ABSTRACT
OBJECTIVE: Using the deep learning model to realize tongue image-based disease location recognition and focus on solving two problems: 1. The ability of the general convolution network to model detailed regional tongue features is weak; 2. Ignoring the group relationship between convolution channels, which caused the high redundancy of the model. METHODS: To enhance the convolutional neural networks. In this paper, a stochastic region pooling method is proposed to gain detailed regional features. Also, an inner-imaging channel relationship modeling method is proposed to model multi-region relations on all channels. Moreover, we combine it with the spatial attention mechanism. RESULTS: The tongue image dataset with the clinical disease-location label is established. Abundant experiments are carried out on it. The experimental results show that the proposed method can effectively model the regional details of tongue image and improve the performance of disease location recognition. CONCLUSION: In this paper, we construct the tongue image dataset with disease-location labels to mine the relationship between tongue images and disease locations. A novel fully-channel regional attention network is proposed to model the local detail tongue features and improve the modeling efficiency. SIGNIFICANCE: The applications of deep learning in tongue image disease-location recognition and the proposed innovative models have guiding significance for other assistant diagnostic tasks. The proposed model provides an example of efficient modeling of detailed tongue features, which is of great guiding significance for other auxiliary diagnosis applications.
Subject(s)
Neural Networks, Computer , Tongue , Image Processing, Computer-Assisted , Tongue/diagnostic imagingABSTRACT
Objective To explore characteristics of the elements of syndrome, the disease position and the relationship between chemical indicators and TCM syndromes of type 2 diabetic encephalopathy. Methods 2 501 cases of type 2 diabetes clinical data were collected from Guang'anmen Hospital of China Academy of Chinese Medical Sciences, Beijing University of Chinese Medicine, Dongzhimen Hospital, Dongfang Hospital, etc. in nearly 3 years, among which, 342 cases were type 2 diabetic encephalopathy. The original clinical data were double entried in epidata by two people, establishment forms in excel, factor analysis and Bayesian networks were used as data mining research methods. Results 20 elements which characteristic root more than 1 were derived by factor analysis, 68.4% were covered. Of all 20 elements, five factors belong to Yin, five factors belong to blood stasis; lassitude, shortness of breath, stool frequency were appeared when fasting glucose abnormalities; lassitude, hemiplegia were appeared when 2-hour postprandial blood glucose abnormalities;lassitude, feverish palms and soles, stool frequency, more nocturnal enuresis when glycated hemoglobin abnormalities by Bayesian networks. Conclusion The Elements of the syndrome of type 2 diabetes encephalopathy were deficiency of Yin and blood stasis; and the main positions for diabetic patients were liver, spleen and kidney. Patients with impaired fasting glucose were Qi deficiency; Patients with impaired 2-hour postprandial glucose were Qi deficiency or pathogenic wind attacking collaterals; Patients with abnormal hemoglobin were Qi deficiency and Yin deficiency.
ABSTRACT
Pathogenesis theory occupies an important position in TCM theoretical system. Suwen is the pioneer. Enriched by later generations,the traditional pathogenesis has been a significant change. It causes understandable bias to their initial implications and affects modern research. In this paper,based on Suwen's academic background,it discusses the composition,completeness and the basis to confirm Pathogenesis 19,offering some new ideas.
