Classification of Skin Disease using Machine Learning Techniques

Abstract

Due to their complexity and time-consuming process, skin diseases can be difficult to diagnose in the global health community. Undiagnosed and uncontrolled skin diseases have adverse effects on human health and psychological well-being. With today's technology, it is possible to diagnose skin diseases quickly and easily using image processing and machine learning techniques. The article describes a method for diagnosing Actinic Keratosis , Atopic Dermatitis, Dermatofibroma, and Melanoma based on the image of the skin. This model involves five steps, the acquisition of images, the pre-processing, the segmentation, and the feature extraction. We also evaluated the model using machine learning algorithms, such as Support Vector Machines (SVMs), Random Forests (RFs), and K-Nearest Neighbors (K-NNs), and achieved 86.4%, 81.48%, and 59.25% of accuracy, respectively. Comparisons were made between the SVM classification results of the proposed model and those of other papers, and the proposed model generally performed better.