Skin Cancer Classification Using Hybrid Deep Learning Approaches: A Systematic Review

Skin cancer is one of the most common types of cancer worldwide, and early diagnosis is crucial in the fight against the disease. In recent years, deep learning and hybrid approaches have shown significant potential in the diagnosis of skin cancer. This study provides a detailed review of hybrid deep learning methods proposed for skin cancer classification between 2020 and 2024. The methods in the literature are compared in terms of the datasets used, model accuracy rates, and performance- enhancing techniques. The ability of hybrid models to capture time- series dependencies and manage uncertainty makes them an effective tool for skin cancer diagnosis. These findings demonstrate that hybrid approaches provide more comprehensive and reliable results compared to traditional methods in skin cancer diagnosis. In the future, the development of larger, more balanced datasets and the integration of explainable artificial intelligence techniques will further support progress in this field.