Implementasi Arsitektur CNN DenseNet-121 untuk Identifikasi Autoimun Kulit dengan Augmentasi Data

Autoimmune skin diseases are conditions in which the immune system mistakenly attacks healthy skin tissue, causing inflammation, tissue damage, and changes in skin color. The similarity of symptoms among various autoimmune skin diseases, such as psoriasis, lichen planus, vitiligo, hidradenitis suppurativa, and dermatomyositis, presents a challenge for accurate and timely diagnosis. This study was conducted to support the diagnostic process by utilizing deep learning technology, specifically the Convolutional Neural Network (CNN) method with the DenseNet121 architecture. Data augmentation techniques were also applied in this study to increase dataset variation, allowing for a performance comparison between the original dataset and the augmented dataset. The results show that the CNN model with the DenseNet121 architecture, configured with a batch size of 32 and 60 epochs on the augmented dataset, achieved a high accuracy rate of 92.43%. The model was then implemented into a web-based application and integrated using the Flask framework.