Validity of inertial measurement units in assessing segment angles and mechanical energies of elderly persons during sit-to-stand motion

Abstract

The aim of this study was to compare segment angles and mechanical energies obtained via inertial measurement unit (IMU) sensors attached to body segments and an optoelectronic motion capture (OMC) system during sit-to-stand motion, relevant for the assessment of movement disorder in elderly persons. seven elderly subjects were included in this study. Each subject performed sit-to-stand motions under the following two conditions: (1) rising with natural speed and posture (NORMAL) and (2) rising with trunk inclined forward (TILT). We calculated not only the segment angle but also the mechanical energy of each segment. Temporal profiles of the angles and mechanical energies of each segment were compared between the OMC data and the IMU data based on the Pearson's correlation coefficient (r) and root mean squared error (RMSE). The estimation error of the peak values of the data was also investigated with Bland-Altman bias and 95% limits of agreement. It was found that the value of r was >0.90 and the RMSE was <5° for all segments except thorax under the NORMAL condition. The Bland-Altman analysis produced an estimation error in 95% of measurements within two standard deviations. This study demonstrates that IMU based motion analysis for elderly could provide accurate kinematic and kinetic assessment of body segments compared to OMC system. However, the validity of the measurement of the thoracic angle by the IMU sensor was lower than other segments while temporal pattern of the thoracic angle during the motion was similarly detected by IMU and OMC system. Therefore, caution should be needed when interpreting research data of angular measurements of elderly and diseased persons obtained using IMU sensors especially thorax.