<![CDATA[The skin disease tinea, caused by a dermatophyte fungal infection, is a significant health concern and can affect the quality of life of sufferers. Early detection of this disease is essential to prevent its spread, especially in areas with limited specialized medical personnel. This study aims to develop a tinea skin disease detection system using the Convolutional Neural Network (CNN) method, which can classify three types of tinea, namely Tinea Pedis, Tinea Manuum, and Tinea Corporis. The dataset used consists of 1,146 images of skin lesions equally divided into three categories, with each category containing 382 images representing different stages of disease symptoms. This dataset was processed through preprocessing techniques, including image cropping, scaling, contrast adjustment, and data augmentation to improve the training quality of the model. The developed CNN model has a structure of 8 convolutional layers and was trained using 80% training data and 20% validation data. The training results showed that the model achieved 75% accuracy on the training data and 85% on the validation data after 20 epochs, with consistent loss reduction. These results show that the CNN model can detect tinea skin disease with high enough accuracy and can be used as a diagnosis aid for medical personnel, especially in areas that lack specialists. The developed web-based application allows users to upload images and receive diagnosis results directly, providing convenience in early detection of tinea skin disease. This research makes an important contribution to the development of technological solutions in the improvement of health services in areas with limited medical resources.]]>
Copyrights © 2026
