通信英文文献.docVIP

Tongue diagnosis based on Bayesian network computer Peng Bo, Zhang David *, Senior Member, IEEE, Ming is Li, Wang Kuanquan, Member, IEEE Abstract:Tongue diagnosis is an important tradional Chinese medicine (TCM)diagnosticmethods.However,becauseofitsualitative,subjectiveandexperience-basednature,traditionalton guemedicine.Moreover has very limited clinical application of traditional tongue diagnosis has always been concerned with the connection identification card instead of the tongue look abnormal and disease. This is not a good understanding of Western medicine, which greatly hindered its widespread use in the world. In this paper, we propose a new method of computer tongue, to address these issues. First, second and quantitative features, color and texture types of measures extracted from the tongue using popular digital image processing technology. Then, the Bayesian network model employing the quantitative relationship between these functions and disease. Test the validity of the method 13 common diseases by 455 patients and 70 healthy volunteers and other diagnostic results from the previous forecast of Bayesian network classifier training reported. Index, Bayesian networks, computer tongue, Traditional Chinese Medicine. I. Introduction Tongue diagnosis [1], [2] is widely used in conjunction with the most valuable diagnostic methods of traditional Chinese medicine (TCM). On the tongue of the charm lies in its simple and direct: there is a complex disease as long as full of contradictions, tongue examination immediately to clarify the main pathological processes. Therefore, it is of great value in both clinical applications and self-diagnosis. In addition, tongue diagnosis is one of the few diagnostic techniques deal with the 21st century, the most promising direction: no pain and no injuries. As the tongue played such a prominent role in diagnosis and subsequent treatment of the disease, it has in terms of attracting more and more clinical medicine and bio