AUTOMATED DIAGNOSIS OF RINGWORM INFECTION THROUGH A WEB APPLICATION

Abstract

Ringworm, a fungal infection caused by dermatophytes, is a common and contagious skin disorder that can cause significant discomfort and embarrassment. Traditional diagnosis of ringworm often involves visual inspection, microscopy and laboratory cultures, which may be time-consuming, resource-intensive, and subject to human error. This paper presents an innovative solution: an automated web application for diagnosing ringworm infections using machine learning and image processing techniques. The proposed system leverages a Convolutional Neural Network (CNN) to analyze clinical images of skin lesions, accurately identifying ringworm infections. The web-based platform is designed to be user-friendly, allowing both healthcare professionals and the general public to easily upload images for diagnosis. The model is trained on a diverse dataset of skin lesion images, using image preprocessing techniques to enhance quality and consistency. The results demonstrate high accuracy, precision and recall, indicating the potential of this approach to improve diagnostic speed and reliability. Future enhancements are discussed, including the expansion of the dataset, improvement of the model's accuracy  and integration with telemedicine platforms.