University of Southern California Launches Center on AI Research for Health

November 28, 2022 - Researchers from the University of Southern California (USC) Information Sciences Institute (ISI) announced the launch of the Center on Artificial Intelligence Research for Health (AI4Health), a collaboration aimed at improving healthcare and patient outcomes by enabling breakthroughs in ethical AI use, fighting misinformation, and analyzing big data.

“ISI has already been using AI for health research, one of the goals of AI4Health is to do it more systematically to make it easier for medical school researchers to find people with expertise in AI,” said Michael Pazzani, PhD, director of AI4Health and principal scientist at ISI, in the press release.

To foster these connections between AI and medicine, AI4Health will hold several events in collaboration with the Keck School of Medicine at USC. These events will help “find intersections between Keck and ISI and increase the number of collaborations,” Pazzani explained.

AI4Health’s research is also set to take advantage of the increasing amount of health data available for analysis. These data will support work in multiple research areas: data management, knowledge discovery and data analytics, precision health, machine learning (ML) for health, and telehealth.

Data management helps to facilitate the useability of health data. AI4Health researchers in this area are working on several projects that involve curating, organizing, annotating, and distributing data for health-related projects, including curating craniofacial and dental development data to advance research into craniofacial development and malformation.

AI4Health’s work in knowledge discovery and data analytics aims to allow researchers to analyze EHRs, medical images, and other health data using AI to tease out relevant patterns that may lead to breakthroughs. One project in this area is leveraging cell phone mobility and health data to link food environment and diet-related diseases.

AI4Health researchers are also using AI to find the most appropriate disease treatments for specific populations to advance precision healthcare, which uses genomic data to improve population health. For example, Pazzani notes in the press release that many drugs for Parkinson’s disease are only somewhat effective for a broad patient population but over 90 percent effective for a subset of those patients. AI can help researchers understand the relationship between a patient’s genetic background and the drug, which can lead to tailored treatments.

AI4Health researchers are also applying ML to health research, with one scientist using the technology to support facial recognition analysis to predict congenital adrenal hyperplasia, a disease that causes mild facial changes.

Finally, the researchers are using AI to support telehealth and remote healthcare. AI4Health researchers are using AI algorithms to analyze text in chat, voice, and images to provide quick feedback to clinicians or patients, which can improve care quality and patient satisfaction.

“This could be a doctor’s visit on chat to decide which type of doctor you need to see,” Pazzani explained. “Or perhaps we can give reassurance and say ‘take two aspirin and call me in the morning’ for some people. And others, we might see that it’s an emergency and we’ll get them to the right specialist.”

More than a dozen researchers have already joined the AI4Health initiative. The team will continue to work to catalyze research and look for breakthroughs to improve health outcomes, the press release notes.

Other institutions are also launching initiatives focused on leveraging AI for healthcare.

Earlier this month, the University of Maryland Strategic Partnership: MPowering the State (MPower), a research and innovation collaboration between the University of Maryland, Baltimore, and the University of Maryland, College Park, launched a partnership with the University of Maryland Medical System and Montgomery County, Maryland, to establish the University of Maryland 3 - Institute for Health Computing, which will leverage AI and computing to diagnose, prevent, and treat diseases.