MIT’s new artificial intelligence technology can detect Parkinson’s early using breathing patterns

The new tech can be used at home while the patient sleeps.

MIT researchers developed a tech that will allow patients to detect Alzheimer's Disease early on and at home. Bloomberg

A new MIT-developed artificial intelligence model can make an early detection of Parkinson’s Disease — which is notoriously hard to diagnose — from a person’s breathing patterns, the university announced Monday.

A news release about the technology said that Parkinson’s disease is hard to diagnose because it relies primarily on the appearance of motor symptoms, such as tremors, stiffness, and slowness, which often appear several years after the disease onset.

But Dina Katabi, an MIT electrical engineering and computer science professor, and her team have now developed an artificial intelligence model that can detect Parkinson’s from a person’s breathing patterns, the release said.


The tech is a neural network — a series of connected algorithms that mimic the way a human brain works — capable of assessing whether someone has Parkinson’s from how they breathe while they sleep.

The neural network, which was trained by MIT PhD student Yuzhe Yang and postdoc Yuan Yuan, is also able to discern the severity of someone’s Parkinson’s and track the progression of their disease over time, the release said.

“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,” Katabi said in the 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.”


Over the years, researchers have tried to detect Parkinson’s using cerebrospinal fluid and neuroimaging, but such methods are invasive, costly, and require access to specialized medical centers, the release said. This makes these methods unsuitable for frequent testing, which could allow for early diagnosis and continuous tracking of the disease’s progression.


But the researchers knew that with the new AI model, Parkinson’s detection could be done every night at home while the patient is asleep and without touching their body.

So they developed a device that looks like a Wi-Fi router, but instead of providing internet access, the device emits radio signals, analyzes reflections off the surrounding environment, and monitors the person’s breathing patterns without any bodily contact, the release said. The breathing signal is then fed to the neural network to assess for Parkinson’s.

The research team’s algorithm was then tested on 7,687 individuals, including 757 Parkinson’s patients.

The fastest-growing neurological disease in the world, Parkinson’s is the second most common neurological disorder, after Alzheimer’s disease, the release said. In the U.S. alone, over a million people live with the disease.

“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 said in the release.

Yang is the first author and Katabi is the senior author of a new paper describing the tech that was published Monday in Nature Medicine. Other authors include researchers from Rutgers University, the University of Rochester Medical Center, the Mayo Clinic, Massachusetts General Hospital, and Boston University.


This discussion has ended. Please join elsewhere on