<![CDATA[Skin diseases are a common health problem around the world and affect different age groups. Data show that skin diseases are still a significant health problem in Indonesia, especially among the population. A quick and accurate diagnosis is essential to ensure proper treatment and prevent further complications. The research aims to develop a system of classification of skin diseases in humans using CNN with the EfficientNet B2 architecture, addressing the challenges in often time-consuming and error-prone manual diagnosis and leveraging advances in deep learning technology. The study used experimental methods to classify skin diseases using the MobileNetV2 Convolutional Neural Network (CNN) algorithm. The evaluation results showed that the EfficientNet B2 model had accuracy of 84,0%, precision of 85,0%, recall of 83,0%, and F1-score of 84.0%. This showed the model had a good performance in classifying different types of skin disease. Contributions to this research are in the fields of dermatology and artificial intelligence, as well as opening up opportunities for advanced research in image-based medical classifications.]]>
Copyrights © 2024
