Detection of skin rashes using image processing and deep learning architectures

Abstract

Skin diseases have become a growing issue all over the world, skin rashes is one of the most common among them. This paper focuses on establishing a viable deeplearning approach to distinguish between skin rashes. For this purpose, Deep Neural Network Models, such as CNN, MobileNetV2, ResNet50V2, and DenseNet201 algorithms were implemented and analyzed for suitability in this kind of image dataset. In this research, a synthetic dataset was prepared which consisted of 1381 images of three distinct Skin rash diseases, which were preprocessed using image resizing and dataset augmentation process. Among the implemented algorithms, multiple algorithms gave high accuracy scores, that suggested the high quality of the dataset.The models were also interpreted using an Explainable AI technique named LIME. ResNet50V2 provided the highest accuracy among the implemented algorithms followed by MobileNetV2