| Autor: |
Hegde N, Bries M, Swibas T, Melanson E, Sazonov E, Hegde N, Bries M, Swibas T, Melanson E, Sazonov E |
| Jazyk: |
angličtina |
| Zdroj: |
IEEE journal of biomedical and health informatics [IEEE J Biomed Health Inform] 2018 Jul; Vol. 22 (4), pp. 979-988. Date of Electronic Publication: 2017 Aug 01. |
| DOI: |
10.1109/JBHI.2017.2734803 |
| Abstrakt: |
Automatic recognition of activities of daily living (ADL) is an important component in understanding of energy balance, quality of life, and other areas of health and well-being. In our previous work, we had proposed an insole-based activity monitor-SmartStep, designed to be socially acceptable and comfortable. The goals of the current study were: first, validation of SmartStep in recognition of a broad set of ADL; second, comparison of the SmartStep to a wrist sensor and testing these in combination; third, evaluation of SmartStep's accuracy in measuring wear noncompliance and a novel activity class (driving); fourth, performing the validation in free living against a well-studied criterion measure (ActivPAL, PAL Technologies); and fifth, quantitative evaluation of the perceived comfort of SmartStep. The activity classification models were developed from a laboratory study consisting of 13 different activities under controlled conditions. Leave-one-out cross validation showed 89% accuracy for the combined SmartStep and wrist sensor, 81% for the SmartStep alone, and 69% for the wrist sensor alone. When household activities were grouped together as one class, SmartStep performed equally well compared to the combination of SmartStep and wrist-worn sensor (90% versus 94%), whereas the accuracy of the wrist sensor increased marginally (73% from 69%). SmartStep achieved 92% accuracy in recognition of nonwear and 82% in recognition of driving. Participants then were studied for a day under free-living conditions. The overall agreement with ActivPAL was 82.5% (compared to 97% for the laboratory study). The SmartStep scored the best on the perceived comfort reported at the end of the study. These results suggest that insole-based activity sensors may present a compelling alternative or companion to commonly used wrist devices. |
| Databáze: |
MEDLINE |
| Externí odkaz: |
|
