No abstract

Occupational musculoskeletal disorders, particularly chronic low back pain (LBP), are ubiquitous due to prolonged static sitting or nonergonomic sitting positions. Therefore, the aim of this study was to develop an instrumented chair with force and acceleration sensors to determine the accuracy of automatically identifying the user's sitting position by applying five different machine learning methods (Support Vector Machines, Multinomial Regression, Boosting, Neural Networks, and Random Forest). Forty-one subjects were requested to sit four times in seven different prescribed sitting positions (total 1148 samples). Sixteen force sensor values and the backrest angle were used as the explanatory variables (features) for the classification. The different classification methods were compared by means of a Leave-One-Out cross-validation approach. The best performance was achieved using the Random Forest classification algorithm, producing a mean classification accuracy of 90.9% for subjects with which the algorithm was not familiar. The classification accuracy varied between 81% and 98% for the seven different sitting positions. The present study showed the possibility of accurately classifying different sitting positions by means of the introduced instrumented office chair combined with machine learning analyses. The use of such novel approaches for the accurate assessment of chair usage could offer insights into the relationships between sitting position, sitting behaviour, and the occurrence of musculoskeletal disorders.

Soft tissue artefact affects the determination of skeletal kinematics. Thus, it is important to know the accuracy and limitations of kinematic parameters determined and modelled based on skin marker data. Here, the curvature angles, as well as the rotations of the lumbar and thoracic segments, of seven healthy subjects were determined in the sagittal plane using a skin marker set and compared to measurements taken in an open upright MRI scanner in order to understand the influence of soft tissue artefact at the back. The mean STA in the flexed compared to the extended positions were 10.2±6.1 mm (lumbar)/9.3±4.2 mm (thoracic) and 10.7±4.8 mm (lumbar)/9.2±4.9 mm (thoracic) respectively. A linear regression of the lumbar and thoracic curvatures between the marker-based measurements and MRI-based measurements resulted in coefficients of determination, R2, of 0.552 and 0.385 respectively. Skin marker measurements therefore allow for the assessment of changes in the lumbar and thoracic curvature angles, but the absolute values suffer from uncertainty. Nevertheless, this marker set appears to be suitable for quantifying lumbar and thoracic spinal changes between quasi-static whole body postural changes.