Artificial Intelligence in Dermatology: Challenges and Perspectives

Abstract

Artificial intelligence (AI) based on machine learning and convolutional neuron networks (CNN) is rapidly becoming a realistic prospect in dermatology. Non-melanoma skin cancer is the most common cancer worldwide and melanoma is one of the deadliest forms of cancer. Dermoscopy has improved physicians’ diagnostic accuracy for skin cancer recognition but unfortunately it remains comparatively low. AI could provide invaluable aid in the early evaluation and diagnosis of skin cancer. In the last decade, there has been a breakthrough in new research and publications in the field of AI. Studies have shown that CNN algorithms can classify skin lesions from dermoscopic images with superior or at least equivalent performance compared to clinicians. Even though AI algorithms have shown very promising results for the diagnosis of skin cancer in reader studies, their generalizability and applicability in everyday clinical practice remain elusive. Herein we attempted to summarize the potential pitfalls and challenges of AI that were underlined in reader studies and pinpoint strategies to overcome limitations in future studies. Finally, we tried to analyze the advantages and opportunities that lay ahead for a better future for dermatology and patients, with the potential use of AI in our practices. Artificial intelligence (AI) is the development of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, and translation between languages. The research on the use of AI in dermatology includes the ability of a machine to correctly classify a skin lesion. Skin cancer is the most common cancer worldwide and melanoma is the deadliest form of skin cancer. All skin cancers have a better prognosis when detected early in their development, hence their early detection is of paramount importance. Dermatologists use a dermatoscope—a specialized magnifying lens to improve their diagnostic capacity. However, even with the use of the dermatoscope, their ability to recognize skin cancer is far from perfect. AI has the ability to learn from dermoscopic images and subsequently provide an image-based diagnosis. Several studies compared the performance of machines and humans in classifying skin lesions from these images and showed that machines can classify skin lesions as good (and sometimes better) than humans. However, the application of AI in everyday clinical practice remains a challenge. In this article, we attempt to summarize the limitations and challenges that researchers found in their studies, and we provide strategies to improve the design of future studies. Finally, we describe the advantages and opportunities that could lay ahead for a better future for dermatology and patients.