Courses in technical skills for AI in medicine coming soon
The lessons will teach the programming skills and analysis required for deeper understanding of AI
Nov. 15, 2022— After the College of Medicine successfully launched a course on the fundamentals of artificial intelligence terminology and applications in medicine earlier this year, additional courses will soon enable clinicians and physician-scientists to dive deeper into the topic.
Courses II and III in AI applications from the College of Medicine are expected to become available early next year to all faculty and trainees at the college who want to learn about the programming skills and analysis required for a deeper understanding of AI algorithms and how they can be applied to medical research to improve patient care.
The courses are being developed by Christopher R. Giordano, M.D., chief for the division of liver and transplant anesthesiology and a professor of anesthesiology; François Modave, Ph.D., a professor of AI in the department of anesthesiology and Patrick Tighe, M.D., M.S., associate dean for AI application and innovation and the Donn M. Dennis, M.D., Professor in Anesthetic Innovation.
Modave said Course II: A Basic Understanding of Coding for AI in Medicine is expected to be available by early February 2023 and will focus on programming and coding in the Python language, with four hands-on modules that may take participants approximately five to six hours to complete.
“It won’t make you an expert, but it will give physicians the fundamental skills they will need for programming in AI,” Modave said.
He said the course build upon key algorithms discussed in course I — including logistic regression, support vector machines and random forests — to show how physicians can develop those algorithms and apply them to their own research.
Course III: An Introduction to Deep Learning in Medicine is expected to become available in May 2023 and focuses on deep learning algorithms as they apply to image analysis, an important component of interpreting results obtained by AI algorithms.
“A clinician can only assess so many images in a given day, whereas AI tools can process thousands of images in a matter of minutes or faster,” Modave said. “AI allows us to automate pre-screening to reduce the number of images clinicians have to review. There are always cases that are borderline, where a clinician’s eye becomes essential, but AI can sort out very efficiently all the cases of a clear ‘no issue’ or clear ‘yes, there is something here that requires a closer look.’”
All three AI courses from the College of Medicine will eventually be offered for continuing medical education credit and will count toward UF’s distance learning AI certificate.
Modave said colleagues outside the College of Medicine have expressed interest in learning more about these topics. A rollout among other colleges in the academic health center will likely take place soon, before expanding to include UF affiliates. The courses will eventually be available for purchase by the public.
“We have a significant number of rich data sets from patients that can be analyzed, but some of it is impossible for providers to comb through themselves,” Modave said. “With recent technologies, we’re able to make those analyses quickly and provide real-time feedback to researchers and providers, which can then be used to improve health care delivery.”