination reports, large-scale model technology can automatically parse complex data and generate aesthetically pleasing, easy-to-read, and clearly structured professional health analysis reports, greatly improving the efficiency of information delivery and user experience.

[0003] Existing methods for generating health reports typically rely on manual development, extracting data indicators through rule-based approaches, and generating report content primarily by piecing together predefined templates.

[0004] Disadvantages of existing technology: High development and maintenance costs: It requires a lot of manual involvement in rule writing and template design, and the development cycle is long.

[0005] Low level of intelligence: Existing methods can usually only provide general health advice, lacking personalized and scientifically sound health guidance based on individual characteristics (such as age, gender, medical history, lifestyle habits, etc.), thus limiting their practicality and reference value. Summary of the Invention

[0006] In view of the above problems, the present invention is proposed to provide a method for generating health analysis reports based on a large model that overcomes or at least partially solves the above problems.

[0007] According to one aspect of the present invention, a method for generating health analysis reports based on a large model is provided, the report generation method comprising: Enter your medical examination report; The medical examination report is preprocessed to obtain the preprocessing result; Visual language model analysis is performed based on the preprocessing results; Perform structuring; Compress prompts; Generate HTML. Use a large language model to generate web page content; Rendering and exporting.

[0008] Optionally, the input medical examination report specifically includes: the system supports receiving medical examination report files in various formats as input, including text or scanned PDF and image files.

[0009] Optionally, the preprocessing of the medical examination report to obtain the preprocessing result specifically includes: After receiving a file uploaded by a user, the system first performs a validity check to ensure that the file format is correct and the content is complete. Files that pass the verification will enter the preprocessing process, which involves removing redundant information, invalid data, and invisible characters.

[0010] Optionally, the visual language model analysis based on the preprocessing results specifically includes: For medical examination reports containing images and complex charts, advanced visual language models are used for in-depth analysis. Extract key data indicators based on pre-set prompts; Output structured text data.

[0011] Optionally, the compressed prompt words specifically include: optimizing the prompt words using a small language model, removing redundant tokens, and shortening the length of the prompt words.

[0012] Optionally, the generation of HTML using a large language model to generate webpage content specifically includes: Based on the specified webpage design style, technical specifications, and output requirements, and combined with optimized prompts, the HTML code is automatically generated.

[0013] Optionally, the rendering and exporting specifically include: Render the generated HTML code into a visual webpage; It can selectively export as HTML files, PDFs, or images to meet the needs of different application scenarios.

[0014] This invention provides a method for generating health analysis reports based on a large-scale model. The method includes: inputting a physical examination report; preprocessing the report to obtain preprocessing results; performing visual language model analysis based on the preprocessing results; performing structured processing; compressing prompts; generating HTML; generating webpage content using the large-scale language model; rendering and exporting. By introducing a large-scale model with natural language understanding, data reasoning, and code generation capabilities, an intelligent report generation mechanism is constructed. This reduces system development workload and shortens the development cycle while achieving professional, accurate, and personalized output of health reports, enhancing the scientific rigor, readability, and practical value of the report content.

[0015] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of a method for generating health analysis reports based on a large model, provided as an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0019] The terms "comprising" and "having," and any variations thereof, in the specification, embodiments, claims, and drawings of this invention are intended to cover non-exclusive inclusion, such as including a series of steps or units.

[0020] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0021] This invention comprehensively utilizes prompting engineering, visual language models, and large language models, combined with other related technologies, to achieve a balance between cost, efficiency, and stability while ensuring system performance and output quality.

[0022] like Figure 1 As shown, the specific process is as follows: Medical examination report input. The system supports receiving medical examination report files in various formats as input, including but not limited to PDF (text version or scanned version), image files, etc. Iterative processing of user input files is performed from (2) to (3).

[0023] Preprocessing. Upon receiving a file uploaded by a user, the system first performs a validity check to ensure the file format is correct and the content is complete. Files that pass the check will then enter the preprocessing stage, which involves removing redundant information, invalid data, and invisible characters to optimize file size and improve subsequent processing efficiency.

[0024] Visual language model analysis. For medical examination reports containing images and complex charts, the system employs an advanced visual language model for deep analysis. During this process, key data indicators are extracted based on pre-defined prompts. After this step, the system can output structured text data for further processing.

[0025] Structured processing. Based on the results of steps 2 and 3, all extracted information is transformed into a specific data structure.

[0026] For example: [file name]: {file1_name}[file content begin]{file1_content}[filecontent end][file name]: {file2_name}[file content begin]{file2_content}[filecontent end] Prompt compression. Considering the limitations of large language model context windows, the system uses a small language model to optimize prompts, removing redundant tokens, thereby shortening prompt length and improving processing efficiency.

[0027] Generate HTML. This step utilizes a large language model to generate webpage content. Based on the specified webpage design style, technical specifications, and other output requirements, and combined with the optimized prompts from the previous step, the system automatically generates high-quality HTML code.

[0028] Rendering and Exporting. Finally, the system renders the generated HTML code into a visual webpage, or selectively exports it as an HTML file, PDF, image, or other formats to meet the needs of different application scenarios.

[0029] Beneficial effects: By introducing a large model with natural language understanding, data reasoning, and code generation capabilities, an intelligent report generation mechanism is constructed. This reduces the workload of system development, shortens the development cycle, and enables professional, accurate, and personalized output of health reports, thereby improving the scientific rigor, readability, and practical value of the report content.

[0030] The above specific embodiments further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for generating health analysis reports based on a large model, characterized in that, The report generation method includes: Enter your medical examination report; The medical examination report is preprocessed to obtain the preprocessing result; Visual language model analysis is performed based on the preprocessing results; Perform structuring; Compress prompts; Generate HTML. Use a large language model to generate web page content; Rendering and exporting.

2. The method for generating a health analysis report based on a large model according to claim 1, characterized in that, The input medical examination report specifically includes: The system supports receiving various formats of medical examination report files as input, including text or scanned PDFs and image files.

3. The method for generating a health analysis report based on a large model according to claim 1, characterized in that, The preprocessing of the medical examination report to obtain the preprocessing result specifically includes: After receiving a file uploaded by a user, the system first performs a validity check to ensure that the file format is correct and the content is complete. Files that pass the verification will enter the preprocessing process, which involves removing redundant information, invalid data, and invisible characters.

4. The method for generating a health analysis report based on a large model according to claim 1, characterized in that, The visual language model analysis based on the preprocessing results specifically includes: For medical examination reports containing images and complex charts, advanced visual language models are used for in-depth analysis. Extract key data indicators based on pre-set prompts; Output structured text data.

5. The method for generating a health analysis report based on a large model according to claim 1, characterized in that, The compressed prompt words specifically include: optimizing the prompt words using a small language model, removing redundant tokens, and shortening the length of the prompt words.

6. The method for generating a health analysis report based on a large model according to claim 1, characterized in that, The generation of HTML, specifically using a large language model to generate webpage content, includes: Based on the specified webpage design style, technical specifications, and output requirements, and combined with optimized prompts, the HTML code is automatically generated.

7. The method for generating a health analysis report based on a large model according to claim 1, characterized in that, The rendering and exporting specifically include: Render the generated HTML code into a visual webpage; It can selectively export as HTML files, PDFs, or images to meet the needs of different application scenarios.

Images