Diseased Skin Dermis Approximation and Validation in Remote Space Using a Cognitive Network with Multiple Hidden Layers
摘要
The human body’s largest organ is its skin, thus, skin diseases are also one of the most common health problems around the world. In developing countries like India, there is 1 doctor for every 834 people and an alarming 1 dermatologist for every 1.40 lakh people. This is way below the current requirement. Most of the dermatologists are concentrated in the urban areas of the country. People in remote areas generally remain unaware of the problems or choose to ignore small symptoms like lesions on their skin until it is too late, which often leads to life-endangering situations. To help solve the problem of skin diseases, we have developed a deep learning model using CNN for differentiating between normal and diseased skin and further classifying the skin disease into 8 classes with an accuracy of 93.1%.