AI assists in detecting cognitive impairment

To improve accessibility to cognitive assessments, a research team at the University of Missouri has developed a portable system designed to measure various aspects of motor function efficiently. This affordable and easy-to-use device integrates a depth camera, a force plate, and an interface board to assess movement patterns associated with cognitive decline. 

The interdisciplinary research team includes Trent Guess, Associate Professor at the College of Health Sciences, Jamie Hall, Associate Teaching Professor at the College of Health Sciences, and Praveen Rao, Associate Professor at the College of Engineering. In a recent study, older adults, including individuals with diagnosed MCI, were asked to complete three physical tasks: standing still, walking, and standing up from a bench. Simultaneously, they had to count backward in intervals of seven - a cognitive challenge designed to engage memory and concentration. 

Data collected from these tasks were analyzed using a machine learning model, a form of artificial intelligence, which successfully identified 83% of participants with MCI. 

“The areas of the brain involved in cognitive impairment overlap with areas of the brain involved in motor function, so when one is diminished, the other is impacted as well,” Guess explained. “These can be very subtle differences in motor function related to balance and walking that our new device is able to detect but would go unnoticed through observation.” 

With Alzheimer’s disease cases in the U.S. projected to more than double by 2060, according to the Centers for Disease Control and Prevention (CDC), early detection of MCI is becoming increasingly important. Hall emphasized the potential impact of the device: “Alzheimer’s disease is a significant problem here in the U.S. We know that if we can identify people early, we can provide early intervention to halt or slow the progression of the disease. Only about 8% of people in the U.S. who are believed to have MCI receive a clinical diagnosis.” 

The research team envisions this portable system being implemented in various settings, including county health departments, assisted living facilities, community centers, physical therapy clinics, and senior centers. Hall noted, “Our portable system can detect if a person walks slower or doesn’t take as big of a step because they are thinking very hard. Some people have more sway and are less balanced or are slower to stand up when they are sitting. Our technology can measure these subtle differences in a way that you could not with a stopwatch.” 

Future research will expand on these findings, with a focus on the system’s ability to detect fall risk and frailty among older adults. 

MEDICA-tradefair.com; Source: University of Missouri-Columbia