Feasibility Study Comparing Physical Activity Classifications from Accelerometers with Wearable Camera Data

Device-based assessments are frequently used to measure physical activity (PA) but contextual measures are often lacking. There is a need for new methods, and one under-explored option is the use of wearable cameras. This study tested the use of wearable cameras in PA measurement by comparing intensity classifications from accelerometers with wearable camera data. Seventy-eight 18–30-year-olds wore an Actigraph GT9X link accelerometer and Autographer wearable camera for three consecutive days. An image coding schedule was designed to assess activity categories and activity sub-categories defined by the 2011 Compendium of Physical Activities (Compendium). Accelerometer hourly detailed files processed using the Montoye (2020) cut-points were linked to camera data using date and time stamps. Agreement was examined using equivalence testing, intraclass correlation coefficient (ICC) and Spearman’s correlation coefficient (rho). Fifty-three participants contributing 636 person-hours were included. Reliability was moderate to good for sedentary behavior (rho = 0.77), light intensity activities (rho = 0.59) and moderate-to-vigorous physical activity (MVPA) (rho = 0.51). The estimates of sedentary behavior, light activity and MVPA from the two methods were similar, but not equivalent. Wearable cameras are a potential complementary tool for PA measurement, but practical challenges and limitations exist. While wearable cameras may not be feasible for use in large scale studies, they may be feasible in small scale studies where context is important.

[1]  Kimberly A. Clevenger,et al.  Development of cut-points for determining activity intensity from a wrist-worn ActiGraph accelerometer in free-living adults , 2020, Journal of sports sciences.

[2]  C. Lindgren,et al.  GWAS identifies 14 loci for device-measured physical activity and sleep duration , 2018, Nature Communications.

[3]  David R Bassett,et al.  2011 Compendium of Physical Activities: a second update of codes and MET values. , 2011, Medicine and science in sports and exercise.

[4]  Jiang Zhou,et al.  Kids'Cam: An Objective Methodology to Study the World in Which Children Live. , 2017, American journal of preventive medicine.

[5]  L. Mâsse,et al.  Physical activity in the United States measured by accelerometer. , 2008, Medicine and science in sports and exercise.

[6]  P. Kelly,et al.  Evaluating the Feasibility of Measuring Travel to School Using a Wearable Camera , 2012, American journal of preventive medicine.

[7]  L. Sylvia,et al.  Practical guide to measuring physical activity. , 2014, Journal of the Academy of Nutrition and Dietetics.

[8]  I-Min Lee,et al.  Impact of accelerometer data processing decisions on the sample size, wear time and physical activity level of a large cohort study , 2014, BMC Public Health.

[9]  Adrian Bauman,et al.  Accelerometer-based measures in physical activity surveillance: current practices and issues , 2014, British Journal of Sports Medicine.

[11]  Hannah Badland,et al.  Using wearable cameras to categorise type and context of accelerometer-identified episodes of physical activity , 2013, International Journal of Behavioral Nutrition and Physical Activity.

[12]  H. P. van der Ploeg,et al.  Advances in population surveillance for physical activity and sedentary behavior: reliability and validity of time use surveys. , 2010, American journal of epidemiology.

[13]  A. Bauman,et al.  Examining the Frequency and Contribution of Foods Eaten Away From Home in the Diets of 18- to 30-Year-Old Australians Using Smartphone Dietary Assessment (MYMeals): Protocol for a Cross-Sectional Study , 2018, JMIR research protocols.

[14]  M. Allman-Farinelli Nutrition Promotion to Prevent Obesity in Young Adults , 2015, Healthcare.

[15]  Steve E Hodges,et al.  Wearable cameras in health: the state of the art and future possibilities. , 2013, American journal of preventive medicine.

[16]  A. Doherty,et al.  A validation study of the Eurostat harmonised European time use study (HETUS) diary using wearable technology , 2019, BMC Public Health.

