Season 3 · Episode 7
F048 AI in healthcare 2/6: Radiology disrupted? (Woojin Kim)
The idea that AI will replace radiologists comes from the fact that today’s AI models models are very good at pattern recognition. But in reality, the “rich data” coming from radiologists is in the radiology reports, which are to a large extent unstructu...
September 19, 201953m 26s
Audio is streamed directly from the publisher (podtrac.com) as published in their RSS feed. Play Podcasts does not host this file. Rights-holders can request removal through the copyright & takedown page.
Show Notes
The idea that AI will replace radiologists comes from the fact that today’s AI models models are very good at pattern recognition. But in reality, the “rich data” coming from radiologists is in the radiology reports, which are to a large extent unstructured. As elsewhere, the 80:20 rule applies here, says Woojin Kim. So the interesting thing are the NLP models mining radiology reports, he said in this interview. So what do radiologists actually do, are they going to start talking to the patients more, why they will not be obsolete anytime soon and more. Woojin Kim is the Chief Medical Information Officer at Nuance, former Chief of Radiography Modality, Director of Center for Translational Imaging Informatics, Associate Director of Imaging Informatics, and Assistant Professor of Radiology at the Hospital of the University of Pennsylvania.