Skip to Main content Skip to Navigation
Conference papers

Reproducibility and reliability of embedded inertial motion sensors integrated to smart eyeglasses during Sit-to-Stand test

Abstract : Purpose: Kinematic and kinetic assessments of Sit-to-Stand (STS) usually done in laboratory settings involve strong technical expertise and expensive equipment. As an alternative, embedded inertial motion sensors can be used to quantify key parameters of selected everyday life movements such as STS. Sensors can be integrated in different wearable devices like smartwatch, wristband and smart eyeglasses. It has been shown that placing inertial sensors on the head allows obtaining better spatial and temporal parameters than those obtained by sensors placed on the hip or the trunk [1]. In this study, we assessed the reliability and reproducibility of inertial motion sensors integrated to smart eyeglasses during Sit-to-Stand movements. Methods: Ten healthy adult participants (5 men and 5 women; 27 ± 5.2 years old) completed two identical test sessions (separated by one week) comprising STS measures. Throughout the tests, participants wore smart eyeglasses (Ellcie-Healthy, France) with a 3-axis accelerometer inserted in the right branch. Each testing session was composed of Sit-to-Stand movements performed at two different self-selected speeds (e.g., slow and comfort) with and without cervical collar. Fifteen repetitions were realized per condition. The maximum and minimum peak values of the vertical acceleration were obtained with the accelerometer of the eyeglasses. Separate repeated measures ANOVAs were performed to assess the changes across the 15 trials for each variable within, and across, testing sessions. Significance level was set as p<0.05 and effect size was computed from partial eta² values (η²p). Relative reliability of each variable across the trials was assessed using intraclass correlation coefficient (ICC). Variability of each variable was assessed with the coefficient of variation (CV) calculation. Results: The maximum and minimum acceleration values of STS did not differ across trial nor sessions (p>0.05). ICC values were comprised between 0.63 and 0.93 for the slow speed whether or not the cervical collar was worn. The ICC values ranged between 0.70 and 0.93 for the comfort speed. Coefficients of variation varied from 4.8% to 12.9% for all the conditions during both sessions. Discussion and Conclusion: This study clearly indicated that intra and inter-session reproducibility and reliability of maximum and minimum peak values are satisfactory. Based on the ICC values, our data show that the measured variables were reliable regardless of the testing conditions (i.e., speed and cervical collar). The accelerometer data of the smart eyeglasses used in the present study were reproducible across days and conditions. These smart eyeglasses could be used to recognize everyday life movements. Further studies should investigate the reproducibility and reliability of this technology over a higher number of participants as well as for other movements and populations (e.g., health compromised individuals and/or elderly). [1] Lindemann, U., Hock, A., Stuber, M., Keck, W., & Becker, C. (2005). Evaluation of a fall detector based on accelerometers: A pilot study. Medical and Biological engineering and computing, 43(5), 548-551.
Document type :
Conference papers
Complete list of metadata
Contributor : Justine Hellec <>
Submitted on : Tuesday, April 14, 2020 - 12:41:31 PM
Last modification on : Monday, December 28, 2020 - 5:24:26 PM


  • HAL Id : hal-02541999, version 1



Justine Hellec, Frédéric Chorin, Andrea Castagnetti, Serge Colson. Reproducibility and reliability of embedded inertial motion sensors integrated to smart eyeglasses during Sit-to-Stand test. International Conference on Human Interaction and Emerging Technologies, Aug 2019, Nice, France. ⟨hal-02541999⟩



Record views