“Can you imagine a robot telling somebody that they had amyotrophic lateral sclerosis? Or that they had Alzheimer disease? Or that they had Creutzfeldt-Jacob disease and were destined to die? …There’s nothing formulaic about it. It’s something that you learn over the course of years, how to convey that information.”

With health systems beginning to invest in the use of artificial intelligence (AI), the development of a physician-AI interface is an ongoing topic in healthcare. Physicians who are not already interacting with these systems will need to understand the role that AI can, and likely will, play in medicine. In neurology, however, there may be some difficulties.

As Joseph Berger, MD, pointed out in a conversation with NeurologyLive at the 2019 American Academy of Neurology (AAN) Annual Meeting, May 4-10, in Philadelphia, the neurologist’s reliance on physical examinations and rudimentary-style tools such as reflex hammers and tuning forks will make them difficult to be replaced by AI altogether. However, the associate chief of the MS division and professor of neurology at Penn Medicine does envision some relationship between the two, with AI augmenting the ability of the physician to do their job.

He detailed how AI might be integrated into the diagnostic process, acting somewhat as a failsafe in the instance that some bit of information slips through the cracks. As well, he shared insight into some of the challenges this dyad will face, including the relaying of information to, and development of a relationship with, the patient.

For more coverage of AAN 2019, click here.