Estimation of the center of rotation using wearable magneto-inertial sensors

J Biomech. 2016 Dec 8;49(16):3928-3933. doi: 10.1016/j.jbiomech.2016.11.046. Epub 2016 Nov 16.

Abstract

Determining the center of rotation (CoR) of joints is fundamental to the field of human movement analysis. CoR can be determined using a magneto-inertial measurement unit (MIMU) using a functional approach requiring a calibration exercise. We systematically investigated the influence of different experimental conditions that can affect precision and accuracy while estimating the CoR, such as (a) angular joint velocity, (b) distance between the MIMU and the CoR, (c) type of the joint motion implemented, (d) amplitude of the angular range of motion, (e) model of the MIMU used for data recording, (f) amplitude of additive noise on inertial signals, and (g) amplitude of the errors in the MIMU orientation. The evaluation process was articulated at three levels: assessment through experiments using a mechanical device, mathematical simulation, and an analytical propagation model of the noise. The results reveal that joint angular velocity significantly impacted CoR identification, and hence, slow joint movement should be avoided. An accurate estimation of the MIMU orientation is also fundamental for accurately subtracting the contribution owing to gravity to obtain the coordinate acceleration. The unit should be preferably attached close to the CoR, but both type and range of motion do not appear to be critical. When the proposed methodology is correctly implemented, error in the CoR estimates is expected to be <3mm (best estimates=2±0.5mm). The findings of the present study foster the need to further investigate this methodology for application in human subjects.

Keywords: Accelerometers; Centre of rotation; Gyroscope; Human movement; Magneto-inertial sensing; Wearable devices.

MeSH terms

  • Algorithms
  • Biomechanical Phenomena
  • Calibration
  • Humans
  • Joints / physiology
  • Magnetometry
  • Models, Biological
  • Movement*
  • Orientation
  • Physical Examination / methods
  • Posture
  • Range of Motion, Articular
  • Signal-To-Noise Ratio