Device May Help Diagnose Parkinson’s From Breathing Patterns
MIT-developed artificial intelligence can detect respiratory symptoms early
The Massachusetts Institute of Technology (MIT) has developed a device that uses artificial intelligence (AI) to detect the presence and severity of Parkinson’s disease through patients’ breathing patterns.
Because respiratory symptoms occur at early stages of the disease, researchers proposed that breathing assessment may help in diagnosing Parkinson’s at early stages.
“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,” Dina Katabi, PhD, professor in the department of electrical engineering and computer science at MIT and principal investigator at MIT Jameel Clinic, said in a press release.
“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,” she said.
This new approach was described in the study, “Artificial intelligence-enabled detection and assessment of Parkinson’s disease using nocturnal breathing signals,” published in the journal Nature Medicine.
The diagnosis of Parkinson’s is usually made by evaluating symptoms mainly related to motor function, such as tremor and rigidity. However, these symptoms tend to appear in a late stage, delaying these patients’ diagnosis. As such, there is an urgent need for new diagnostic biomarkers, particularly those that can detect the disease early.
Additionally, over the last years, researchers have studied cerebrospinal fluid (the fluid that surrounds the brain and spinal cord) and neuroimaging as potential methods to detect Parkinson’s. However, such approaches are invasive, expensive, and require access to specialized medical centers, hampering regular testing that could lead to an early disease detection or continuous monitoring of its progression.
Testing the device
A team of researchers from MIT, in collaboration with the University of Rochester, Mayo Clinic, and Massachusetts General Hospital, have developed an AI device that can assess whether a person has Parkinson’s by analyzing breathing patterns that occur during sleep.
The device was tested on a large database of 7,671 people, 757 of whom had Parkinson’s disease. The data came from several hospitals in the U.S., as well as public databases.
The model uses a neural network, a set of algorithms that mimic the human brain. The device, which looks like a Wi-Fi router, is designed to emit radio signals and analyze reflections around it to obtain a person’s breathing signals.
There are two ways for breathing signals to be extracted while sleeping. One is through a belt placed around a person’s chest. The other is through a wall-installed wireless device. The signals are then sent to the neural network, which will evaluate the presence of Parkinson’s without requiring interaction with the patient and caregiver.
Once a person is diagnosed with Parkinson’s, the device is able to evaluate the severity of the disease — in compliance with Movement Disorder Society Unified Parkinson’s Disease Rating Scale — and track its progression over time.
The study’s results “show the potential of a new digital biomarker” for Parkinson’s that “is noninvasive and easy to measure in the person’s own home. Further, by using wireless signals to monitor breathing, the measurements can be collected every night in a touchless manner,” the researchers wrote.
According to Katabi, this study may be an important trigger for development of new therapies, as “the results can enable clinical trials with a significantly shorter duration and fewer participants, ultimately accelerating the development of new therapies,” she said.
“We’ve had no therapeutic breakthroughs this century, suggesting that our current approaches to evaluating new treatments is [sic] suboptimal,” said Ray Dorsey, a professor of neurology at the University of Rochester and the study’s co-author.
“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,” he added. “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.”
This new approach can also help around 40% of Parkinson’s patients who are geographically dispersed and far from Parkinson’s specialists in urban areas.
“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,” Katabi added.
This research was sponsored by the National Institutes of Health and partly supported by the National Science Foundation and the Michael J. Fox Foundation for Parkinson’s Research.