Skin Disease Classification Using Combined Machine Learning And Deep Learning Models

Abstract

Skin disease is a very common disease of living organisms. In the medical world, tracking and classifying skin diseases is a complex process. Due to the complexity of individual skin tone and the near visible effect of infections, recognizing the exact type can sometimes be challenging.  As a result, it is important to detect skin disease and identify it as soon as possible. Artificial intelligence (AI) is rapidly expanding into therapeutic areas in the modern environment For diagnostic purposes, more deep learning (DL) and machine learning (ML) methods are used. These techniques drastically improve the diagnostic process while also speeding it up. In this work, a combined deep learning (DL) and machine learning (ML) is developed to improve skin disease classification. The Convolution Neural Networks mode is used for feature extraction and classified using Machine Learning models include Decision Tree, Nearest Neighbor, Support Vector Machines and Light Gradient Boosting classifier.   To identify the best predictive model, a comparative study was carried and the hybrid method CNN with SVM gives the optimum results of 91.04% accuracy.