SHASHWAT GUPTA RAY
CUNCOLIM: Researchers at NIT Goa have devised an AI-enabled model that studies brain signals to detect the onset of Parkinson’s disease at an early stage, using a low-cost, non-invasive method, which is now undergoing tests.
If you enjoyed reading this article, you can help support our journalism by adding OHeraldo as a preferred source.
“AI-powered Deep Learning (DL) can analyse massive brain signal data like Electroencephalogram (EEG). It can quickly and accurately identify patterns and anomalies that humans might miss, and help diagnose brain disorders like epilepsy, Alzheimer’s, or Parkinson’s,” Melina Maria Afonso, PhD scholar, Department of Computer Science and Engineering, NIT Goa and Assistant Professor in the Department of Computer Engineering, Goa Engineering College, told O Heraldo.
EEG detects abnormalities in brain waves, such as those caused by seizures, tumors, brain injury, or inflammation.
“It can even assist in creating brain-computer interfaces, like mind-controlled devices. Basically, AI can unlock insights into how our brain works and help improve treatments,” Afonso said.
“Researchers across the world are studying Parkinson’s disease to better understand how it affects the brain. Our work is part of this global research effort. We study the brain’s electrical activity, recorded using a safe and painless method called EEG, to look for patterns that may be linked to Parkinson’s disease. By using deep learning (DL) models that can study large amounts of data, we aim to understand whether these brain signals contain useful information that could help researchers learn more about the disease and, in the long term, contribute to the development of brain-computer interface-based tools for Parkinson’s research. By learning more about how Parkinson’s affects the brain, doctors could get clearer insights into the disease. It might also lead to gentler ways of studying brain conditions, methods that don’t require surgery or anything invasive,” Afonso said.
Eventually, this research could help create simple, technology-based tools (like lightweight handheld devices) that doctors could use to support their decisions when caring for patients.
“This technology was created to see if the brain’s electrical signals could give doctors new clues about the disease. Many scientists around the world are asking the same question, and our work is one piece of that bigger puzzle. In simple terms: it was devised because Parkinson’s is still a mystery, and we’re trying to see if listening to the brain’s signals can help unlock answers,” she said.
Explaining the functioning of this model, the researcher said that it works by first recording the brain’s electrical signals, like listening in on the “chatter” happening inside the brain.
“These signals are very complex, so we use smart computer programs to study them and look for patterns that might be linked to Parkinson’s disease. Because the computer ends up with huge amounts of information, we also use extra tools to simplify and organize it, so the system runs more smoothly and clearly shows what’s important. In everyday terms we ‘listen’ to the brain, let advanced computers find hidden patterns, and then tidy up the results so they’re easier to understand,” she added.
According to Dr Damodar Reddy Edla, Associate Professor, Department of Computer Science and Engineering, NIT Goa, right now, this technology is still in the “early research” stage, where scientists share results with each other to learn more and improve the methods.
“It’s not ready for patients yet, it’s still being studied and refined, but the hope is that, over time, it could grow into something useful in real medical care. As more studies like this add to the understanding of Parkinson’s, the combined insights could eventually help shape new medical tools that reach patients directly,” he informed.
