Detecting early-stage Parkinson's with Artificial Intelligence

Parkinson's is among the most difficult diseases to diagnose in its early stage. This fact has inspired research by a team of MIT scientists, who have developed an artificial intelligence (AI) model that can detect Parkinson's through people's nocturnal breathing patterns, long before it manifests with typical visible symptoms (tremors, stiffness, imbalance, slow movements). This could also be an important breakthrough in supporting the development of new medicines and therapies for the treatment of this neurodegenerative disease that, along with Alzheimer's, is spreading rapidly in contemporary society, with about 1 million people afflicted in the US alone.

The core of this technology is the neural network – a set of algorithms that interact with each other to mimic the functioning of a human brain – that is trained to recognize and interpret breathing patterns. Based on specific evaluation criteria, this AI model can not only confirm the actual presence or absence of the disease, but also assess its severity and track its progression over time. "Some medical studies have shown that respiratory symptoms manifest years before motor symptoms, meaning that breathing attributes could be promising for risk assessment prior to Parkinson's diagnosis," says Dina Katabi, head of the research team. All this diagnostic potential is contained within a box that resembles a common Wi-Fi router in appearance and size. It extracts nocturnal breathing signals either from a specifically designed belt worn by the person or from radio signals that bounce off their body while asleep.

Beyond the benefit of earlier diagnosis, this technology also promises to open up new possibilities in drug development and clinical care, as Katabi points out: "In terms of drug development, the results can enable clinical trials with a significantly shorter duration and fewer participants, ultimately accelerating the development of new therapies. In terms of clinical care, the approach can help in the assessment of Parkinson's patients in traditionally underserved communities, including those who live in rural areas and those with difficulty leaving home due to limited mobility or cognitive impairment."

Did you like this article? Take a look at our healthcare overview to learn more about future challenges and opportunities in healthcare and the innovations that are shaping trends in this industry.?

AUTHOR

Sofia Cosima Pelanda Mazza, Supertrends