Parkinson’s disease is the fastest growing neurodegenerative disease, now affecting more than 10 million people worldwide, yet clinicians still face significant challenges in tracking its severity and progression.
Doctors usually assess patients by testing their motor skills and cognitive function during clinic visits. These semi-objective measurements are often skewed by external factors – the patient may be tired after a long drive to the hospital. More than 40 percent of people with Parkinson’s disease never see a neurologist or a Parkinson’s specialist, often because they live too far from an urban center or have difficulty traveling.
In an effort to address these problems, researchers from MIT and elsewhere have demonstrated a home device that can monitor a patient’s movement and gait speed, which can be used to assess Parkinson’s disease severity, disease progression, and a patient’s response to medications. .
The device, about the size of a Wi-Fi router, collects data passively using radio signals that are reflected off the patient’s body as they move through their home. The patient does not need to wear a device or change his or her behavior. (A recent study, for example, showed that this type of device could be used to detect Parkinson’s disease from a person’s breathing patterns during sleep.)
The researchers used these devices to conduct two studies involving a total of 50 participants. They show that by using machine learning algorithms to analyze the data sets they collected (more than 200,000 walking pace measurements), a clinician can track the progress of Parkinson’s disease more effectively than periodic in-clinic assessments.
“By being able to have a device at home that can monitor the patient and tell the doctor remotely the progression of the disease, and the patient’s response to the medications so they can care for the patient even if the patient can’t attend the clinic — they now have real, reliable information — that goes a long way toward improving equity.” and access,” says senior author Dina Katabi, Thuan and Nicole Pham Professor in the Department of Electrical Engineering and Computer Science (EECS), and principal investigator in the Computer Science and Artificial Intelligence Laboratory (CSAIL) and MIT Jameel Clinic.
Co-lead authors are EECS graduate students Yingcheng Liu and Guo Zhang. The search was published in Translational Medicine Sciences.
This work uses a wireless device previously developed in Katabi’s lab that analyzes radio signals that bounce off people’s bodies. It transmits signals that use a small portion of your Wi-Fi router’s power – these ultra-low-power signals don’t interfere with other wireless devices in the home. As radio signals pass through walls and other solid objects, they are reflected back to humans due to the water in our bodies.
This creates a “human radar” that can track a person’s movement in a room. Radio waves always travel at the same speed, so the length of time it takes for signals to reflect back on the device indicates how a person is moving.
The device includes a machine learning classifier that can pick up minute radio signals that are reflected back to the patient even when other people are moving around the room. Sophisticated algorithms use this motion data to calculate walking speed – how fast you walk.
Since the device is running in the background and working all day, every day, it can collect a huge amount of data. The researchers wanted to see if they could apply machine learning to these data sets to gain insights into the disease over time.
They collected 50 participants, 34 of whom had Parkinson’s disease, and performed two observational studies of gait measurements at home. One study lasted two months and the other was conducted over two years. Through the studies, researchers collected more than 200,000 individual measurements that they aggregated, on average, to smooth out variance due to device condition or other factors. (For example, the device may accidentally turn off while cleaning, or the patient may walk more slowly when talking on the phone.)
They used statistical methods to analyze the data and found that walking speed at home can be used to effectively track the progress and severity of Parkinson’s disease. For example, they showed that walking speed decreased by almost twice as much for individuals with Parkinson’s disease, compared to those without it.
“Continuously observing the patient as they moved around the room enabled us to get really good measurements of their gait speed. With so much data, we were able to do the aggregation that allowed us to see very small differences,” Zhang says.
Better and faster results
Research into these variables provided some key insights. For example, researchers can see that daily fluctuations in a patient’s gait speed correspond to how they respond to their medication — walking speed may improve after a dose and then begin to decrease after a period of time.
“This really gives us the possibility to objectively measure how your movement responds to your medication. Previously, this was almost impossible because this drug effect could only be measured by having the patient keep a diary,” says Liu.
Your doctor can use this data to adjust your medication dose more effectively and accurately. This is especially important because many of the medications used to treat symptoms of the disease can cause serious side effects if the patient receives too much.
The researchers were able to show statistically significant results regarding the development of Parkinson’s disease after studying 50 people for just one year. By contrast, an often-cited study by the Michael J. Fox Foundation included more than 500 individuals and observed them for more than five years, Katabi says.
“For a pharmaceutical company or biotech company trying to develop drugs for this disease, this could significantly reduce the burden and cost and speed up the development of new treatments,” she adds.
My book credits the study’s success to the dedicated team of scientists and clinicians who worked together to address the many difficulties that arose along the way. For example, they started studying before the Covid-19 pandemic, so engineers initially entered people’s homes to set up the devices. When this was no longer possible, they developed a method for remote deployment of devices and created an easy-to-use application for participants and clinicians.
During the study period, they learned to automate processes and reduce effort, especially for participants and the clinical team.
This knowledge will prove useful as it looks to deploy the devices in home studies of other neurological disorders, such as Alzheimer’s disease, amyotrophic lateral sclerosis, and Huntington’s. They also want to explore how these methods, along with other work from Katabi’s lab showing that Parkinson’s disease can be diagnosed through breathing monitoring, can be used to gather a comprehensive set of markers that can diagnose the disease early and then use it to track and treat it.
This work is supported in part by the National Institutes of Health and the Michael J. Fox Foundation.