Computer-aided diagnosis (CAD) of the skin disease based on an intelligent classification of sonogram using neural network

Today skin diseases and lesions are the most common diseases that people suffer in different age groups, such as eczema, scalp ringworm, skin fungal, skin cancer of different intensity (basal cell carcinoma and squamous cell carcinoma and melanomaaEuro broken vertical bar), diabetic ulcers, and etc. There are different ways to evaluate and diagnose mentioned diseases. For example, most dermatologists prescribe the biopsy to diagnose them. This is a simple method to identify the type of skin disease, but that is an invasive method, and in prolonged time leads to pain and discomfort for patients. Another method that can be used to diagnose is based on non-ionizing radiation such as acoustic or ultrasound waves, which is being investigated in this study. It should be noted that ultrasound imaging is one of the best and useful medical diagnostic tool to scan soft tissue. Therefore, the aim of this study is to diagnose the diseases by studying and analyzing sonography images using intelligent artificial neural network, in order to eliminate any need for radiography and pathobiology process in dermatology. Our main diagnostic tool in this study is a sonography image acquisition system that uses non-ionizing ultrasound waves for skin imaging. Intelligent artificial neural network has been used to study and intelligently classify the skin sonograms. The results of this study show the high capability of this method in diagnosis and classification of the skin diseases.