Title: Evaluation of a modelling workflow to obtain subject specific spine geometry for sitting postures

Authors: Ilias Theodorakos; Xuguang Wang

Addresses: Université de Lyon, Lyon, France; Université Claude Bernard Lyon 1, Villeurbanne, France; Université Gustave Eiffel, IFSTTAR, LBMC UMR_T 9406, 24 avenue François Mitterrand, Bron, France ' Université de Lyon, Lyon, France; Université Claude Bernard Lyon 1, Villeurbanne, France; Université Gustave Eiffel, IFSTTAR, LBMC UMR_T 9406, 24 avenue François Mitterrand, Bron, France

Abstract: A three steps modelling workflow is introduced to obtain personalised spine models at standing and sitting postures. Initially, a principal component analysis based model is employed to locate internal points in a standing posture, using the external trunk shape as input. Then, these estimated internal points are used to personalise a generic anatomical spine template at a standing posture, using the AnyBody Modeling System. The personalised spine model is repositioned into a sitting posture using anatomical landmarks (AL). The proposed workflow was verified with MRI scans obtained from nine participants in standing and sitting postures. The mean prediction error of the spine joint centres was 18.7 (3.6) mm for the standing and 13.1 (5.0) mm for the sitting posture when all AL were used. The suggested workflow can provide accurate spine models at sitting postures when enough information (AL) is available regarding the spine, for the model positioning.

Keywords: personalisation; spine; subject-specific model; sitting biomechanics; sitting posture; AnyBody Modeling System; AMS.

DOI: 10.1504/IJHFMS.2019.105427

International Journal of Human Factors Modelling and Simulation, 2019 Vol.7 No.2, pp.87 - 99

Received: 05 Mar 2019
Accepted: 01 Aug 2019
Published online: 28 Feb 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article