[17]  S. Keinänen-Kiukaanniemi,et al.  Measuring Physical Activity in Free-Living Conditions—Comparison of Three Accelerometry-Based Methods , 2017, Front. Physiol..

[18]  Gareth Hagger-Johnson,et al.  Sitting Time, Fidgeting, and All-Cause Mortality in the UK Women's Cohort Study. , 2016, American journal of preventive medicine.

[19]  Steve Hodges,et al.  Can we use digital life-log images to investigate active and sedentary travel behaviour? Results from a pilot study , 2011, The international journal of behavioral nutrition and physical activity.

[20]  Ulf Ekelund,et al.  Age group comparability of raw accelerometer output from wrist- and hip-worn monitors. , 2014, Medicine and science in sports and exercise.

[21]  U. Ekelund,et al.  Evaluation of raw acceleration sedentary thresholds in children and adults , 2017, Scandinavian journal of medicine & science in sports.

[22]  P S Freedson,et al.  Calibration of the Computer Science and Applications, Inc. accelerometer. , 1998, Medicine and science in sports and exercise.

[23]  M. Tremblay,et al.  A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review , 2008, The international journal of behavioral nutrition and physical activity.

[24]  J. Larry Durstine,et al.  Chronic disease and the link to physical activity , 2013 .

[25]  Gert R. G. Lanckriet,et al.  Hip and Wrist Accelerometer Algorithms for Free-Living Behavior Classification. , 2016, Medicine and science in sports and exercise.

[26]  Charles E Matthew,et al.  Calibration of accelerometer output for adults. , 2005, Medicine and science in sports and exercise.

[27]  D. J. van der Valk,et al.  How accurately can sitting and the intensity of walking and cycling be classified using an accelerometer on the waist for the purpose of the “Global recommendations on physical activity for health”? , 2015 .

[28]  K. Khunti,et al.  Sedentary time in adults and the association with diabetes, cardiovascular disease and death: systematic review and meta-analysis , 2012, Diabetologia.

[29]  A. Rangan,et al.  Validity of self-reported weight and height for BMI classification: A cross-sectional study among young adults. , 2019, Nutrition.

[30]  Jiang Zhou,et al.  Semantic Indexing of Wearable Camera Images: Kids'Cam Concepts , 2016, iV&L-MM@MM.

[31]  Huaijun Wang,et al.  Wearable Sensor-Based Human Activity Recognition Using Hybrid Deep Learning Techniques , 2020, Secur. Commun. Networks.

[32]  P. Dixon,et al.  A Primer on the Use of Equivalence Testing for Evaluating Measurement Agreement , 2017, Medicine and science in sports and exercise.

[33]  S. Guthridge,et al.  Socio-economic indexes for areas (SEIFA) of administrative health districts and urban centres/localities in the Northern Territory , 2005 .

[34]  Aiden R. Doherty,et al.  High group level validity but high random error of a self-report travel diary, as assessed by wearable cameras , 2014 .

[35]  William Leung,et al.  Studying third-parties and environments: New Zealand sun-safety research. , 2019, Health promotion international.

[36]  Oisin Mac Aodha,et al.  Unsupervised Monocular Depth Estimation with Left-Right Consistency , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[37]  S. Godbole,et al.  Convergent validity of ActiGraph and Actical accelerometers for estimating physical activity in adults , 2018, PloS one.

[38]  Klaas R. Westerterp,et al.  Assessment of physical activity: a critical appraisal , 2009, European Journal of Applied Physiology.

[39]  T. Baranowski,et al.  Utility of eButton images for identifying food preparation behaviors and meal-related tasks in adolescents , 2018, Nutrition Journal.

[40]  S. Marshall,et al.  Using the SenseCam to improve classifications of sedentary behavior in free-living settings. , 2013, American journal of preventive medicine.

[41]  Paul Kelly,et al.  Developing a Method to Test the Validity of 24 Hour Time Use Diaries Using Wearable Cameras: A Feasibility Pilot , 2015, PloS one.