SDD-ConvNet: A Novel Convolutional Neural Network Architecture for Multi-class Skin Disease Detection

The skin is the body’s most effective barrier against external threats. It serves as a barrier to prevent harm to our interior organs. However, this vital component of the body can experience such severe diseases brought on by fungi, viruses, or even dust. Numerous skin conditions affect thousands of people worldwide. People endure a lot of suffering, from eczema to acne issues. A thorough identification can lead to a correct course of treatment, which can lessen the suffering of those who are ill and make people aware. In this study, we aimed to create a prototype for a neural network-based skin disease detection system. We have selected CNN, or convolutional neural network, as our neural network of selection with Mish Activation function and proposed a model for Skin Diseases Detection using Convolutional Neural Network namely “SDD-ConvNet”. This research can help people to recognize common skin conditions such as Fungal, eczema, etc. We operated our model on the Dermnet dataset of images of diverse skin diseases specifically Eczema, Psoriasis, Nail Fungus, Tinea, and Seborrheic Keratoses. Further, we prepare the data to scale down the computation complexity and trained our model using the training dataset. Finally we tested our model and compared our model performance with some other popular model in order to evaluate the effectiveness in terms of Precision, Recall, and F1-Score.