A model that has been developed by research led by Carnegie Mellon University can properly forecast how the mental health of patients with chronic neurological conditions like multiple sclerosis will be affected by stay-at-home orders like those implemented during the COVID-19 epidemic.
Before and during the first wave of the epidemic, researchers from CMU, the University of Pittsburgh, and the University of Washington collected information from MS patients’ smartphones and fitness trackers.
Specifically, during the record stay-at-home time, they developed machine learning models using the passively acquired sensor data to forecast despair, fatigue, poor sleep quality, and worsening MS symptoms.
The initial research topic concerned whether digital information from MS patients’ smartphones and activity trackers could forecast clinical consequences before the pandemic started.
When research participants were told they had to stay at home starting in March 2020, their daily behavior patterns underwent a substantial change. The research team understood that the information being gathered could help determine how the stay-at-home orders would affect MS patients.
“It presented us with an exciting opportunity,” said Mayank Goel, head of the Smart Sensing for Humans (SMASH) Lab at CMU. “If we look at the data points before and during the stay-at-home period, can we identify factors that signal changes in the health of people with MS?”
We were able to capture the change in people’s behaviors and accurately predict clinical outcomes when they are forced to stay at home for prolonged periods. Now that we have a working model, we could evaluate who is at risk for worsening mental health or physical health, inform clinical triage decisions, or shape future public health policies.
Mayank Goel
Over the course of three to six months, the researchers gathered data passively, including details such as the quantity and length of calls made and received on the participants’ smartphones, the number of missed calls, as well as information on their whereabouts and screen usage. From their fitness trackers, the team also gathered statistics on heart rate, sleep, and step count.
The research, “Predicting Multiple Sclerosis Outcomes During the COVID-19 Stay-at-Home Period: Observational Study Using Passively Sensed Behaviors and Digital Phenotyping,” was recently published in the Journal of Medical Internet Research Mental Health. Goel, an associate professor in the School of Computer Science’s Software and Societal Systems Department (S3D) and Human-Computer Interaction Institute (HCII), collaborated with Prerna Chikersal, a Ph.D. student in the HCII; Dr. Zongqi Xia, an associate professor of Neurology and director of the Translational and Computational Neuroimmunology Research Program at the University of Pittsburgh; and Anind Dey, a professor and dean of the University of Washington’s Information School.
The work was based on previous studies from Goel’s and Dey’s research groups. Using data from smartphones and activity trackers, a CMU team published research in 2020 that described a machine-learning model that may detect depression in college students at the end of the semester.
When compared to the greater population outside the university, the participants in the earlier study, notably the 138 first-year CMU students, were fairly similar to one another. The researchers worked with Xia’s MS research program because they wanted to see if their modeling approach could correctly predict clinically important health outcomes in a real-world patient group with more demographic and clinical diversity.
The researchers had the opportunity to investigate if its model might predict negative health outcomes like extreme fatigue, poor sleep quality, and increasing MS symptoms in addition to depression because people with MS can experience a number of chronic comorbidities.
Using the findings of this study as a foundation, the team aims to develop precision medicine for MS patients by enhancing early disease progression detection and applying digital phenotyping-based tailored therapies.
The research may also assist decision-makers entrusted with giving future orders to stay at home or other similar measures in the event of pandemics or natural disasters. The economic effects of the original COVID-19 stay-at-home orders were initially a source of worry, but it wasn’t until much later that the toll on people’s mental and physical health particularly among vulnerable populations like those with chronic neurological conditions became apparent.
“We were able to capture the change in people’s behaviors and accurately predict clinical outcomes when they are forced to stay at home for prolonged periods,” Goel said. “Now that we have a working model, we could evaluate who is at risk for worsening mental health or physical health, inform clinical triage decisions, or shape future public health policies.”
