Communications of the ACM
Wearable Devices Accurately Predict Fitness Levels
Non-exercise prediction models are an alternative to exercise testing in clinical settings.
Credit: Fast Company
Scientists at the University of Cambridge used wearable devices to accurately measure the fitness of individuals without requiring the users to exercise.
The researchers compiled activity data from more than 11,000 study participants using wearable sensors, then tested a subgroup of 2,675 participants seven years later. They used the data to create a model for predicting maximal oxygen consumption (VO2max) — an overall fitness metric and predictor of heart disease and mortality risk — for validation against a third cohort of 181 participants engaged in a laboratory-based exercise test.
The algorithm agreed closely with measured VO2max scores at baseline and in follow-up testing.
"This study is a perfect demonstration of how we can leverage expertise across epidemiology, public health, machine learning, and signal processing," said Cambridge's Ignacio Perez-Pouelo, co-author of a study on the work published in Digital Medicine.
From University of Cambridge
View Full Article
Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA
No entries found