Deep learning has received significant attention from the medical community as an important technology in radiology. At a recent virtual conference, however, the medical director for AI at the Mayo Clinic said RPA might have more impact on the medical specialty.
Radiology is using AI in imaging studies for things like segmenting and identifying pixels in an image and predicting the nature of a group of pixels for diagnosis. Bradley Erickson said radiologists could leverage RPA to determine, given predefined criteria, which AI tools should be applied to each imaging study. It can also be employed to create data used to train AI algorithms.
“I think this is also going to accelerate our ability to develop AI tools, because the data collection is going to be consistent, it’s going to be robust,” Erickson said during his address at the virtual conference. “And the same way it collects data for training purposes will also then be used in the clinical implementation, so the challenge of ‘drift’ is likely to be lower.”