If your physician worries about your risk of falling, she soon may tell you to take a walk. A team from Neuroscience Research Australia (NeuRA) and the University of New South Wales in Sydney (NSW Sydney) are developing a smart watch algorithm called Watch Walk to evaluate an individual’s risk of falling by monitoring walking speed and stability. The researchers published their findings in Scientific Reports.
Building the Watch Walk algorithm entailed two stages. First, the team recorded data to create the digital gait biomarker algorithm. A study group of 101 19-to-81-year-old subjects wore a wrist wearable with an on-board accelerometer while running in a lab setting and while performing specific actions and routines in their own homes. The team prescribed the movements to measure structured mobility tasks. Stage one culminated with an algorithm ready for validation based on gait including posture, cadence, walking speed, and stride length. According to the researchers, these factors can help predict the subject’s risk of falling. The data can also provide insight into overall health and functional decline.
The team validated the Watch Walk gait biomarker algorithm with a larger test group of 77,822 people in the UK Biobank database. The UK Biobank is a vast and globally accessible medical research resource containing genetic and health data from more than a half million people from the UK. Subjects in the validation stage of the Watch Walk development project wore the wrist device on their dominant wrist for seven days. The subjects, who were 46 to 77 years old, recorded an aggregate of 11,646 four-second movement recordings. The algorithm validation test classified the recordings in four categories. The Watch Walk algorithm showed high levels of precision for walking (93%), running (98%), stationary (86%), and unspecified arm activities (74%).
Earlier gait biomarker analysis studies relied on data that was obtained solely while walking on treadmills or fixed-length walkways, the study authors said. According to Professor Stephen Lord, Senior Principal Research Fellow at NeuRA and UNSW Medicine & Health, “the growth of wearable device technology in recent years has provided an accessible and approachable method of fall prevention in older people. Watch Walk demonstrates that this technology can also be highly accurate in real-world settings.” The Australian team intends to release the Watch Walk app in late 2023.
