Innovative Algorithm Enhances Fitness Tracking for People with Obesity

Fitness trackers play a vital role in helping individuals monitor daily activity levels and caloric expenditure. However, for people living with obesity, these devices often produce inaccurate readings due to differences in gait, movement patterns, and device positioning. Recognizing this challenge, researchers at Northwestern University have developed a novel algorithm that significantly improves the accuracy of activity and calorie measurement in this population.

Published in Scientific Reports, the study entitled "Developing and comparing a new BMI inclusive energy burn algorithm on wrist-worn wearables" introduces an open-source, transparent algorithm specifically designed for wrist-worn devices. Led by Nabil Alshurafa from Northwestern’s HABits Lab, this technology fills a critical gap by providing tailored data that better reflects the energy expenditure of individuals with obesity during various physical activities.

Current algorithms primarily target people without obesity, often miscalculating calorie burn due to gait changes and device tilt caused by higher body weight. While wrist-worn trackers are preferred for comfort and adherence, they have not been rigorously calibrated for those with obesity, leading to significant inaccuracies. This new algorithm, however, leverages research-grade data and wearable cameras to validate its precision, achieving over 95% accuracy in real-world scenarios.

The inspiration for this innovation came from a personal experience when Alshurafa observed his mother-in-law, who has obesity, working hard in an exercise class but not receiving appropriate feedback from fitness trackers. Motivated to improve health monitoring, the team used commercial fitness tracker data to create a model that accurately estimates minute-by-minute energy use, making activity tracking more inclusive.

The study employed two main methods: one involved participants wearing fitness trackers and metabolic carts during physical tasks to compare results, while the other used wearable cameras to visually confirm instances of over- or under-estimation of calorie expenditure. The team also challenged participants with activities like wall push-ups to test the algorithm’s robustness in diverse conditions.

This breakthrough enables more precise activity monitoring for people with obesity, which can lead to better-tailored interventions and improved health outcomes. Furthermore, the accessible nature of the open-source algorithm encourages other researchers to build upon this work, fostering advancements in health tracking technology.

Projected to be integrated into upcoming activity-monitoring apps for both iOS and Android later this year, this innovation promises to make fitness tracking more equitable and effective. By providing accurate data that reflects the hard work of individuals with obesity, it supports a more inclusive approach to health and wellness.