AN ERYTHEMATO SQUAMOUS DISEASE (ESD) DETECTION USING DBN TECHNIQUE

ESD is a serious type of skin disease that increases over the past decades in the world and as a sequel to curing strategy in the medical field, automatic detection of ESDs using dermoscopic images has been still challenging and complicated task. This kind of difficulty occurs in the diagnosis of ESD owing to the following factors such as indistinct ESD borders, poor color contrast, location-dependent, shape variations, and complex structures of the ESDs. The progressing public health burden issues have to be detected early and treated in proper ways to prevent further spreading to other organs of the body through which medical professionals and researchers can save several lives. When there is an abnormal change in the appearance of the skin, then there is a chance for the subject that may be affected by ESD. To obtain better solutions, the computer vision methods must be paired with dermatology knowledge for efficient ESD detection. Hence, it is important to develop Deep Belief Network (DBN) based detection techniques to assist clinicians to diagnose ESD at early stages.