In this post, Anne Sereno, a professor of psychological sciences in the College of Health and Human Sciences and a member of the Purdue Institute for Integrative Neurosciences, discusses her research “A machine-learning method isolating changes in wrist kinematics that identify age-related changes in arm movement” which was recently published in Scientific Reports with the support of the National Institutes of Health and the Indiana Department of Health.
What did you want to know?
Using wrist-worn motion sensors and machine learning algorithms, we wanted to know whether we could develop a system to detect and identify physiological tremors—subtle, naturally-occurring tremors that increase with age.
What did you achieve?
Using our novel system in a study comparing younger (mean age 19) and older (mean age 56) participants performing simple tasks, we were able to detect and identify key tremor features (frequency and kinematic variables) of age-related physiological tremors.
What is the impact of this research?
Our system was sufficiently sensitive to detect increased physiological tremors and slowed movement initiation times in healthy older as compared to younger adults. Such objective methods have great potential in the clinic: to detect and distinguish between different neurological tremor disorders, to track disease progression, or to evaluate the effects of interventions. Furthermore, such a system could facilitate remote monitoring, reducing the need for patients to travel to the hospital or clinic as frequently, as well as aid longitudinal monitoring, giving physicians a better more accurate picture of the variation of tremors over longer time periods.