AI identifies Parkinson’s from breathing patterns
A new artificial intelligence (AI) model from researchers at MIT can detect Parkinson’s disease, a notoriously difficult disease to detect, through breathing patterns.
The tool could help alleviate the disease onset gap between when symptoms of Parkinson’s first appear and when most people receive a diagnosis. The MIT team that developed the AI model uses a neural network to analyze breathing patterns that occur during sleep and published their findings in Nature Medicine.
The study included 7,671 individuals and used data from several U.S. hospitals and multiple public datasets. In addition to detecting Parkinson’s to a reliable degree, the AI model could also estimate the severity and progression of the disease in accordance with the Movement Disorder Society Unified Parkinson’s Disease Rating Scale.
Roughly 1 million people in the U.S. live with Parkinson’s, and 60,000 people are diagnosed every year. According to MIT researchers, it is the fastest-growing neurological disease in the world, with a yearly economic burden of $52 billion. Currently, most people are diagnosed based on physical symptoms, such as tremors and rigidity. However, symptoms do tend to appear years after the onset of the disease, typically a late diagnosis. MIT researchers aimed to prove that AI could be a useful tool for noninvasive, at-home assessment of Parkinson’s and provide risk assessment before diagnosis.
“A relationship between Parkinson’s and breathing was noted as early as 1817, in the work of Dr. James Parkinson. This motivated us to consider the potential of detecting the disease from one’s breathing without looking at movements,” said Dina Katabi, one of the authors of the paper and Thuan and Nicole Pham Professor of Electrical Engineering and Computer Science at MIT. “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.”
Pulling from multiple datasets and an observational study conducted by MIT, researchers analyzed 11,964 nights from 757 Parkinson’s subjects and 6,914 control subjects. The data were divided into two groups, including the breathing belt datasets, which comes from sleep studies that uses a breathing belt to record a person’s breathing through the night, and the wireless datasets, which collects nocturnal breathing through a contactless radio device that analyzes radio reflections from the environment to extract breathing signals.
The study found that AI can identify people who have Parkinson’s disease from their nocturnal breathing, revealing a potential new digital biomarker for the disease. According to researchers, the findings could lead to an easier way to diagnose and assess Parkinson’s patients through measurements at home, even in contactless measurements through wireless signals to detect breathing. The approach could reduce the time and cost of Parkinson’s clinical trials and potentially facilitate drug development, in addition to aiding in earlier detection of the disease. Current Parkinson’s disease drug development has a $13 billion price tag and 13-year timeline.
In the future, researchers envision their AI model being deployed in the homes of Parkinson’s patients and those at high risk for the disease to monitor their status and provide feedback to providers.
“We’ve had no therapeutic breakthroughs this century, suggesting that our current approaches to evaluating new treatments is suboptimal,” Ray Dorsey, a professor of neurology at the University of Rochester and Parkinson’s specialist and co-author of the paper, told MIT News. Dorsey adds that the study is likely one of the largest sleep studies ever conducted on Parkinson’s. “We have very limited information about manifestations of the disease in their natural environment and [Katabi’s] device allows you to get objective, real-world assessments of how people are doing at home. The analogy I like to draw [of current Parkinson’s assessments] is a street lamp at night, and what we see from the street lamp is a very small segment. [Katabi’s] entirely contactless sensor helps us illuminate the darkness.”
Amy joined TriMed Media as a Senior Writer for HealthExec after covering home care for three years. When not writing about all things healthcare, she fulfills her lifelong dream of becoming a pirate by sailing in regattas and enjoying rum. Fun fact: she sailed 333 miles across Lake Michigan in the Chicago Yacht Club "Race to Mackinac."