Skin Disease Diagnosis System Using Machine and Deep Learning Algorithms

Automating the diagnosing process has shown promising results in recent years due to the advancements in deep learning approaches. In this work, we provide a unique skin disease diagnosis system that uses Convolutional Neural Networks (CNN) and Generative Adversarial Networks (GAN) to provide effective diagnosis. In our work, the system is trained on the HAM10000 dataset, which contains a diverse collection of skin lesion images categorized into seven classes, including melanoma, basal cell carcinoma, and others. GANs are used to address the challenges of dataset imbalance and missing data by generating realistic synthetic images that augment the existing dataset and CNNs are used for image pre-processing, image classification, etc. Utilizing these techniques, our proposed system achieves an impressive classification accuracy of 90.7%. In conclusion, using deep learning methods helps our research improve the skin disease diagnosis system and also gives medical professionals a useful tool for the accurate identification of skin diseases.