Teenage Tech Stories: Alay Shah

Each month, Tech for Good speaks to one teenage entrepreneur about their incredible achievements in the world of tech, and how they’re contributing to making the world a better place.

Name: Alay Shah
Age: 17
Born: Texas, USA

Achievements: Alay is a researcher, aspiring innovator and high school senior. He earned seventh place in this year’s Regeneron Science Talent Search, for the development of a diagnostic tool that tracks eye movement to identify neurological disorders that he hopes can become a low-cost alternative to MRIs.

My research surrounds how we can use the eyes and the way our eyes move in order to understand deeper brain function; things like how diseases work and what we can do to prevent them, treat them and diagnose them.

I’ve been working on this research for about three years, and I’m currently working on turning that research into a product that actually affects people worldwide.

My goal is to create the most portable diagnostic tool for concussions, that can be used in third world countries, or in places like the sidelines of sports games or the military.

My interest in eye movement stems from football, funnily enough. When a player gets hit really hard on the football field, they’re taken aside, and the medic does a quick test on the patient to determine if there’s some kind of underlying concussion or minor traumatic brain injury.

That made me think: If this works for concussions, why are we not using it for other similar neurological diseases?

The first thing my device does is movement tracking. It has two cameras that go underneath the eyes and they’re basically tracking how the eyes move. If there’s a tremor or a twitch in the eye, this is able to catch that.

The second step is called gaze estimation. And this is using information from movement tracking, along with the way a light reflects on the eye, in order to create this mathematical computation of where you’re looking at on the screen. So, if you put a picture of the Mona Lisa in front of you, gaze estimation is looking at whether you’re looking at the background of the painting or at the foreground.

To tie everything together, I use what’s called a recurrent neural network. It’s a type of pattern recognition algorithm; and it’s taking both movement, tracking and gaze estimation metrics and attempting to find abnormal patterns between the two of them.

Patients have to look at a dot that moves across the screen and complete a task. From that, the algorithm is able to identify these really specific and minute abnormalities that occur in patients. But, to test it, I actually did need human participants.

I worked with the Lone Star neurology clinic in Frisco. And they basically supplied me with any patient that was willing to take a really quick five-minute test, so I was able to get about 200 patients.

I noticed that Parkinson’s patients have a hard time reacting when the dot changes direction, and there is almost always a tremor that occurs. In dementia patients, their eyes kind of just wander around the screen. And then, for multiple sclerosis patients, when there are two dots on the screen, they tend to shift between looking at the two instead of choosing one to follow, like most people do.

I’ve also been working on how we can use eye-tracking data to basically generate MRI scans. Everybody has heard of MRIs. But the issue is that it uses this big, clunky machine, and you can’t get it to the places that need it the most, like rural areas, third world countries and the military.

I want to create what’s called a functional MRI, which is basically a type of MRI that works on the brain that attempts to plot based on blood oxygen levels, where activities occurred.

I created this completely new tool that shows a patient 16 strategic images to induce different emotions and different kinds of cognitive and motor functions. And then, by looking at the way our eyes move, we can create a baseline for what normal eye behaviour looks like.

It can be pretty daunting for a 15-year-old to walk up to a clinic and be like ‘Hey, I’m a 15-year-old with an experimental neurological test. Can I test it on some of your most vulnerable patients?’ I had to work on things like pitches and proposals, and contact many people to make sure that the patients were safe.

The Regeneron Science Talent Search is like this big thing that everybody wants to get to, so I was very surprised when I got the top 300, and then even more when I got to the top 40.

I’m going to Yale, under the Hahn scholarship. And there, I want to work with professors at the Wu Tsai Institute and the Yale New Haven Hospital so that any concussion patient that comes through the hospital can be matriculated into my clinical trial.

My plan is to make the biggest-ever concussion clinical trial with eye-tracking technology so that, when I come out of college, I can attempt to do FDA approval and take it to market.

I want to work on democratising healthcare and collaborate with companies and startups that can effectively bring healthcare to the masses rather than keeping it for an elite few.