Automated Skin Disease Diagnosis Using CNN

Abstract

Skin disorders are among the most prevalent medical conditions in the world, and accurate diagnosis frequently requires skilled dermatologists. Due to the lack of access to dermatologists and the similarity of various skin conditions, early diagnosis is challenging. Recent developments in deep learning, specifically in Convolutional Neural Networks (CNNs), have enabled autonomous image identification. In this paper, a CNN-based method for dermoscopic image-based automated skin problem diagnosis is proposed. By learning from images that contain features like color, textures, and skin lesions, the suggested method distinguishes between various skin conditions. According to the results, the method has been shown to support telemedicine, extensive dermatological screening, and even early-stage diagnoses.