An analysis of tongue shape to identify diseases by using supervised learning techniques

Abstract

The tongue is a basic organ in Traditional Chinese Medicine (TCM). Tongue image diagnoses the health status of a human based on shape characteristics of the tongue. Tongue shape is a very important feature in tongue diagnosis. Different tongue shapes like circle, rectangle and triangle indicate presence of certain pathologies. In this paper, we propose a method for automated recognition of diseases based on the shape of a tongue image using supervised machine learning techniques. Experimental results obtained with a set of tongue images demonstrate that the SVM classifier improves the current state of the art method.