Presentation on Artificial Intelligence in Dermatology (Video on Demand) : LINK
-Continuing growth in the incidence of melanoma, alongside increased mortality associated with advanced disease is prevalent in Australia and many other populations highlighting the clinical need for improved early detection.
-Dermatology is about to undergo a leap forward with the introduction of artificial intelligence (AI) to assist with clinical care including diagnosis, surveillance and patient triage (1).
-Less than 5 years ago, convolutional neural networks designed for the image classification of skin lesions were shown to be able to deliver diagnostic support of comparable or greater accuracy than skin doctors and dermatologists.
-A recently published AI system distinguishes between 26 common skin conditions, representing 80% of cases seen in primary care, while also providing a secondary prediction covering 419 skin conditions with comparable accuracy to dermatologists (2).
-Modern digital infrastructure, advances in research and access to better-quality datasets for training AI are enabling rapid and ongoing growth in this field. There is great potential for AI to transform current dermatological practice and not only minimizes melanoma burden, but also facilitate new research into skin biology and other dermatological conditions.
1. Tschandl, Philipp, et al. \”Human – computer collaboration for skin cancer recognition.\” Nature Medicine 26.8 (2020): 1229-1234.
2. Liu, Yuan, et al. \”A deep learning system for differential diagnosis of skin diseases.\” Nature medicine 26.6 (2020): 900-908.
Presentation and description by Prof. H. Peter Soyer MD, FACD FAHMS (Australia) [World Skin Health Day 2022 -Guinea] (To find the presentation simply use the filter buttons from the event homescreen)