AI4Health: Collaboration for the Future of Healthcare – USC Viterbi

Photo: PhonlamaiPhotos

You need to be screened for a disease – would you rather have a photo taken or undergo extensive, expensive and invasive genetic testing?

It sounds too good to be true, but it’s a real question thanks to advances in artificial intelligence and the work of Wael Abd Almagedscientific director in USC Information Science Institute (ISI) which is using AI and facial recognition analysis to accurately predict congenital adrenal hyperplasia, a disease that causes mild facial changes.

This is just one example of the work being done by researchers at ISI who are coming together to form the Center for Research in Artificial Intelligence for Health (AI4Health).

The center headed by a director Michael Patzanichief scientist at ISI, will focus on research that enables breakthroughs in ethical algorithms and artificial intelligence systems to improve healthcare, fight disinformation and analyze big data.

Finding the intersection of AI and medicine

Patzani said, “ISI is already using AI for health research, one of the goals of AI4Health is to do so more systematically to make it easier for medical school researchers to find people with AI expertise.”

With this goal in mind, AI4Health will hold a number of events in collaboration with Keck School of Medicine at USC. The first event is scheduled for Thursday, December 1, 2022, from 11:00 AM to 1:00 PM on the USC Health Sciences Campus. At this event, six ISI researchers and six Keck researchers will each give five-minute talks on their work. Guys explained that these events will seek to “find intersections between Keck and ISI and increase the number of collaborations.” Register at

A wealth of data

“Health data has become much more abundant in recent years,” Patzani said. Electronic health records, genomic data, information from sensors and wearables, and medical images are all ripe for AI analysis. Information can also be gleaned from scientific journal publications and social media posts, both of which continue to grow rapidly in quantity.

And this level of big data is where AI and machine learning work best: looking for patterns in data, extracting information from text (ie magazines and social media) and making predictions based on data analysis .

AI4Health will use AI to take advantage of growing amounts of health data as well as find solutions to the challenges that come with big data.

AI4Health Research Areas

Data management
For data to be useful, researchers must be able to find it; it is useful if it is selected, organized and annotated; and should be made available or distributed to interested parties. Making it all happen is what is known as data managementand several ISI researchers have been active in this space as it relates to health.

Carl Kesselman, Fellow of ISI and director of the information systems research department, created the pipelines and workflows that enable FaceBase 3 Data Management and Integration Center to collect and process huge data sets on craniofacial and dental development in humans and animal models. All of these are accessible to the wider craniofacial research community with a purpose Advances in Craniofacial Development and Malformation Research.

Yigal AhrensISI’s senior administrative director and interim director of the artificial intelligence division, and his team has worked for years with the National Institutes of Health and the National Institute of Mental Health to create NIMH Repository and Genomic Resources (NRGR). The NRGR is a collection of biospecimens and data from people with diagnosed mental health problems and their relatives. Repository datasets are made available to researchers to stimulate research and development by providing timely access to primary data and biomaterials.

Important work like this – work that facilitates the use of the wealth of health data available – will continue as part of AI4Health.

Knowledge discovery and data analysis
Thanks to the wealth of health data, researchers are able to use AI to understand patterns that can lead to breakthroughs. This often means analyzing electronic health records, medical images, or data from wearable sensors to discover new connections.

What does this look like in practice? The work of the ISI Senior Researcher Greg Ver Steeg that he found predictive factors for Alzheimer’s disease in the patient’s medical records.

Or head of ISI research Abigail Hornework of to understand the behaviors that lead to diet-related illnesses. Horne has connected vast amounts of mobile phone mobility data and health data to show that the food environment is strongly associated with diet-related diseases. The research also analyzes digital restaurant menus to determine the quality of food available to communities, hopefully paving the way to more effective public health policies or interventions for demographic groups affected by poor diet.

But there are some health data that may not look like “health data” at first glance. Social media posts, for example. Emil Ferrara, head of the ISI research team, works to counter social media manipulation and misinformation on a range of public health issues: COVID-19 conspiracies; anti-vax campaigns; tobacco advertising; and the blending of politics and public health policies online.

Another dataset ripe for knowledge discovery and analysis is the ever-increasing number of e-journal publications. With AI, they can be analyzed to create databases of health information.

Precision health
“Knowledge discovery refers to the study of how to use machine learning to find patterns in data,” said Pazzani, who followed by explaining that precise health refers to “finding the risks of disease and the treatments that will work best for each person.”

Priority for Keck School of Medicine at USC, precision health uses the identification of genomic data or other factors to improve the health of a subset of the population. This could mean tailoring treatment to a group of patients, looking at a virus with a specific genome, and more.

Pazzani gave an example: “There are a number of drugs for Parkinson’s disease that unfortunately are only about 25 percent effective, but for a certain group of patients they are 90 percent effective.”

This is where AI steps in. He continued: “So if you can understand the relationship between a patient’s genetic background and a drug, then you can tailor a drug to a specific patient or a specific group of patients.”

And this type of analysis can have tangible consequences: “It’s hard to get FDA approval for something that’s 25 percent effective. Getting something approved that is 90 percent effective for people with a certain genome is much easier.

Machine learning for health
AI and machine learning (ML) can also be used to make clinical decisions by suggesting diagnoses or recommending interventions to clinicians. AbdAlmageed’s work uses facial recognition analysis to predict congenital adrenal hyperplasia is a case in point, and many researchers at ISI are already heavily involved in this area.

Pazani, who has extensive experience in machine learning, has worked on using ML to detect cognitive impairment, recommend treatment for HIV patients, analyze chest X-rays, diagnose glaucoma, and more. AI4Health researchers, including Pazzani, will continue their work on ML for health while seeking new opportunities and applications for ML in the healthcare space, with the goal of creating both better patient experiences and improved health outcomes.

Improving patient experiences can also be done through telehealth by using AI systems to support remote healthcare. AI can analyze chat text, voice and images to provide quick feedback to doctors or patients. Again, AbdAlmageed’s analysis of these facial changes is a great example of this, but there is a wide range of how telehealth can be improved with AI.

Patzani said: “It can be a chat doctor visit to decide what type of doctor you should see. Or maybe we can give reassurance and say “take two aspirin and call me in the morning” for some people. And others we may see as urgent and take them to the right specialist.”

Catalyzing research and looking for breakthroughs

More than a dozen researchers already working on AI research applicable to health will join the AI4Health initiative. With Patzani as director, the center will be co-led by ISI Wael Abd Almaged, Jose-Luis Ambite, Abigail Horne and Greg Ver Steeg as co-directors. This team, along with researchers from ISI and USC will work to catalyze research, seek discovery and most importantly, work to improve health outcomes for patients.

Posted on November 22, 2022

Last updated on November 22, 2022

Leave a Comment

Your email address will not be published. Required fields are marked *