Big promises with big data

Filed Under:
Eye Care
Share on:
Add to Twitter Add to Facebook Add to LinkedIn
As an ophthalmologist with a passion for health IT, P. Lloyd Hildebrand, MD, FACS, has thought a lot about the use of artificial intelligence in clinical practice. So as he and his colleague, H. Jay Wisnicki, MD, delivered a talk on the subject earlier this month at AAO 2017, Hildebrand was fairly certain that he could predict what his audience’s biggest question would be.

“It’s usually something like, ‘Are machines going to replace us?’ And my answer is always the same,” says Hildebrand, professor emeritus in the Department of Ophthalmology at the University of Oklahoma. “This technology isn’t going to take away our jobs, it’s going to empower us and our profession. It’s going to make us better physicians, and it’s going to allow for better healthcare.”

"This technology isn’t going to take away our jobs, it’s going to empower us and our profession. It’s going to make us better physicians, and it’s going to allow for better healthcare."
Leading the “Healthcare Revolution”
According to an analysis by the research firm Frost & Sullivan, the global market for AI and cognitive computing in healthcare will expand from $600 million in 2014 to more than $6.6 billion by 2021. “The need for data mining and decision-making has put AI-enabled solutions at the forefront of the healthcare revolution,” the firm reports. “AI facilitates greater accessibility, relevancy and actionability of healthcare information.”
Hildebrand, who in addition to practicing as an oculoplastic surgeon, founded the retinopathy screening service Inoveon, agrees. “Diabetic retinopathy is a perfect example of a disease we could fight more effectively with AI.” With the right solution, an ophthalmologist might profile individual patients against population-level data sets to optimize their treatment and predict outcomes over time. “I see it as a way of customizing care,” he explains. “It’s taking a patient’s clinical data, putting it into the context of the disease process, and then presenting that information to the clinician so they can provide treatment quickly and accurately.”
Overcoming AI’s Challenges
Despite the predicted boom in the AI market, there are still a number of challenges that must be overcome before the technology gains widespread acceptance. AI systems, for example, will need to be tailored to meet the needs of different specialties. And AI-assisted treatment of any disease will depend on people actually seeing their providers in the first place. “With diabetic retinopathy,” says Hildebrand, “we have one of the most algorithmic diseases in medicine—ideal for machine learning and artificial intelligence. And yet it’s still a leading cause of vision loss and blindness because so many of those who have it never see an ophthalmologist.”
Technology can help solve that problem, he says, by putting diagnostic testing “in the hands of the people who care for patients with diabetes” (November is National Diabetes Month). That would ensure that high-risk patients are at least identified, “so they can cross the bridge over to ophthalmology.”
And once they’re in that ophthalmologist’s office? That’s where big data and artificial intelligence can come into play. “If you think about how healthcare works now,” Hildebrand says, “we have so much information about so many things, so it can be difficult to manage it and use it in meaningful ways.” His vision for the near future involves clinicians of all kinds doing the work that they’ve always done, only far more efficiently and with better results. “It’s coming,” he says. “And I think it’s an opportunity we won’t want to miss.”

Originally posted on: 11/17/2017 7:22:44 AM

<< back to blog home
Alternate text
Shannon Reilly
Shannon Reilly is a Product Marketing Manager for Watson Health Imaging.

Search by Topic

Search the Blog

Never miss a post from Watson Health Imaging. Subscribe to the blog.    Subscribe

Never miss a post from Watson Health Imaging.