Diabetes Diagnostic Method based on Tongue Image Classification Using Machine Learning Algorithms

Abstract

Machine learning algorithms are implemented in variety of medicine applications in real time systems. In this sector, image classification plays a main role. Present days many people are suffering from Diabetes Mellitus, which is a disease caused by high blood glucose levels in the blood. Having too much glucose in your blood can leads to diabetes over time. Diabetes Mellitus also takes a toll on the entire human body, which effects on oral health. In real time, there are several algorithms are implemented for classification of various types of diseases for obtain good accuracy levels in classification levels. Artificial Intelligence shows major role in medicine applications in image classification. As a result, many techniques based on Machine Learning and Deep Learning have been developed. In this process, the classification can be performed with diabetes tongue, which preprocess for, extract features by various standard classification algorithms such as K-Nearest Neighbors, Support Vector Machine, Decision Tree, Logistic Regression, Convolution Neural Networks. The proposed algorithm developed with dimensional reduction-based CNN process, which is more accurate compared with other classification algorithms.