Skin Diseases Classification Using Machine Learning and Deep Learning

Abstract

According to the Global Burden of Disease project, skin diseases are the fourth leading cause of benign sickness. Diagnosis of dermatological diseases presents a challenge alongside the absence of trained dermatologists and access to formal medical care. This presents a critical challenge, especially in countries with a large rural population and minimal development. This report aims to study machine learning-based classifiers for predicting skin infections for three classes from a clinical dataset. Convolutional neural network (CNN) has been proved to perform well in image classification. The performance of the neural network is compared with a benchmark multiclass SVM classifier. Additionally, an android based mobile application was implemented, which integrates the classification architecture into the application. The results analysis and possible future works are also discussed in this paper.