Detection and classification of skin diseasesusing deep learning models

Monkeypox, a skin disease, is one of the diseases that can spread globally in epidemic form. Monkeypox is particularly similar to chickenpox and measles in its symptoms and skin lesions. Therefore, accurate diagnosis is a critical factor directly affecting treatment success. Traditional diagnostic methods rely on manual processes such as visual examination and dermatologist-performed biopsies, which can be time-consuming and require expertise. In this context, Artificial Intelligence (AI) technologies play a significant role in diagnosing monkeypox. This thesis focuses on the use of AI-based systems in the accurate diagnosis of skin diseases. It examines how AI techniques, such as deep learning and machine learning, are used to classify skin disease lesions, distinguish between monkeypox, chickenpox, and measles lesions, and, more generally, to analyze dermatological images. Furthermore, the integration of AI-based diagnostic systems with specialized imaging techniques, particularly dermoscopy images, is discussed. The thesis highlights the potential of deep learning algorithms to achieve high accuracy in diagnosing skin diseases and compares different models. Furthermore, the potential of AI-assisted systems as a helpful tool in dermatologists' decision-making processes is discussed. Obstacles to the use of AI-based systems in clinical practice, such as data privacy, algorithm accuracy, and clinical approval processes, are also among the important issues addressed in the thesis. Ultimately, this study reveals the potential of AI for detecting skin diseases and offers suggestions on how these technologies can be used more widely and effectively in the healthcare sector.