Fairfax Co. students use AI to create an algorithm for classifying brain signals – WTOP

2 minutes, 26 seconds Read
Fairfax County students Gautham Ramachandran and Sriram Nalini used AI to create an algorithm that can be used to classify EEG signals, which can be used to control prosthetics. (Courtesy Gautham Ramachandran)

About seven months ago, Chantilly High School student Gautham Ramachandran came across a man in a wheelchair on the sidewalk.

Ramachandran said he was inspired by the man who was navigating life without his legs. Ramachandran approached him, and they exchanged pleasantries.

In the days that followed, Ramachandran reflected on his desire to help people like the man he spoke to.

“I was wrestling with a pressing question: How can we leverage technology to provide autonomy and empower individuals like him?” he said.

Ramachandran and his classmate, Sriram Nalini, realized the answer to that question involved artificial intelligence. They used AI to create an algorithm that can be used to classify electroencephalogram (EEG) signals, which can be used to control prosthetics.

“It’s a key to a new realm of interaction, where your thoughts seamlessly translate into action; where technology is an invisible, intuitive extension of the human experience,” Ramachandran told WTOP.

Ramachandran has been programming since he was in third grade, and used to create video games using HTML and JavaScript. Instead of playing the video games, though, he said he “got addicted to creating them,” and spent hours coding.

When he was in middle school, Ramachandran said he released a game on the platform Steam. But none of that compared to his latest efforts.

Ramachandran said he and Nalini considered different types of available technology for analyzing brain signals. In some cases, patients get implants, which Ramachandran said are innovative but “require a level of invasiveness that isn’t really ideal for everyone.”

A traditional EEG headset is another alternative, but Ramachandran said those are complicated “due to needing so many electrodes, and frankly, not as accurate as we’d hope, especially when it comes to classifying movements.”

So, the two students took the EEG technology and streamlined it.

“We basically used this AI algorithm as the embodiment for this vision,” Ramachandran said. “It’s designed to work with just a handful of electrodes.”

The innovation is helping with accuracy, he said. Previously, the maximum accuracy he’s seen is 98%, “which means that there’s going to be two mistakes out of every 100 movements.”

With the help of the algorithm, Ramachandran said there’s “classification accuracy that reaches up to the zenith, up to 100% accuracy, which is really not just an improvement, it’s a revolution.”

The technology, Ramachandran said, “allows people to basically control prosthetics or control devices using their brain signals.”

The pair published their findings in a research paper.

Their work, he said, helps “provide a cost-effective solution to prosthetics.”

Get breaking news and daily headlines delivered to your email inbox by signing up here.

© 2024 WTOP. All Rights Reserved. This website is not intended for users located within the European Economic Area.

This post was originally published on this site

Similar Posts