Durham, N.C.-based Duke Health is putting healthcare artificial intelligence to use in surgery scheduling and predictive analytics, DukeMed Alumni News reported Nov. 13.
Across the system, AI models predict operating room times needed for each procedure, optimizing surgical suite use and surgeon time, according to the story. A shared calendar is built into the EHR.
"The secret sauce is that we went to the end user at the beginning and asked, 'Is this going to make their life more efficient?" Wendy Webster, interim assistant vice president for Duke Health perioperative services and director of clinical operations and healthcare analytics for Duke Surgery and Duke Neurosurgery, told the magazine. "Then you're side by side with the surgeon or the anesthesiologist or the nurse validating the data."
AI is also analyzing mammograms to predict disease earlier, with imaging a big part of Duke Health's recent AI partnership with Microsoft, according to the story. But researchers are also developing predictive analytics models that are both more effective and unbiased, including via a simple algorithm that asks patients about their risk factors, health status and impressions of their own health.
"That's telling me that patient engagement in risk prediction is very important," Michael Pencina, PhD, director of Duke AI Health and chief data scientist for Duke Health, told the news outlet. "We need to better understand which variables are important, which ones we need to be collecting, and how to build these algorithms in a more equitable way."