A Deep Learning Approach for Automated Skin Disease Classification and Diagnosis

Everyone is ware that a wide range of problems affect skin structure, function, and appearance. Dermatological issues or dermatoses are common names used to represent the skin related illness. Early detection of skill llness is possible by employing Artificial Intellengence. A wide range of artificial intelligence techniques have been devised to promptly detect and diagnose skin problems. In this article we are proposing Mobilenetv2 and LSTM algorithms to identify skin disorders at an early stage. The combination of MobileNetV2 and LSTM (Long Short-Term Memory) for skin disease detection involves leveraging a convolutional neural network (CNN) for image feature extraction and a recurrent neural network (RNN) with LSTM units for sequence modeling, this architecture aims to capture both spatial and temporal dependencies in skin image data, making it well-suited for skin disease detection tasks. The method shown in this study can identify five distinct diseases with ease, and with accuracy rates of 99.2%.