Personalization of Energy Expenditure Estimation in Free Living Using Topic Models

Abstract

We introduce an approach to personalize energy expenditure (EE) estimates in free living. First, we use topic models to discover activity composites from recognized activity primitives and stay regions in daily living data. Subsequently, we determine activity composites that are relevant to contextualize heart rate (HR). Activity composites were ranked and analyzed to optimize the correlation to HR normalization parameters. Finally, individual-specific HR normalization parameters were used to normalize HR. Normalized HR was then included in activity-specific regression models to estimate EE. Our HR normalization minimizes the effect of individual fitness differences from entering in EE regression models. By estimating HR normalization parameters in free living, our approach avoids dedicated individual calibration or laboratory tests. In a combined free-living and laboratory study dataset, including 34 healthy volunteers, we show that HR normalization in 14-day free-living data improves accuracy compared to no normalization and normalization based on activity primitives only ( 29.4% and 19.8 % error reduction against lab reference). Based on acceleration and HR, both recorded from a necklace, and GPS acquired from a smartphone, EE estimation error was reduced by 10.7 % in a leave-one-participant-out analysis.

Evaluating physiological signal salience for estimating metabolic energy cost from wearable sensors

“…Heart rate is another common physiological signal used to estimate energy cost, and a linear relationship between heart rate and energy expenditure during submaximal physical activity has been established (28,37,41). However, given that heart rate is tightly controlled by the nervous system, changes in heart rate can also be attributed to changes in emotional arousal and may require individual calibration (2,10,28,34). Some of these factors have been mitigated by combining heart rate monitors with accelerometers (10,11,19) or biological parameters (e.g., sex, age, weight) (28) in predictive algorithms.…”

“…Some of these factors have been mitigated by combining heart rate monitors with accelerometers (10,11,19) or biological parameters (e.g., sex, age, weight) (28) in predictive algorithms. Some more recent studies have incorporated autonomic nervous system signals, such as skin temperature, humidity, and electrodermal activity into their estimates of energy cost (2,3,57). Electrodermal activity (also called galvanic skin response) is a measure of skin conductance, which changes when an individual begins to sweat.…”

A Survey on Energy Expenditure Estimation Using Wearable Devices

“…Despite this, Fitbit, Inc., sued Jawbone for the infringement of U.S. Patent 7 . Although they eventually settled their dispute 8 , this may have led to the end of Jawbone 9 . Apple is also selling millions of smartwatches 10 and protecting his knowledge using patents [116].…”

How to Measure Sedentary Behavior at Work?

“…Heart rate (HR) can therefore be used to estimate energy expenditure. Coupling HR monitoring and accelerometers leads to a better accuracy in the assessment of SB and physical activity (30, 39). Historically, electrical HR sensors detect the electric impulses that are linked with the myocardial contraction.…”