Unlock instant, AI-driven research and patent intelligence for your innovation.

Traditional Chinese medicine dialectical treatment recommendation system and method based on graph neural network, and electronic equipment

A neural network and recommendation system technology, applied in the field of Chinese medicine diagnosis and treatment recommendation system, can solve the problems of imprecise Chinese medicine diagnosis, affecting the accuracy of prediction results, and unable to effectively learn key features, so as to achieve good evaluation and prediction, and ensure reliability. sexual effect

Pending Publication Date: 2021-11-09
海南榕树家信息科技有限公司
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] How to use information technology to carry out diagnosis and treatment is a problem that has been explored in various medical and health fields. There are two types of methods in the existing technology: 1. Use algebra or clustering methods to solve the matching problem of symptoms and medicines, and some even skip dialectics directly The process of matching symptoms and medications is not rigorous in TCM diagnostics
2. It is more effective to predict the result by training the CNN convolutional neural network, but in the process of feature and weight training and learning, the key features cannot be effectively learned, which affects the accuracy of the prediction result

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Traditional Chinese medicine dialectical treatment recommendation system and method based on graph neural network, and electronic equipment
  • Traditional Chinese medicine dialectical treatment recommendation system and method based on graph neural network, and electronic equipment
  • Traditional Chinese medicine dialectical treatment recommendation system and method based on graph neural network, and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] On the one hand, if figure 1 As shown, the present invention provides an embodiment of a Chinese medicine diagnosis and treatment recommendation system based on a graph neural network, including:

[0067] The knowledge map acquisition module 301 is used to obtain the knowledge map of TCM diagnosis and treatment based on structured TCM diagnosis and treatment data.

[0068] Specifically, a large amount of structured information obtained by performing NLP (Natural Language Processing) on ​​the collected medical record data. After manual labeling and data processing, large-scale graph information is obtained. The graph information includes related attributes such as diseases, syndrome types, and traditional Chinese medicines, as well as related relationships.

[0069] Exemplary, collation of graph data: using manual and systematic processing of a large amount of medical record data. By sorting out raw materials such as medical records, the sorted data and the correspondi...

Embodiment 2

[0079] Based on the above-mentioned embodiment, the parts that are the same as those in the above-mentioned embodiment in this embodiment will not be repeated one by one. This embodiment provides a TCM diagnosis and treatment recommendation system based on a graph neural network, including:

[0080] The knowledge map acquisition module is also used for:

[0081] Structural processing of diseases, syndrome types, and characteristic attributes and correlations of traditional Chinese medicine through NLP to obtain structured data of traditional Chinese medicine diagnosis and treatment.

[0082] The building blocks include:

[0083] An extraction module, configured to sample the knowledge map through a graph neural network to obtain at least one subgraph, and extract the feature vector of the subgraph;

[0084] The training module is used to use the eigenvectors of the sub-graphs to convolve based on the average value aggregator, and use the negative sampling algorithm to perform...

Embodiment 3

[0213] Based on the above-mentioned embodiments, the parts that are the same as those in the above-mentioned embodiments in this embodiment will not be described one by one. This embodiment provides a TCM diagnosis and treatment recommendation method based on a graph neural network. Debate recommended methods, such as figure 2 shown, including:

[0214] S100 is based on the structured data of TCM debate and treatment, and obtains the knowledge map of TCM debate and treatment.

[0215] Specifically, a large amount of structured information obtained by performing NLP (Natural Language Processing) on ​​the collected medical record data. After manual labeling and data processing, a large-scale graph information is obtained. The graph information includes related attributes such as diseases, syndrome types, and traditional Chinese medicines, as well as related relationships.

[0216] Exemplary, collation of graph data: using manual and systematic processing of a large amount of ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the technical field of intelligent medical treatment, and provides a traditional Chinese medicine dialectical treatment recommendation system and method based on a graph neural network and electronic equipment, and the system comprises: a knowledge graph obtaining module which is used for obtaining a knowledge graph of traditional Chinese medicine dialectical treatment based on structured traditional Chinese medicinedialectical treatment data; a construction module which is used for extracting feature vectors from the knowledge graph through a graph neural network so as to construct low-latitude characterization information of the traditional Chinese medicine dialectical treatment; and a result acquisition module which is used for obtaining a prediction result of the traditional Chinese medicine dialectical treatment according to the low-latitude representation information of the traditional Chinese medicine dialectical treatment by utilizing a similarity evaluation algorithm, wherein the historical traditional Chinese medicine dialectical treatment data comprises symptoms, syndrome types, and characteristic attributes and association relationships of traditional Chinese medicines. The invention discloses a system for performing dialectical treatment by using a graph neural network based on a spatial domain. According to the system, a deep learning algorithm of a graph is utilized to extract traditional Chinese medicine disease vectors, traditional Chinese medicine syndrome type vectors and characterization information of traditional Chinese medicine vectors of large-scale graph expression relation data. By calculating a representation vector relationship, recommendation recall of traditional Chinese medicine dialectical treatment is achieved among diseases, syndromes and medicines.

Description

technical field [0001] The invention relates to the technical field of intelligent medical treatment, in particular to a graph neural network-based TCM diagnosis and treatment recommendation system, method and electronic equipment. Background technique [0002] In TCM diagnostics, it is pointed out that the basic principles of TCM diagnosis are: overall observation, combination of four diagnostic methods, combination of symptoms, and unity of movement and static. To comprehensively determine the medication. The discipline of traditional Chinese medicine is relatively large, and dialectics is divided into eight key principles, six evils, yin and yang, qi and blood, body fluid, viscera, six meridians, Wei qi and blood, triple burner, meridian dialectics, etc., so how to accurately and effectively dialectize diseases , classification and treatment are the core problems of traditional Chinese medicine. [0003] Due to the strong correlation between the analysis and input refer...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G16H20/90G16H50/70G06F16/36G06N3/04G06N3/08
CPCG16H20/90G16H50/70G06F16/367G06N3/08G06N3/045
Inventor 郝文晖刘毅冯军张少良
Owner 海南榕树家信息科技有限公司