Predictive Analytics for Tongue Diseases: A Comparative Study of Deep Learning Models

Use of AI in medical care has made remarkable strides by providing novel solutions for the quick identification and treatment of illness. The goal of this research is to use a variety of Deep Learning models to predict five varieties of tongue diseases by analyzing their images. The primary objective is to create a worthy model that is able to diagnose healthy tongue and tongue-related diseases such as Leukoplakia, Oral thrush, Oral lichen planus, Oral cancer and Hairy tongue. This research aims to reduce the cost of medical services by early recognition and therapy of oral infections. Advances in Deep Learning have revolutionized the conventional clinical diagnostics by providing devices that are capable of learning from dataset and distinguish the patterns which are unlikely noticeable to human. Here, Deep Learning models like Densenet121, EfficientnetB0, Resnet50V2 and Nasnetmobile are used to classify and predict the tongue disease. This research is conducted using publically available dataset which contains 1649 images placed in 6 classes. A tongue, which is frequently considered as a sign of health, can reveal a lot about a person's overall well-being and intended for early detection of infection, when treatment is most effective, by dissecting tongue images.