Adam Dubis, PhD, is driving advances for artificial intelligence in eye care and vision research.
John A. Moran Eye Center researcher Adam Dubis, PhD, works at the intersection of big data, artificial intelligence (AI) algorithms, and hope.
The breakthrough tools and platform technologies he developed are helping ophthalmologists determine treatment plans and creating new opportunities for vision scientists to explore pressing questions. His most recent project aims to improve care and speed drug development for people and their families who are facing blinding inherited eye diseases.
“We have an incredible opportunity to use AI tools and ethically sourced data to improve how healthcare is accessed and delivered on a global scale,” said Dubis, a data expert who has traveled the world advising international bodies on health technology regulation. “At the end of the day, I’m creating tools that give patients hope for the future.”
AI FOR AMD
Age-related macular degeneration (AMD) is the leading cause of blindness for people 55 and older, impacting central vision that’s important for seeing grandchildren clearly and enjoying hobbies in your golden years.
There is no cure for it, but ophthalmologists do have treatment options for the neovascular or “wet” form of the disease in which abnormal blood vessels growing in the eye can leak and rupture. Retinal specialists can inject medication to prevent vessel growth and use retinal imaging to evaluate a patient’s progress.
The problem: Not all patients respond to the same treatment regimen equally, so care must be personalized. If injection appointments are scheduled too far apart, or if patients feel they are coming in too frequently and miss some appointments or even drop out of treatment, they could lose vision.
Dubis made headlines in 2025 when the company he co-founded launched the Treatment Planning Support (deepeye® TPS) algorithm, which uses tens of thousands of 3D scans of the retina combined with medical records to analyze disease progression and correlate them to treatment regimens. The tool, now available in Europe, was the first of its kind approved for use in the Western world. Its algorithm helps physicians make two decisions, acting like a second expert reader of the scans:
- Does the patient need to receive a treatment today or not?
- How many treatments will the patient need over the next 12 months?
The AI also displays the basis for the recommendation.
“The idea is that it will allow patients to keep their vision as long as possible and ensure they are only spending time going to a clinic when they need to,” said Dubis. “For physicians, it optimizes clinic flow, provides reassurance for the treatment course, and possibly helps in patient education or justifying changes to prescribed therapies.”
Discussions about evaluating the AI at the Moran Eye Center as part of a U.S. clinical trial are ongoing.
A WIDER DOOR FOR RESEARCH
Dubis is among the first to say AI is only as good as its vetted data sources and a person’s ability to easily access it.
This idea is at the center of the new Moran Phenotyping, Imaging, and Advanced Technologies (PHIAT) AI database, which contains close to 2 million health records. PHIAT allows researchers to easily ask myriad questions related to ophthalmology. For example, a clinician scientist might try to predict which patients with diabetes are most at risk for getting diabetic eye disease, or to create an algorithm that helps doctors identify when one drug isn’t working and it’s time to try another.
Dubis emphasizes that ethically sourced data powers his AI. In PHIAT, individual health records are de-identified, and the confidentiality of the data is strictly protected within data computing facilities using industry-leading security measures. PHIAT is unique among academic medical centers.
“While other universities have limited databases open to research, their data is preconfigured, whereas PHIAT is not,” said Dubis. “This allows for a much broader range of inquiries and investigations. We are opening the door wider for research in diseases that can be time-consuming and expensive to study.”
PRIORITIZING PATIENT GOALS
“There is so much out there in the world of research that maybe in my lifetime I’ll be able to get some vision back. But even if not, at least I will have done the work as best I can to try to ensure my daughter’s children or other people’s children don’t go blind."
Bob Pettipaw, 62, was just 15 years old when he learned he would slowly go blind from choroideremia, a rare, inherited eye disease impacting men in his family. His grandfather on his mother’s side was blind, as was his uncle.
Choroideremia advances slowly, starting with peripheral vision loss that moves inward later in life, so at first, Pettipaw did his best to ignore his diagnosis. He got married, started a family, built a successful plumbing career, and started his own business.
But as the disease advanced and he made the first of many visits to the Moran Eye Center, he realized he had to modify his approach to managing the disease.
“I made changes in how I lived my life and pursued my career,” said Pettipaw, now an advocate active with the Choroideremia Research Foundation. “There is a point when you just say, ‘So what? I’m not afraid, I’m looking forward.’ What helps me the most is when I can help others, especially people who are just starting their journey and may need some help from us old guys who have been through it.”
Pettipaw participated in an East Coast clinical trial for a new drug that didn’t make it to the market. While he says he experienced improvements in his vision and day-to-day functioning, the trial gauged success primarily through improvements in a single measure: visual acuity. Taken using an eye chart, visual acuity looks for improvement in a small region of central vision.
“Choroideremia is a condition that you will retain good central vision well into your 70s,” explained Pettipaw. “We’ve got 20/20 in our center, it’s our peripheral, our contrast, our light sensitivity that are impacted. So, what I am looking for in terms of a successful outcome is more functional-based, like mobility tests and things of that nature.”
These types of outcomes, known as patient-reported outcomes (PROs), focus on tasks most relevant to patients. Before and following treatment, is a patient able to navigate a room in dim light or drive at dusk better? Can they participate in sports, or recognize faces across a room more easily?
The use of PROs in blinding eye diseases has been fragmented at best. That’s where Dubis comes in.
He’s working with patients like Pettipaw, caregivers, clinicians, and researchers to create a unified set of PROs for multiple conditions and an accompanying comprehensive, multicenter clinical database. He envisions Moran as ground zero to use this information to create a new center for inherited eye disease research.
“Visual acuity measures central vision in a very small part of the eye, yet we see across a much larger field of vision,” explained Dubis. “Quantifying how vision occurs, how it changes during disease, and how it works post-treatment across the entire visual space is essential. PROs better capture post-treatment changes, and without them, drug companies must start from scratch for each trial. That slows development and adds cost.”
Pettipaw’s daughter, Anna, 25, is hopeful about research to find new treatments and a cure. She thinks about having children one day, but it’s a complicated question, knowing she could pass choroideremia on to a son. On some days, her answer is yes; on others it is a no.
“Research is my future, my family’s future, my kids’ future,” she said. “It is ever developing, always a positive light when things aren’t so positive, when they are dark.”
