Image
Real-World Activity Sensing for Patients with Alzheimer’s Disease
People with Alzheimer’s disease commonly report that their main priority is to maintain their ability to continue carrying out the activities they do today. The experts in digital health, geriatrics, and engineering leading this pilot study see great potential in using unobtrusive wearable sensors to measure a range of behaviors—including walking, balance, and getting up from a chair—to assess how older adults are functioning in the real world.
Launched in January 2023 as part of the first round of MassAITC-funded pilots, the study aims to develop and validate algorithms to capture measures of real-world walking behavior in patients with AD, with the potential to benefit clinical trials in the future. Currently, reliable methods to capture aspects of walking behavior do not exist for older adults with Alzheimer’s disease.
“Instead of relying on self-reports or requiring participants to come to a clinic to do a battery of tests that assess their physical function and capacity, wearable sensors could allow researchers to measure aspects of real-world behaviors and understand how participants are functioning in their home environments while imposing a much lower burden. That increases the potential to recruit participants who live farther away from metropolitan areas where clinical trials typically occur, and those from other traditionally underrepresented groups,” said Jen Blankenship, senior research scientist at VivoSense, a small California-based company, which develops and validates real-world digital clinical measures for use in regulated clinical trials. This is the company’s first venture into developing such measures for AD specifically, according to Blankenship, who earned her bachelor’s, master’s, and PhD degrees in kinesiology from UMass Amherst.
In the pilot study, participants with and without mild AD perform walking tasks while wearing activity monitors in the Human Motion Lab at UMass’s Center for Human Health and Performance (CH2P) Core Facility. Participants are also asked to wear activity monitors in their homes for two weeks. The study team will use data collected in the lab to develop machine learning algorithms to derive measures of real-world walking behavior in AD, then, those algorithms will be applied to the at-home monitoring data obtained in the study to determine if there are differences in real-world walking behaviors in patients with and without AD.
“There’s a lot of interest in using wearable sensors in regulated clinical trials, but the barrier to adoption is acceptance by the FDA,” explained collaborator Michael Busa, CH2P director. “Through this research, we aim to provide the evidence needed for the FDA to accept the validity of these measures.”