Parkinson’s Disease (PD) is a neurological disorder that causes uncontrollable movements, balance issues, and coordination difficulties. Millions of people live with Parkinson’s, yet there is no standard way to screen for it. As a result, many patients are misdiagnosed or diagnosed only after the disease has progressed. Research scientists from the Massachusetts Institute of Technology (MIT) have developed an artificial intelligence (AI) system capable of detecting Parkinson’s disease by analyzing nocturnal breathing patterns. This AI technology identifies the presence of Parkinson’s based on how a person breathes while sleeping, potentially allowing for early detection, which is crucial in preventing further brain damage.
There are two methods of collecting nocturnal breathing signals. The first is through a breathing belt, and the second involves using a wireless signal device that transmits low-power radio waves, which bounce off the body while the individual sleeps. The data collected is used to train a neural network that predicts whether or not a person has Parkinson’s. The model showed 90% accuracy based on one night of data and 95% accuracy when using 12 nights of breathing patterns. Additionally, this AI-based system could predict the severity and track the progression of Parkinson’s Disease over time for individuals.
This AI offers several potential benefits. It aims to provide a non-invasive, low-cost, at-home assessment for diagnosing and monitoring the progression of PD. Individuals would be able to track their condition from anywhere. Pharmaceutical companies could also use this system to monitor changes in Parkinson’s patients during clinical trials, helping to shorten trial durations, reduce costs, and accelerate the drug development process. While this AI isn’t a substitute for a biological test, it can assist physicians in diagnosing Parkinson’s until such a test is available.
Although this AI system represents a significant advancement in the early diagnosis of Parkinson’s, concerns remain about the potential for false positives. This highlights the need for further testing to refine the algorithm. Another concern is accountability—if the AI produces an incorrect diagnosis, who would be held responsible? In cases of misdiagnosis, people often seek to hold someone accountable for the mistake. But when artificial intelligence is involved, it’s unclear whether blame would fall on the developers or the physicians using the AI. Despite these challenges, this AI may represent the first step toward a biological test for Parkinson’s, emphasizing the importance of nocturnal breathing patterns in its diagnosis.
WellPath Partners is your senior resource referral guide. Follow us on ALL social media platforms and join us weekly for more content and public health discussions.
By: Cindy Birrueta
Office Support Specialist at WellPath Partners
B.S. in Healthcare Administration at California State University, Long Beach
