An Eye on Artificial Intelligence

December 7, 2022 by T. Y. Alvin Liu, MD, assistant professor of ophthalmology at the Wilmer Eye Institute

Combining human strengths with technological innovations to provide better patient care

Many people are familiar with virtual assistants, like Tesla’s self-driving function or Amazon’s Alexa and Apple’s Siri. But did you know ophthalmologists are using similar artificial intelligence (AI) technologies to enhance patient care?

Johns Hopkins assistant professor of ophthalmology Alvin Liu
Dr. Liu is partnering with the Johns Hopkins Whiting School of Engineering to integrate artificial intelligence into precision medicine as part of a university-wide initiative. The team is starting with the retina with the goal of employing images and AI to personalize ophthalmic care for all patients.

In general, the concept behind AI is to make a computer or machine act as though it were human. Though there are many different types of AI, deep learning is one form that is particularly good at pattern recognition in image and language analysis. The very nature of deep learning was inspired by how the human brain processes visual information, and its rise is particularly important in medical fields that employ imaging.

Ophthalmology is one of the leaders in the medical deployment of deep learning, and the retina — which we examine via imaging — is at the forefront. We’re using this technology to better treat major blinding diseases like age-related macular degeneration, diabetic retinopathy, and retinal vein occlusion. All impact clear, sharp vision and can negatively affect day-to-day activities. Age-related macular degeneration is especially common and is the leading cause of central vision loss in people over the age of 50 in the U.S., when patients often engage in very visual activities, like reading, watching television, and enjoying their grandchildren.

We are using AI to produce more precise predictions, tailor treatment, and give better prognosis for patients. These aren’t just shiny research tools — they’re powerful technologies that we are integrating into clinical practice to provide patients with instantaneous feedback and predictions. With this information, we can fine-tune an individual patient’s management and improve final vision outcomes.

Our goal is not to replace doctors. Instead, we’re adding more tools to our toolbox. Like our use of superhuman X-rays and CT scans, what’s most powerful is how we employ these tools to enhance the patient-provider interaction. These technologies give clinicians more time to do what we, as humans, are good at. Building trust, strong connections, and good relationships with patients is the advantage humans have over computers.

This story first appeared in the Fall 2022 edition of Planning Matters. 

Make Your Gift

Interested in supporting cutting-edge research at the Wilmer Eye Institute?

Topics: Friends of Johns Hopkins Medicine, Johns Hopkins Medicine, Fuel Discovery