Optimal Edge Detection Technique for Diagnosis & Treatment of Leukoplakia in Oral Mucosa

Tongue inspection plays a very important role to monitor the health of patient and it is commonly used in Traditional Chinese Medicine (TCM). Texture analysis of Tongue Contours is an important issue in development of disease diagnostic expert system using analysis of tongue images. In this paper we present an approach of medical biometrics to detect and diagnose Leukoplakia. The aim of this paper will be to enhance the extracted part of the tongue, finding the severity of affected part and decide the proper treatment of the same. In this proposed system we not only extract the affected area but also calculate the same using geometrical and textural features of image. physiological and pathological changes and get a patient physical condition. Traditional tongue diagnosis mainly depends on the doctor's experience and knowledge, this can't cause the diagnosis of subjectivity and difficult to repeatability. It is required to combine TCM expert’s clinical experience with modern information technology. In order to analyze tongue image, the important premise is to accurately segment tongue from original tongue image. Because of the various shapes and colors, and lots of noise information from the mouth, nose and face, segmenting the tongue effectively becomes a difficult problem. And the segmentation result will directly affect the accuracy of image analysis. So we need to improve the efficiency of the technique used to segment the tongue image. M. Dhanalakshmi et al. [1] had introduced a sequential image processing technique for automated tongue segmentation in which a gradual, step by step sequential process for extraction of the shape feature, color feature and so on for the tongue analysis. The aim of this method was to reduce the complexity in tongue segmentation. Lam Keywords-Tongue Image, Leukoplakia, ISEF, TCM, Tongue Extraction.