AN EFFICIENT MACHINE LEARNING BASED TONGUE COLOR ANALYSIS FOR AUTOMATED DISEASE DIAGNOSIS MODEL

The human tongue comprises several characteristics which can be utilized for disease diagnosis, especially the color features. Conventionally, the physicians investigate the color features depending upon the expert's knowledge and experience. In order to eradicate the qualitative aspects of the tongue analysis, the tongue color analysis can be objectively examined by the use of color features, which paves a way for effective disease diagnosis models. This paper introduces a new machine learning (ML) based tongue color analysis model for automated disease diagnosis. The presented model contains three distinct sub processes namely preprocessing, feature extraction, and classification. For feature extraction purposes, histogram of gradient (HOG) and Scale Invariant Feature Transform (SIFT) models are used. In addition, support vector machine (SVM) and decision tree (DT) classifiers are employed to assign the class label of the respective input images. For examining the effectual results of the presented model, a set of simulations were carried out and the results are investigated with respect to distinct performance measures. The resultant experimental values denoted the superior results of the presented model with the maximum precision of 91.65%, recall of 91.13%, accuracy of 91.13%, and F1-score of 91.09%.