Aaron Masino, Dr. Gary Spitzer Endowed Distinguished Professor in Genomics, uses AI to support rare disease research

1. November 18, 2024

Aaron Masino isn't a researcher pursuing a cure or new treatment for a rare disease. He doesn't conduct or set up clinical trials.

But his research supports those who do.

Masino is the Clemson University Center for Human Genetics Dr. Gary Spitzer Endowed Distinguished Professor in Genomics and an associate professor in the School of Computing.

"His research fits perfectly with Clemson's commitment to the advancement of human health and Dr. Gary Spitzer's belief that collaboration, genomics, and the power of computer and AI technology are key to earlier diagnoses and better treatment of diseases," said Trudy Mackay, director of the Center for Human Genetics. "Dr. Masino was an obvious choice for the Spitzer endowed professorship because he truly embodies Gary's vision."

Dr. Gary Spitzer was one of the primary physician architects of stem cell transplantation at M.D. Anderson Hospital in Houston, Texas, and helped develop treatment regimens for Hodgkin's lymphoma.

The endowed professorship was made possible through a $1 million gift from Suzanne Spitzer.

Power of collaboration

"Gary believed in the power of collaboration. He envisioned Clemson and Upstate South Carolina as leaders in genomic research," Suzanne Spitzer said of her late husband, who died of cancer in 2021.

The Spitzers are originally from Australia. They moved to Texas in 1973 when the United States made a large investment in cancer treatment and issued a worldwide call for doctors and scientists to work on a cure. He worked in academic medicine for 26 years, including at M.D. Anderson, the Saint Louis University School of Medicine and Georgetown University Hospital. He published more than 300 manuscripts and articles and built strong relationships nationally and internationally.

In 1999, he left academia and moved to Greenville, South Carolina, to work in clinical practice at the Cancer Center of the Carolinas. From there, he started a private practice and then helped establish a new cancer program at St. Francis Hospital before leaving to start his own consulting and diagnostic company, working in biotechnology with a number of national molecular companies.

"Gary always wanted to be on the cutting edge of research and revolutionize treatments in clinical practice," Suzanne Spitzer said.

Masino's research focuses on developing new artificial intelligence methods to support biomedical research, particularly for rare diseases.

The AI methods he develops aim to help researchers discover disease-causing variants, optimize clinical trials and assist clinicians in diagnosing rare diseases faster. The research integrates formal biology knowledge with patient data from sources such as electronic health records and public biobanks.

AI's impact

"We're at this inflection point of AI's impact in the biomedical space broadly, and genomic diagnostic medicine in particular," said Masino, who joined the Clemson faculty in January.

Masino continued, "My research interests are largely at the intersection of AI and diagnostic medicine and disease understanding. I was looking for an institution with a strong research program and a strong underlying education program. With the Center for Human Genetics and its exciting precision medicine program, along with the biomedical data science and informatics program in the School of Computing, Clemson offered a perfect synergy of research and education in one environment."

Masino said his work supports clinicians to recognize and diagnose somebody with a rare disease.

"AI would help researchers predict which variants would cause disease and also determine the association between those variants and the resulting phenotype, meaning the observable characteristics of the disease, so what the disease looks like in a person on the clinical diagnostic side," he said. "For many people with rare diseases, it can take multiple years to get a molecular diagnosis, one based on a genetic cause, if ever. We're trying to accelerate that process."

40 million

There are estimated to be between 7,000 and 10,000 rare diseases impacting 30 to 40 million people in the United States.

Masino said if AI recognizes earlier that somebody may have a rare disease, even if it doesn't know what that rare disease is, it could make recommendations to clinicians about which diagnostic tests they should consider based on knowledge stored computationally about rare diseases.

"We're not trying to develop AI that is going to say, 'This is the answer.' But rather, AI is going to work hand-in-hand with the physician to get to the answer faster," he said.

The methods Masino is developing are disease agnostic, meaning they are not tailored to one specific disease but rather trying to address all phenotypes and their associations to various diseases.

He's also working on developing methods to help researchers design optimal clinical trials, something specifically called comparative effectiveness research.

"There are characteristics about rare disease mainly driven by small study cohorts that have implications about how you design a clinical trial," he said.

Masino will validate the methods based on a rare disease that scientists know something about. A proposal he's recently submitted would include cystic fibrosis and sickle cell disease.

"They affect different groups of people in the population. The diseases themselves have very different characteristics. The idea behind working with those two particular diseases is that if we can show the methods work for both of those, it provides more evidence for their generalizability," he said. "Plus, those happen to be important diseases and they're more common than other rare diseases."

Visions aligned

Masino said he feels his work aligns with Dr. Spitzer's vision.

"In directly integrating formal biology into new AI methods, we'll enable discovery of new knowledge for rare disease, which will lead to more efficient diagnosis and new treatments. That would reduce the burden for patients," he said. "I think that aligns very well with what Dr. Spitzer envisioned."

Before coming to Clemson, Masino served as vice president of clinical data science at AiCure from 2021-2023, where he led research on AI methods to characterize central nervous system disorders.

Prior to that, he served as an assistant professor in the Department of Anesthesiology and Critical Care in the Perelman School of Medicine at the University of Pennsylvania. He was a biomedical informatics scientist at the Children's Hospital of Philadelphia, where he researched the application and development of machine learning methods to challenges in pediatric medicine. Topics included machine learning model development for sepsis prediction, utilization of wearable device data to recognize stress in individuals with autism and deep learning methods to detect latent concepts in electronic health record data.

He received his Ph.D. in applied mathematics from the University of Central Florida. He also holds a master's degree in aerospace engineering from the University of Colorado and a bachelor's in mathematics from Rutgers University.

^{Want to Discuss?}

Get in touch and we will connect you with the author or another expert.

Or email us at [email protected]

Public link

Please use the above public link if you want to share this noodl on another website.