University of South Wales


Biomedical Engineering and Computing
Research Group

Algorithms for Anthropometric Feature Extraction from Cardiff Body Match Measurements


Body shape data for specific individuals is important as it dictates the shape of a custom contoured seat that can be made to meet their complex needs. Conventional radiological techniques require direct exposure of the spine to the measurement instrumentation, preferably in the erect standing position. Radiographs require medical approval, are costly (both monetarily and temporal), and some techniques run the risk of excessive exposure. The work focuses on the development of novel algorithms for anthropometric measurements and feature extraction from the body shape data acquired using the CBM shape sensor shown in Figure 1


CBM shape sensor
Figure 1. The CBM Rig (left); subject sat in the CBM Rig (top-right); the corresponding CBM measurement matrix for the back (bottom-right)

Algorithms are developed for the measurement of anthropometric features such as pelvic orientation, direction of the spinal curve, trunk asymmetry, orientation of the legs, back and base contact area and location of the maximum pressure points. At Cardiff and Vale UHB’s REU individual’s body shapes have been recorded via the CBM system, since 1996. Collectively, these files represent a database of hundreds of individual’s body shapes. The CBM measurements are linked with clinical information related to the individual; such as, medical diagnosis, musculoskeletal conditions, age, gender etc. The clinical information related to individuals that have been measured in the CBM shape sensor will be used to validate the performance of algorithms. More specific tests relating to the accuracy of extracted anthropometric features will be performed using the clinical data obtained during the planned observational clinical trial.

Symmetry Line
Figure 2. (top) estimated Symmetry Line in the coronal plane, (bottom) the corresponding backrest profile in the sagittal plane

Publications:

  1. Kulon, J., Partlow, A., Gibson, C., Wilson, I. and Wilcox, S. J., "Rule-Based Algorithm for the Classification of Sitting Postures in the Sagittal Plane from the Cardiff Body Match Measurement System," Journal Med. Eng. Tech., 2013. (doi: 10.3109/03091902.2013.844208).

  2. Partlow, A., Gibson, G., Kulon, J., Wilson, I., Wilcox, S. J. “Pelvis feature extraction and classification of Cardiff body match rig base measurements for input into a knowledge-based system,” Journal Med. Eng. Tech., vol. 36, no. 8, pp. 399-406, Nov. 2012. (doi: 10.3109/03091902.2012.712202).

  3. Partlow, A., “Approximating the 3D Position of the Pelvis in Cardiff Body Match Measurements,” All-Wales Medical Physics and Clinical Engineering Meeting 2013 held at Village Hotel, Swansea, UK, 2013 June 21.

  4. Partlow, A., “Feature Extraction from Cardiff Body Match Rig Base and Back Measurements to Assess a Client’s Postural Needs,” All-Wales Medical Physics and Clinical Engineering Meeting 2012 held at Village Hotel, Cardiff, UK, 2012 June 15 .