Meet the Teenager Who Uses AI and 3D Printing to Diagnose Diabetic Retinopathy
A teenager from Virgina came up with an app that can sensor diabetic retinopathy, an eye disease that can lead to blindness. Using 3D printing helps making diagnosis technology affordable and widely available.
3D printing gets more and more important in medicine. That is why we at drupa like to shift your attention to these unconventional fields that employ 3D printing as a pillar of their technology development. We have often covered exciting student research programs on our blog but none as important as this one: 16-year old Virgina high school student Kavya Kopparapu develops an app that uses a 3D-printed lens in order to detect diabetic retinopathy, an eye disease coming along with diabetes for about 30 per cent of the people suffering from it. Without early detection and treatment patients go blind in less than five years.
Necessity Is the Mother of Invention
Kavya’s grandfather is one of the patients living in rural India where there is no easy and affordable way of diagnosis. So the teenager built a solution by herself. Well, not entirely, of course. She had help from her older brother and a classmate. Together they use an architecture called a convolutional neural network (CNN) in order to build an app that learns to detect signs of diabetic retinopathy as well as microaneurysms and blood vessels by taking a simple photo with a smartphone camera.
Practice Makes Perfect and Leads to First Clinical Results
To help Eyeagnosis, so the name of the app, learn how to detect diabetic retinopathy, the team used 34,000 retina scans provided by the EyeGene database from the National Institutes of Health. The scans were often out of focus, blurry, and under exposed. “But that was actually a good thing. It’s very representative of the real-world conditions you’d get with using a smartphone,” says Kavya. After figuring out the technical teething they were quick to arrange an agreement with the Aditya Jyot Eye Hospital in Mumbai. The hospital received the prototype last November and managed to make five accurate diagnoses.
3D Printing Makes Complex Diagnosis Technology Affordable
These numbers don’t look like a lot at first sight but this technology isn’t about making the most accurate diagnosis technically possible. It’s about making a decent diagnosis technology as affordable as and thereby widespread as possible thanks to 3D printing. Of course, Eyeagnosis needs a lot more practice to work even more accurate, but one thing is for sure: Kavya Kopparapu is off to a great start.
Now Kavya Kopparapu is not a 3D printing entrepreneur but focuses her work rather on AI and coding. But since her technology is using 3D printing to such a strong degree, we’d be happy to welcome her and her team at drupa 2020.
What do you think about 3D printing in medicine? Is it a good way to make high-tech equipment affordable? Discuss it with us in the comment section!