IRIS – Saving Sight

By Rick Outzen

Ask Dr. Sunil Gupta what he does, and he will say, “I take care of patients with some very sophisticated sort of therapeutics that have come into the market both in terms of the biologics as well as the surgeries and the laser and things like that.”

Huh?

He explained that 40 years ago, the leading cause of blindness in working-age patients was diabetic retinopathy, which damages the blood vessels in the tissue at the back of the eye. Today, diabetic retinopathy is still the leading cause.

“So we’ve come 40 years, and we haven’t really moved the needle significantly,” said Dr. Gupta in a recent phone interview. “However, in terms of innovation, we now can basically stop people from going blind.”

He continued, “So people talk about these moonshots with cancer and other things, but for diabetic retinopathy, we’ve got a solution that if we identify those patients at the right time, we can actually prevent them from losing any sight.”

If an ophthalmologist can diagnose diabetic retinopathy early enough, the same medicine used to treat the complication in later stages can actually reverse it by two levels. Dr. Gupta said, “If you can identify the patients just in time before they’re going to go into that high-risk phase of losing vision, we can then reverse their disease and prevent them from going over that edge.”

To help more patients keep their eyesight, more screenings need to be done earlier, which is why he founded Intelligent Retinal Imaging Systems, also known as IRIS. 

When he started IRIS in 2014, about 40% of the patients were getting screened, even though all patients with diabetes should have their eyes checked once a year for diabetic retinopathy. The fact that 60% of the patients were going undetected bothered Dr. Gupta.

“Why am I getting patients at a stage when I can’t fix them?,” he asked himself. “And I hate not being able to fix things. I hate losing to a disease state. I’m very competitive.”

Dr. Gupta and IRIS see diabetic retinopathy as their competition. He wants to get his patients as close to 20/20 vision as possible. The genesis of IRIS was to catch the complication earlier so that treatment can be started sooner and blindness avoided. 

“I’m an engineer by training, and I grew up since seventh grade with my father actually aggregating data on oil wells,” said Dr. Gupta. 

An environmental engineer with degrees from UC Davis and Princeton, his father built a computer code that tracked data from 10,000 wells. Dr. Gupta said, “We would analyze that data, and then we’d be able to say, ‘Look, there’s environmental contaminants that are going to contaminate the drinking water in X amount of time.’”

With the onset of cloud computing—a network of remote servers hosted on the internet to store, manage and process data—Dr. Gupta saw an opportunity to create a similar data analysis system for IRIS that would expand screening for patients.

“Microsoft introduced a system that created HIPAA compliance that made it very easy for entrepreneurs to set up a system in the cloud that could do things that IBM does for large hospital systems,” he said. “It also allowed somebody like me to create a system in the cloud that would connect IRIS to camera systems in primary care clinics.”

A technician in the primary care doctor’s office could take a picture of the retina with a completely automated camera. The photo is uploaded to the cloud where the software analyzes it and identifies the pathology. Then retina specialists and ophthalmologists can read the image.

“It’s a triage model that allows us to get the patient to the ophthalmologist and retina specialists just in time to be treated,” Dr. Gupta explained. “And so that is what IRIS does.”

Game Changer
He credited Quint Studer, founder of the Studer Group and an original board member of IRIS, for helping him test his systems and work out any kinks. Studer told him, “You got to do this thing one in a row.”

When Dr. Gupta asked him to explain what he meant, Studer said, “Let’s just pick one client and get it all right with them.”

IRIS took Studer’s advice and picked one client. The company submitted a bid for Harris County, Texas, which covers Houston and is the largest federally-qualified health clinic in the country. The patient population included underserved, underinsured people for all sorts of demographic groups. 

“They’re truly doing population health where the government, the federal government and Harris County, give them $1 billion,” he said. “But for that $1 billion, they got to take care of all the people that are underserved, that don’t have insurance or have Medicaid. And so they were struggling with what to do with these patients.”

IRIS won the contract and embedded the cameras in all 12 of their primary care clinics that were spread out in different neighborhoods all over Houston. Gupta said, “We trained up their technicians to do this. And lo and behold, their rate of screening for diabetic retinopathy was 40%. Today, they’re at 90%.”

He boasted, “Literally an underserved, underfunded organization has taken their quality measure for primary care—which is, ‘Hey, how many of your diabetics are you getting evaluated for retinal damage?’—from a one-star rating to a five-star rating, and I can tell you, there are very few five-star rated HEDIS compliant for diabetic retinopathy.”

Dr. Gupta knew that if they could turn around Harris Health, IRIS could be a game-changer.

“The beauty was that the retina clinic in that institution had a three-month wait for their patients that were severe,” he said. “We were able to pull out all the normal exams out of that clinic and put in there all the abnormal exams for the doctors to treat. And now, within a week, if somebody’s got a hemorrhage in the retina, they can get into that clinic.”

The work with Harris Health proved that if you could create the right structure and the vertical integration of primary care to a specialist centered around the patient, the model allows you to really have an effect on the population. 

IRIS also has been successful in catching other medical conditions during the screenings.

“Because diabetics also have hypertension and the number two cause of vein occlusion is diabetes, we’re picking up patients that have vein occlusions and they don’t even know they have it, and we can grab it early and treat it if it’s affecting their sight,” said Dr. Gupta. “We are now picking up things like scar tissues on the surface of the retina.”

The second biggest target for the screenings is a condition called glaucoma. He said, “The optic nerve gets damaged because the pressure in the eye is too high and you lose peripheral vision, but you don’t realize it because you’re losing it so slowly. So we are actually picking up patients that we suspect have glaucoma now as well.”

Other tech giants have noticed IRIS’s success. In late January, Microsoft’s AI for Good initiative announced its new $40 million, five-year program, AI for Health, that aims to use artificial intelligence to help solve health problems around the world. IRIS was one of a handful organizations selected for the grant program. 

Today, the IRIS team continues to rally around the mission to end preventable blindness. Since 2014, the IRIS solution has scanned over 564,000 diabetic patients, identifying 205,000 cases of pathology among them. The Pensacola-grown organization now employs close to 50 people, a number that continues to grow.