A Machine Learning Classifier for Detection of Physical Activity Types and Postures During Free-Living

Abstract

Introduction: Accelerometer-based measurements of physical activity types are commonly used to replace self-reports. To advance the field, it is desirable that such measurements allow accurate detection of key daily physical activity types. This study aimed to evaluate the performance of a machine learning classifier for detecting sitting, standing, lying, walking, running, and cycling based on a dual versus single accelerometer setups during free-living. Methods: Twenty-two adults (mean age [SD, range] 38.7 [14.4, 25–68] years) were wearing two Axivity AX3 accelerometers positioned on the low back and thigh along with a GoPro camera positioned on the chest to record lower body movements during free-living. The labeled videos were used as ground truth for training an eXtreme Gradient Boosting classifier using window lengths of 1, 3, and 5 s. Performance of the classifier was evaluated using leave-one-out cross-validation. Results: Total recording time was ∼38 hr. Based on 5-s windowing, the overall accuracy was 96% for the dual accelerometer setup and 93% and 84% for the single thigh and back accelerometer setups, respectively. The decreased accuracy for the single accelerometer setup was due to a poor precision in detecting lying based on the thigh accelerometer recording (77%) and standing based on the back accelerometer recording (64%). Conclusion: Key daily physical activity types can be accurately detected during free-living based on dual accelerometer recording, using an eXtreme Gradient Boosting classifier. The overall accuracy decreases marginally when predictions are based on single thigh accelerometer recording, but detection of lying is poor.

Modified prolonged exposure therapy as Early Intervention after Rape (The EIR-study): study protocol for a multicenter randomized add-on superiority trial

“…Physical activity and sleep will be objectively measured using two AX3 (Axivity, Ltd., UK) accelerometers attached to the skin at the right thigh and lower back. The sensor streams are analyzed using a machine learning model, providing an overall accuracy of about 95% in detecting total sleep time (TST), movements per time unit (indicator of fragmented sleep), and time spent sitting, standing, walking, running, cycling, and lying down [ 67 , 68 ].…”

Assessment of 24-hour physical behaviour in adults via wearables: a systematic review of validation studies under laboratory conditions

“…In general, we expect the fast-technological development of wearables to affect the future of physical behaviour data evaluation and processing. In particular, supervised learning approaches, such as machine learning or deep learning algorithms, are gaining popularity [ 35 – 37 ]. The inclusion of supervised learning approaches in health behaviour research has been slow, but this may change in the upcoming years [ 38 ].…”

Sensitivity and specificity of measuring children's free-living cycling with a thigh-worn Fibion® accelerometer

“…In the present study we observed sensitivity of 72.2% and specificity of 81.9% (i.e., for time between 0 and 34 s), which is a similar magnitude to the study by Brønd et al, but included free-living, also longer duration, trips. Most recently, Bach et al included 22 adults in their study and recorded their activities (e.g., sitting, standing, lying, walking, running, and cycling) during 1.5–2 h of free-living setting using direct video recording from chest, and dual accelerometers (Axivity AX3; worn on lower back and thigh) ( 47 ). Using machine learning methods, they observed that dual accelerometry can provide accurate estimation of free-living activities, but that a single thigh-worn accelerometer could also provide the same estimation.…”