By analysing breathing patterns, a recently created artificial intelligence model can identify Parkinson's illness. The programme can also determine how severe Parkinson's disease is and monitor its development over time.
Parkinson’s disease is famously challenging to diagnose since it mostly depends on the emergence of motor symptoms like tremors, stiffness, and slowness—symptoms that frequently develop years after the disease first manifests itself.
Parkinson’s disease, the second most prevalent neurological disorder after Alzheimer’s disease, is the neurological condition with the fastest global growth rate. It affects more than 1 million people in the US alone, putting a $51.9 billion annual strain on the economy. The device developed by the study team was put to the test on 7,687 people, including 757 Parkinson’s patients.
The principal investigator at the MIT Jameel Clinic and the Thuan and Nicole Pham Professor in the Department of Electrical Engineering and Computer Science (EECS) at MIT, Dina Katabi, and her group have created an artificial intelligence model that can identify Parkinson’s disease simply by observing a person’s breathing patterns.
The tool in question is a neural network, which consists of interconnected algorithms that simulate how the human brain functions and can determine a person’s nocturnal breathing—or, more specifically, the breathing patterns that take place while sleeping—to determine if they have Parkinson’s disease.
The neural network, which Yuzhe Yang, an MIT Ph.D. student, and Yuan Yuan, a postdoc, trained, can also determine a person’s Parkinson’s disease severity and monitor the disease’s development over time. A new study outlining the technique was co-authored by Yang and Yuan and was just published in Nature Medicine. The senior author is Katabi, head of the Center for Wireless Networks and Mobile Computing and a member of the MIT Computer Science and Artificial Intelligence Laboratory.
Cerebrospinal fluid and neuroimaging have been investigated as potential screening tools for Parkinson’s disease over the years, but these techniques are invasive, expensive, and require access to specialised medical facilities, making them unsuitable for routine testing that would otherwise allow for early diagnosis or ongoing monitoring of disease progression.
The MIT researchers showed that a Parkinson’s assessment using artificial intelligence may be carried out each night at home while the patient is sleeping and without having to touch them. To achieve this, the researchers created a device that resembles a Wi-Fi router for a home, but instead of giving internet access, it emits radio signals, analyses how they are reflected off the environment, and then, without any physical touch, extracts the subject’s breathing patterns. The breathing signal is then fed to the neural network to assess Parkinson’s in a passive manner, and there is zero effort needed from the patient and caregiver.
“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 says. “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.”
Katabi notes that the study has important implications for Parkinson’s drug development and clinical care.