A SYSTEMATIC REVIEW OF AI-BASED IMAGE PROCESSING MODELS FOR PERSONALIZED DIAGNOSIS AND SEVERITY ASSESSMENT OF SKIN DISEASES

This systematic review provides a thorough analysis of the current state of AI-based image-processing models used in diagnosing and assessing the severity of skin diseases. The review synthesizes recent advancements in deep learning models, exploring various methodologies employed in dermatological image analysis. While significant progress has been made in developing AI tools for skin disease diagnosis, the review identifies critical challenges that hinder the clinical adoption of these technologies. Among the most pressing issues are the lack of data diversity, insufficient integration of patient-specific information, and limited generalizability of models across different skin types and conditions. The review also highlights a major gap in current research: the frequent omission of demographic and clinical data, which are essential for creating personalized diagnostic tools. Furthermore, there is a notable absence of models that can accurately assess disease severity—a crucial component for effective treatment planning and management. These shortcomings underline the necessity for more comprehensive data collection strategies, including the incorporation of multi-modal datasets that encompass diverse patient populations. In addition to data improvements, the review emphasizes the need for the development of more robust and generalizable AI frameworks. Such frameworks would enhance the accuracy and reliability of AI diagnostics in dermatology, making them more applicable in real-world clinical settings. By addressing these gaps, the review offers valuable insights and practical recommendations for future research. Ultimately, this work aims to contribute to the advancement of equitable, personalized, and effective dermatological care through the integration of cutting-edge AI technologies.