Identifying Types of Physical Activity With a Single Accelerometer: Evaluating Laboratory-trained Algorithms in Daily Life

• Published in: IEEE transactions on biomedical engineering Vol. 58; no. 9; pp. 2656 - 2663
• Main Authors: Gyllensten, Illapha Cuba, Bonomi, Alberto G.
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.09.2011; Institute of Electrical and Electronics Engineers; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Acceleration; Accelerometers; Accuracy; Activities of Daily Living - classification; Adult; Algorithms; Artificial neural networks; Assessment of daily physical activity; Biological and medical sciences; Classification; Classification algorithms; Decision Trees; Female; Fundamental and applied biological sciences. Psychology; Humans; intelligent device for energy expenditure and physical activity (IDEEA); Laboratories; Locomotion - physiology; Male; Models, Statistical; Monitoring, Ambulatory - instrumentation; Monitoring, Ambulatory - methods; Motor Activity - physiology; Neural Networks (Computer); physical activity; Principal Component Analysis; Reproducibility of Results; Support Vector Machine; Support vector machines; Training data; triaxial accelerometer; Vertebrates: body movement. Posture. Locomotion. Flight. Swimming. Physical exercise. Rest. Sports; Performance evaluation; Physical exercise; Human; classification algorithms; Daily living; Assessment of daily physical activity; Algorithm; intelligent device for energy expenditure and physical activity (IDEEA); physical activity; Energetic cost; Reproducibility; Classification; Accelerometer; triaxial accelerometer; Intelligent system; Biomedical engineering
• ISSN: 0018-9294; 1558-2531; 1558-2531
• DOI: 10.1109/TBME.2011.2160723

Accurate identification of physical activity types has been achieved in laboratory conditions using single-site accelerometers and classification algorithms. This methodology is then applied to free-living subjects to determine activity behavior. This study is aimed at analyzing the reproducibility of the accuracy of laboratory-trained classification algorithms in free-living subjects during daily life. A support vector machine (SVM), a feed-forward neural network (NN), and a decision tree (DT) were trained with data collected by a waist-mounted accelerometer during a laboratory trial. The reproducibility of the classification performance was tested on data collected in daily life using a multiple-site accelerometer augmented with an activity diary for 20 healthy subjects (age: 30 ± 9; BMI: 23.0 ± 2.6 kg/m 2 ). Leave-one-subject-out cross validation of the training data showed accuracies of 95.1 ± 4.3%, 91.4 ± 6.7%, and 92.2 ± 6.6% for the SVM, NN, and DT, respectively. All algorithms showed a significantly decreased accuracy in daily life as compared to the reference truth represented by the IDEEA and diary classifications (75.6 ± 10.4%, 74.8 ± 9.7%, and 72.2 ± 10.3 %; p <; 0.05). In conclusion, cross validation of training data overestimates the accuracy of the classification algorithms in daily life.