A Novel Approach for Analyzing Diabetes Mellitus and Non Proliferative Diabetic Retinopathy Using Tongue

Objectives: To find out the diabetes mellitus disease with better accuracy by comparing the tongue features of diseased persons with the tongue features of normal persons.

Methods: Median filter is used for remove the unwanted noise from tongue image. The proximal support vector machine is a classification technique which is used for separate the diabetes mellitus and non proliferative diabetic retinopathy samples from the healthy samples.

Findings: The proposed method achieves high performance in terms of precision, recall, accuracy and F-measure.

Application/Improvements: Median filter is used for preprocessing the tongue image and PSVM classifier used for classification which results in better accuracy than the SVM classifier.