In a podcast from the RSNA 2017 medical imaging conference, a veteran radiologist reflects on AI hype versus reality and the new generation of radiologists growing up with AI.
CHICAGO -- With nearly three decades of experience as a radiologist, Stephen Chan, M.D., is a wise mentor for younger radiologists coming into the field.
But Chan, an associate clinical professor of radiology at Columbia University Medical Center's Harlem Hospital, is also a tech-savvy veteran who embraces AI in radiology, a movement that is sweeping the medical imaging community.
At RSNA 2017, medical imaging's signature gathering -- with annual attendance of about 50,000 making it North America's biggest healthcare conference -- Chan moderated an educational panel of young radiologists and radiology residents.
AI, machine learning and deep learning
The topic was AI in radiology. Of course, AI in radiology is an umbrella that encompasses various forms of machine learning, including deep learning, the AI variant that is probably the most widely used in today's AI-assisted healthcare imaging applications.
After the panel, SearchHealthIT recorded a podcast with Chan in which he reflects on AI and machine learning technologies and how natural they appear to be for the new generation of radiologists who are incorporating them into their work.
Chan notes that the new doctors are already well-equipped to assimilate AI in radiology because they are already so familiar with technology itself.
"The technological facility of the new crop of radiology trainees exceeds the average that was in the past. The basic familiarity with things like coding and computer science is much more second nature to this current generation," Chan says in the podcast.
AI speculation and reality
Chan also speaks to the most obvious question that hovers around discussions of AI in radiology.
Stephen Chan, M.D.associate clinical professor of radiology, Columbia University Medical Center's Harlem Hospital
"Clearly we've heard a lot of hype when we've heard predictions, including from the father of AI, Geoff Hinton, who predicted that radiologists would disappear within five years," he says. "I think there are various technical, as well as healthcare reasons why that will not occur, but certainly we expect AI to be a major part of radiologists' everyday professional activities in the future."
Like many imaging professionals at the Radiological Society of North America's 103rd Scientific Assembly and Annual Meeting, Chan says the teaming of man/woman with machines can do better radiology than radiologists working alone, without learning computers.
Chan refers to the analogy invoked in an RSNA 2017 speech by noted radiologist Keith Dreyer, M.D., that pictured a "centaur" chess player proposed by former world champion grandmaster chess player Garry Kasparov, comprised of a human and a machine.
That paradigm has played out in reality, with human-machine chess teams facing off against computers, each other and all-human teams.
"It was clear that these partnerships were better than teams of humans alone or teams of computers alone or an individual player or an individual computer," Chan says in the podcast.
That human-machine construction in radiology "is certainly a viable option for the future," Chan says. "How it unfolds, I'm not sure."