AI has the potential to serve as “an indefatigable super-doctor,” helping us stay healthier and live longer, says Prof. Hengyong Yu of the Department of Electrical and Computer Engineering.
“AI can analyze life and medical data for each individual, offering personalized recommendations for diet, medication delivery and exercise,” Yu says. “AI models, when trained on the clinical diagnosis process, clinical datasets and test results, can surpass top physicians in efficiency, diagnostic accuracy and speed.”
AI can also enhance the performance of medical devices and diagnostic tools, says Yu, who received two grants totaling more than $4.7 million from the National Institutes of Health to improve CT scan images.
The first grant will help Yu’s team develop an AI-based image-reconstruction algorithm that would effectively “freeze” the beating heart during a CT scan. This will eliminate the blurring movement of the coronary arteries in X-ray images and help doctors better analyze plaque buildup on the walls of the arteries, which is the main cause of heart attacks.
The second grant will develop a computer program for a new generation of CT scanners, improving the image quality and resolution of photon-counting CT technology by using the power of AI for 3D color CT imaging and reconstruction.
Computer Science Prof. Hong Yu, founding director of UML’s Center of Biomedical and Health Research in Data Sciences (CHORDS), says the intersection of health care and AI is “full of golden opportunities.”
Yu has won grants for AI research projects to help detect physician errors and to identify people at risk of suicide.
“We can save a lot of lives because of AI,” she says. “All these major advances in medicine, to a large extent, are because of AI.”
Beyond doctors’ offices and hospitals, a team of UML researchers is involved in a national effort to develop AI-based smart home technologies to improve the quality of home care for the country’s rapidly aging population. Called the AI Institute for Collaborative Assistance and Responsive Intervention for Networked Groups (AI-CARING), the initiative was launched in October 2021 with a $20 million grant from the National Science Foundation.
Computer Science Prof. Holly Yanco, who is leading UML’s efforts on the project, says AI shows great promise in helping caretakers and people with minor cognitive impairments manage and perform daily tasks like meal prep, scheduling appointments and medication management.
“We don’t want AI to take the place of somebody having hands-on care from a doctor or nurse. That personal interaction is important,” Yanco says. “AI can assist us with health care, but it doesn’t replace it. It’s a tool, like an X-ray machine or any other diagnostic machine.”—EA and JG