Effective machine learning-based skin disease diagnosis using PyTorch

Abstract The application of machine learning in medical diagnosis has become a trend research. Skin infection is one the most seen diseases and world’s infectious diseases, influencing people all ages. reason for explicit attention researchers skin detection due to that disease more visible compared any other disease. In past, varied methods have been proposed, which rendered remarkable results. However, presently functional models are trained on specific kinds limited 4 5 classes, inefficient detecting large set diseases. paper offers weightless model 23 different PyTorch backend, gives flexibility developing an algorithm. attained 96.37% accuracy training data, 87.75% test expected improve as size dataset increased.