Design And Implementation Of A Mobile Medical Diagnostic Platform Based On Tongue

The rapid development of information science has brought great changes to all aspects of people’s lives. The rise of Telemedicine has a great impact on the traditional medical system. The traditional medical system can only serve a limited number of people for the limitation of public health resources. Thanks to the computer networks, Telemedicine is no longer limited by location and working time of doctors, which is about to benefit more people.Tongue diagnosis is one of four important methods of Traditional Chinese Medicine. Tongue diagnosis only calls for patients` tongue, and there is no want of complex diagnostic instruments which contributes to great effect with low cost. The traditional tongue diagnosis can’t guarantee to get an objective result due to various environmental factors. Therefore, it is of great significance to make the tongue diagnosis objective. Based on smart phones and tongue image, the thesis aims at creating mobile medical diagnosis platform, realizing its computerization, exploring the remote service.This thesis puts forward a set of procedures based on color correction, tongue image segmentation, extraction of tongue attribute and dialectical inference. The main research content is as follows:1. Study the color correction of tongue, and summarizes the causes of the tongue color cast. Considering the tongue color cast is mainly affected by the color temperature and color rendering index of the source light, this thesis uses e gray world algorithm to correct the tongue color cast and verify the algorithm.2. Compare various image segmentation algorithms, and summarize the usage scenario as well as the advantages and disadvantages of these algorithms. Considering the trait of tongue segmentation, this thesis comes up with an algorithm combined by Watershed Algorithm and Snake Algorithm. This algorithm can meet the requirements of tongue segmentation.3. After an in-depth study of tongue diagnosis of traditional Chinese medicine theory, this thesis summarizes the function of tongue quality and tongue fur in tongue diagnosis and a set of tongue attributes which have a great significance on tongue diagnosis. According to the features of different attributes, this thesis put forward their extraction algorithms.4. After a study of tongue diagnosis, this thesis put forward a dialectical algorithm based on weighted sum. The diagnosis result is obtained with the help of TCM Knowledge Base.5. This thesis has designed and implemented the server and client for the platform. The server is implemented by the Node.js, and has a cache based on Redis. The Client has implemented the customized camera, and accomplished the tongue diagnosis, medicine query and step counter.