H22 Artificial intelligence: whose brains were behind it?

Abstract Long considered futuristic, artificial intelligence (AI) has slowly but surely come to prominence in medical spheres over the last 90 years. Dermatology is poised to benefit from emerging AI technologies, with its inherent visual diagnostic methods and the increasing volume of digital photographic data underscoring its suitability for AI-augmented patient care. Algorithms have been around for centuries, but it was Alan Turing who was the first to formalize algorithm and computation in the 1930s, along with cracking the Enigma code during World War II. In 1943, the first artificial neural network of electrical circuits was modelled to simulate biological neurones in the brain, a concept crucial to the success of AI today. The term ‘artificial intelligence’ was first coined by John McCarthy in 1956 at Dartmouth College. Computer-aided diagnosis was first applied in the ana­lysis of pulmonary nodules detected in chest radiographs in 1963. MYCIN, Edward Shortliffe’s doctoral dissertation at Stanford University in the 1970s, was the first powerful use of AI, whereby a computer program correctly identified the causative bacteria in cases of bacteraemia and meningitis and even recommended antibiotic treatments. In 1998 the first Food and Drug Administration (FDA)-approved computer-aided diagnosis system was used in mammography. After the 2010s, a subfield of machine learning called deep learning emerged, and from then on AI flourished, creating meaningful applications across many fields, not least in dermatology. As dermatology is a morphological feature-dependent discipline, machine learning models are increasingly implemented as a diagnostic support tool using image analysis in areas including dermatopathology and skin cancer detection. MelaFind was the first FDA-approved digital cancer detection unit to identify melanoma, succeeded by Nevisense, and recently by DermaSensor, which was approved by the FDA in January 2024. The visual nature of dermatology lends itself well to transformative AI-augmented practice. Although in its infancy, with the correct government and regulation, AI has the potential to help practising dermatologists deliver better skincare.