Intelligent processing method for tongue image information of stomach disease in traditional Chinese medicine based on deep learning

Abstract

The invention provides an intelligent processing method for tongue image information of a stomach disease in the traditional Chinese medicine based on deep learning, which comprises the steps of training data processing, namely, performing hierarchical quantification on collected patient symptom data to obtain quantified information of symptoms; tongue image feature extraction, namely, designing atongue image based convolution neural network according to the collected tongue image symptom data, and extracting tongue image features; multi-modal data fusion, namely, combining the tongue image features and the quantified information of the symptoms in a full connection layer, and carrying out training through a multilayer neural network to obtain a syndrome judgment result of the patient; model training, namely, performing model training according to the syndrome judgment result of the patient and a collected syndrome judgment result of an expert; and symptom and severity judgment. According to the technical scheme of the invention, the intelligent processing method can judge the symptoms and the severity of the patient after obtaining the tongue image of the patient and question answering results of the symptoms of the patient, the diagnosis efficiency is high, and the accuracy of the diagnosis result is high.

Description

Technical field [0001] The invention relates to the technical field of Chinese medicine automation, and in particular to an intelligent processing method for the information of the tongue image of Chinese medicine gastric disease based on deep learning. Background technique [0002] Traditional Chinese medicine conducts dialectics based on the four diagnosis information of looking, smelling, asking, and cutting. Among them, looking means looking at the complexion; smelling means listening to the sound, smelling the smell; asking; referring to the symptoms; cutting; referring to the pulse condition. The clinical dialectics of traditional Chinese medicine gastric disease mainly adopts three judgment methods: looking, asking, and cutting. However, because the collection of pulse information requires higher individualization of the instrument, the existing pulse imager cannot fully and objectively reflect the various pulse information, so the pulse imager has not yet The clinical ap...

AI Technical Summary

Problems solved by technology

In inspection, the information of the tongue image is a key information, and the features used for dialectics involve tongue shape, tongue quality, tongue coating, etc. The method based on simple features such as color cannot achieve intelligent dialectical effect

[0004] At present, the traditional Chinese medicine diagnosis method is inefficient on the one hand, and on the other hand, Chinese medicine requires long-term experience accumulation to obtain high accuracy results, so the diagnostic accuracy of Chinese medicine is low

Images