Citation

Moufawad El Achkar C, Lenbole-Hoskovec C, Paraschiv-Ionescu A, Major K, Aminian K, Büla C. Med. Biol. Eng. Comput. 2018; 56(8): 1403-1412.

Affiliation

Service of Geriatric Medicine, Centre Hospitalier Universitaire Vaudois (CHUV), 1066, Epalinges, Switzerland.

Copyright

(Copyright © 2018, International Federation for Medical and Biological Engineering, Publisher Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/s11517-017-1778-8

PMID

29327335

Abstract

The frequency and quality of sit-to-stand and stand-to-sit postural transitions decrease with age and are highly relevant for fall risk assessment. Accurate classification and characterization of these transitions in daily life of older adults are therefore needed. In this study, we propose to use instrumented shoes for postural transition classification as well as transition duration estimation from insole force signals. In the first part, data were collected with 10 older adults and 10 young participants performing transitions in the laboratory while wearing the instrumented shoes, without arm assistance. A wavelet approach was used to transform the insole force data, and candidate events were selected for transition duration estimation. Transition durations were then validated against a model based on force plate reference. Vertical force estimation was also compared to force plate measurement. In the second part, postural transitions were classified in daily life using the instrumented shoes and validated against a highly accurate wearable system. Transition duration was estimated with an error ranging from 10 to 20% while the error for vertical force estimation was 7%. Postural transition classification was achieved with excellent sensitivity and precision exceeding 90%. In conclusion, the instrumented shoes are suitable for classifying and characterizing postural transitions in daily life conditions of healthy older adults. Graphical abstract "Experimental setup showing instrumented shoes, reference force plate, as well as IMUs used for postural transition classification and duration estimation comparison".

Language: en

Keywords

Instrumented shoes; Postural transitions; Transition classification; Transition duration; Wavelet