Research Updates

Explore more of the Lees’ recent research that uses big data and machine learning to gain exciting new insights into what our eyes can tell us about our brain health.

UW Medicine Will Lead Study Arm of National AI Initiative

Dr. Aaron Lee and Dr. Cecilia Lee will be co-principal investigators of a new UW Medicine-led study arm funded by the NIH as part of its “Bridge to AI (Bridge2AI)” program. The NIH program aims to develop model datasets that medical scientists can broadly explore the potential of artificial intelligence algorithms. Using Type 2 diabetes as its model disease, the Lees’ team will collect data from a population-representative cohort of 4,000 participants across three sites in the US, balanced for three factors: disease severity, race/ethnicity and sex. The goal is to create a publicly accessible dataset that all researchers can access freely.

“The Eye: A Window to Our Health”

Watch Dr. Russell Van Gelder, UW Medicine Department of Ophthalmology chair, in conversation with Dr. Aaron Lee and Dr. Cecilia Lee. Find out how new AI algorithms can predict a patient’s blood pressure, BMI, age and even biological sex using an image of the retina. And learn how the Lees anticipate we’ll be able to use big data to noninvasively study the eye, better understand brain health and predict other diseases in the future.

Listen to this story:

When Cecilia Lee, MD, MS, and Aaron Lee, MD, MSCI, began analyzing the data of thousands of patients, they discovered something that had never been noticed before at the clinical level. Something that could only be revealed by looking at millions of data points: a link between eye disease and Alzheimer’s disease.

At the Roger and Angie Karalis Johnson Retina Center, part of the UW Medicine Department of Ophthalmology, vision researchers are using big data — extremely large data sets that can be analyzed with powerful supercomputers — to better understand, diagnose and treat eye diseases. This emerging field is known as computational ophthalmology, and in the next five years, work in this area could lead to significant breakthroughs in earlier diagnosis and intervention for neurodegeneration.

Making powerful discoveries possible

Just how big is big data? The world’s largest medical database is an ophthalmology registry known as the IRIS Registry; it includes billions of data points associated with 65 million Americans. Using this massive collection of diagnoses, treatment regimens and outcomes, researchers can analyze vast amounts of medical data, helping them to see trends and connections that might be missed at a smaller scale.

As clinician-investigator faculty with the UW School of Medicine Department of Ophthalmology, Cecilia and Aaron use big data to study eye diseases such as age-related macular degeneration, diabetic retinopathy and conjunctivitis. Aaron specializes in artificial intelligence, algorithms and machine learning — the study of how algorithms improve through experience and making mistakes. Cecilia’s interests are more epidemiologic, focusing on how big data can be applied to clinical epidemiology and improving patient care.

“I’m interested in taking the newest technologies being developed at the schools of computer science and engineering and applying them to accelerate medical discoveries,” says Aaron. “And Cecilia brings her epidemiology perspective. So our skill sets are very complementary.”

As husband-and-wife scientists, any topic of interest is an opportunity for collaboration. Indeed, it’s when their research interests overlap — like using algorithms to analyze terabytes of retinal imaging — that they make some of their most exciting discoveries.

Predicting the risk of Alzheimer’s

Cecilia has always been fascinated by the eye as a literal window into the brain. “The eye is unique because it’s a direct extension of the central nervous system,” she says. “It’s developed from the same tissues that make up our brain, so it’s not such a far-fetched idea that the diseases you see in the eye may represent certain pathologies in the brain.”

Cecilia and Aaron’s work on retinal biomarkers — changes to the structures of the retina, such as tissues or blood vessels, that may indicate disease — began with a study of 25 years’ worth of data from over 3,900 participants. The study, a collaboration with Kaiser Permanente Washington, found that people with diabetic retinopathy, glaucoma or age-related macular degeneration had a higher risk of developing Alzheimer’s than people who did not have an eye disease. Her results were so promising that they led to a $17.2 million grant from the National Institute on Aging to take the study further.

Now, Cecilia is diving deeper, asking patients to take part in state-of-the-art ophthalmic and brain imaging and to share detailed clinical histories. Researchers — including Aaron — will use this extensive data to search for new retinal biomarkers that may predict an increased risk of Alzheimer’s. Her ultimate goal is a reliable, non-invasive and affordable screening tool that can identify those at risk of developing Alzheimer’s disease so they can get earlier access to treatment.

A tipping point in medicine

But the Lees don’t just collaborate with each other. They’re also working with Russell Van Gelder, MD, PhD, chair of the UW Medicine Department of Ophthalmology and director of the Karalis Johnson Retina Center, to establish a new center for computational ophthalmology that will help their colleagues.

“Our vision of the center is not only to provide expertise in order to accelerate research in our department, but also to recruit research participants for personalized medicine or precision-medicine projects,” says Cecilia. “We want to make sure that our clinicians and patients can participate in this important research.”

“We are so fortunate to have attracted Cecilia and Aaron to the University of Washington,” says Van Gelder. “Their unique and complementary skills give them the ability to answer questions that have resisted analysis until now. Both are at the forefront of their fields. They are wonderful scientists, physicians and colleagues.”

The Lees’ research home is the Karalis Johnson Retina Center. The new center was opened in 2019 by Angie Karalis Johnson, who is passionate about finding a cure for macular degeneration and helping people with vision problems.

“Our entire clinic space and lab are here because of Angie’s generosity and foresight in supporting innovation,” says Aaron. And continued donor support will also make future projects, like the center for computational ophthalmology, a reality.

“The Seattle philanthropic community has been critical to developing our research programs. Through extremely generous gifts like Angie’s, we are able to provide the infrastructure to allow Aaron and Cecilia to develop their cutting-edge programs. We are so indebted to our donor community,” says Van Gelder.

In the next few years, the Lees anticipate major developments in computational ophthalmology — and more big questions to answer together.

“I believe medicine is at a tipping point with these exciting artificial intelligence techniques,” says Aaron. “But how does that help us understand diseases better? Can we use these technologies to change the way we deliver care? In the next five years, our field is going to be struggling with a lot of those issues. And we want to be at the forefront of guiding how that thought develops.”

By Stephanie Perry
Illustration: David Hoyt