SKIN DISEASE CLASSIFICATION USING MACHINE LEARNING

Abstract

Ellulitis, impetigo, athlete foot, ringworm, cutaneous larva migrans, chickenpox, and shingles. After comparing the performance of the models ResNet50, ResNet101, DenseNet121 and DenseNet169, with respective accuracies of 93%, 94%, 75% and 74%, ResNet50 was selected for deployment. This model not only ensures high accuracy but also minimizes overfitting, making it more suitable for integration into the web application. With a user-friendly interface, the system will allow users to upload images, and it will automatically determine the type of skin disease they need to diagnose. The user-friendly interface, built with Flask and Vue, allows users to upload images, automatically classifying and diagnosing skin conditions. The system will utilize Flask and OpenCV libraries to process requests and handle images. Additionally, deployment techniques for machine learning models on the web will be integrated to ensure quick and reliable feedback. The application of this system is not limited to the medical field; it extends to the online user community, allowing them to quickly make preliminary predictions about the condition of their skin and seek advice from healthcare experts if necessary.