AI meets eye care in USC medical student Ryan Shean’s research

USC medical student Ryan Shean discovered his calling to become a physician through his own health emergency. As a 13-year-old growing up in Orange County, California, he faced the excruciating symptoms of a kidney stone. He still remembers the calm and reassurance of his surgeon, and the difference it made.

“The experience shifted my perspective,” said Shean, now in his fourth year at the Keck School of Medicine of USC. “It introduced me to significant pain, but also to the impact of compassionate providers. That doctor took the time to comfort me, which allowed me to get through a really challenging moment. I hope to emulate that compassionate care as I become a physician, particularly in ophthalmology where restoring vision can profoundly change a patient’s independence and quality of life.”

In addition to training for a career in service of patients, Shean contributes to the future of eye care through his research alongside Benjamin Xu, MD, PhD, associate professor of ophthalmology and clinical scholar at the Keck School of Medicine and chief of the Glaucoma Service at the USC Roski Eye Institute. The team explores the role that artificial intelligence can play in improving clinical diagnosis, patient care and medical education for glaucoma, a disease that damages the optic nerve and can lead to vision loss.

Shean’s work has earned him a Medical Student Eye Research Fellowship from Research to Prevent Blindness, which provides funding support for the year he is dedicating to scientific investigation.

“We’re deeply grateful to Research to Prevent Blindness for investing in Ryan with this extraordinary and well-deserved honor,” said Xu, director of data science and AI at the USC Roski Eye Institute. “Research to Prevent Blindness has a distinguished history of identifying and supporting future leaders in vision science, and this highly competitive award reflects national recognition of Ryan’s exceptional promise at an early stage in his career.”

AI’s potential as a tool for improving eye care

Among a larger portfolio of glaucoma research, Xu, Shean and their colleagues are pursuing an AI-based approach for improving detection of the disease.

Glaucoma diagnosis involves evaluating demographic factors, pressure within the eye, acuity of the vision, and structural and functional tests that reflect the health of the optic nerve. If the disease evades discovery, patients risk missing the chance for early therapies that help prevent vision from deteriorating.

“There’s so much information to synthesize, and different physicians synthesize it differently,” Shean said. “A clinical decision-support tool could lead to more consistent diagnosis, so that all patients receive the highest level of care.”

With Xu, Shean is testing whether large language models and generative AI chatbots can be part of the solution. The researchers feed the range of diagnostic criteria into both existing chatbots and custom algorithms trained using relevant medical information. The machines’ diagnostic decisions are compared with the consensus from human experts.

So far, the performance from AI has approached that of specialists. The sweet spot for the algorithms tends to be in identifying moderate to severe cases of glaucoma, where they could potentially augment physicians’ ability to screen for the disease.

“It shows there’s promise for artificial intelligence in standardizing vision care,” Shean said.

At the same time, the researchers are appropriately guarded about large language models’ tendency to “hallucinate,” confidently relaying information that’s entirely false.

“Hallucinations are a significant barrier to translating our findings to the clinical setting,” Shean said. “The large language models, and the specialized ones as well, need to be as accurate as possible.”

Shean stresses that the goal behind bringing AI into medicine is to offer doctors tools that make them more effective, not to replace human practitioners with machines.

“I want to make sure that this technology can be used to benefit physicians and improve patient care, and that the relationship between physician and patient is uncompromised,” he said. “As I experienced with my own medical emergency, there’s a great role for physicians to play, person to person.”

Learning across the lab and the clinic

Xu has served as Shean’s mentor since his first year as a USC student. Shean credits Xu with showing him everything it takes to go from an idea for a study to a polished manuscript. Under Xu’s mentorship, Shean has developed multiple research projects spanning glaucoma surgery outcomes, medical education in ophthalmology and AI. Xu has also demonstrated how to build a bridge from the research literature to patient care.

“I’ve learned directly from Dr. Xu what it means to be an ophthalmologist,” Shean said. “He’s constantly thinking about how he can apply the latest technologies and research to his clinical practice.”

Shean has even worked on a study that influenced treatment. Investigating a laser procedure for a subtype of glaucoma, the team identified the area of the eye where the procedure was most effective. Xu then integrated that finding into his surgical approach.

“It’s so rewarding to see this research, where I worked behind the scenes, making a difference in patients’ lives,” Shean said.

Following Xu’s example, Shean sees his studies at USC building toward a career that combines patient care and research.

“I’m really excited about the future of ophthalmology,” Shean said. “I want to continue to be a part of the conversation about integrating artificial intelligence into care — and doing it in a way that centers the patient-physician relationship.”

Learn more about the Keck School of Medicine’s MD program.