Vitiligo Image Categorization Using Convolution Neural Network

Abstract

Vitiligo is a widespread skin condition characterized by the loss of melanocytes, resulting in chalky-white patches on the skin. The exact cause of vitiligo is not fully understood, but it is generally considered to be an autoimmune condition in which the body's immune system mistakenly attacks and destroys melanocytes. It affects a significant portion of the global population, with an estimated prevalence of 0.5-2%. This common pigment-related disorder is categorized into two main types: Segmental and Nonsegmental vitiligo. Categorizing vitiligo images using Convolutional Neural Networks (CNNs) involves developing a deep learning model that can automatically identify and classify images based on their features. Below is a high-level overview of the steps you can follow to implement vitiligo image categorization using CNN. The challenges posed by the subjectivity of dermatologist evaluations and the need for accurate diagnosis drive the demand for machine learning solutions. In response to this need, the research introduces an intelligent approach using Convolutional Neural Networks (CNNs) for Vitiligo classification. The proposed model, known as IVC, concentrates on classifying Nonsegmental vitiligo into its subtypes, such as Acrofacial, Focal, Generalized, and Mucosal vitiligo. To support this endeavour, a dataset of 368 vitiligo-infected photos has been thoroughly collected and categorised. The research aims to design a sophisticated framework for the precise detection and classification of Vitiligo. This innovative approach harnesses the power of AI and CNNs to enhance the accuracy of diagnosis, thereby addressing a critical need in the field.