Machine learning and deep learning based psoriasis recognition system: evaluation, management, prognosis—where we are and the way to the future

Psoriasis, a chronic inflammatory skin disease, poses significant diagnostic challenges due to its heterogeneous clinical manifestations and overlap with other dermatological conditions. The inconsistencies in diagnostic standards and variability in disease presentation further exacerbate the controversy surrounding the diagnosis of psoriasis. Recent advances in artificial intelligence (AI), particularly deep learning (DL) techniques such as convolutional neural networks (CNNs) and transformers, have shown promising results in improving recognition of psoriasis. This comprehensive review aims to provide an in-depth examination of the current state of AI applications in the diagnosis of psoriasis, highlighting the strengths and limitations of existing approaches. We evaluate the performance of various AI models on publicly available datasets, including the Psoriasis Image Dataset (PID) and the International Skin Imaging Collaboration (ISIC) dataset, and compare their accuracy, sensitivity, and specificity with traditional diagnostic methods. Furthermore, we discuss the challenges and future directions of AI-powered psoriasis diagnosis, including the need for large, diverse datasets, addressing data quality and generalization issues, and developing explainable AI models. Our review highlights the potential of AI to revolutionize the diagnosis and management of psoriasis and provides a pathway for future research in this field.