An interdisciplinary project has received funding to help ensure that if — or more likely, when — certain imaging tools that use artificial intelligence (AI) are put to clinical use, their inherent uncertainty is considered as part of any subsequent clinical decisions, including guiding treatment.
The project, led by Abhinav Jha, assistant professor of biomedical engineering at Washington University in St. Louis’ McKelvey School of Engineering, will be funded by a $314,807 grant from the National Institute of Biomedical Imaging and Bioengineering, part of the National Institutes of Health (NIH).
The two-part project will involve first developing a technique by which to quantify the amount of uncertainty in AI-based tools used to measure quantitative parameters from patient images, such as tumor volume.
Jha then will work with a team including Anya Plutynski, professor of philosophy in Arts & Sciences. The two worked together previously to better understand patients’ attitudes toward AI. For this project, they also will work with radiation oncologist Clifford Robinson, MD, and nuclear medicine physician Tyler Fraum, MD, both at Washington University School of Medicine, to develop a questionnaire for patients that will determine how much risk someone is comfortable with if using an AI-based tool as part of the clinical decision-making process, given the certainty of the measurement the tool provides.