Automatic Quantification of Spine Parameters from X-ray Images by Means of Morphological Tools
This work presents an automatic technique for computing from an X-ray image the value of the set of parameters that characterizes the morphological structure of a human lumbar spine. To compute these parameters it is necessary to detect, in the X-ray image, each one of the vertebrae that forms the spine, as well as their relative location. Given the low quality of normal X-ray images, a direct detection cannot be performed. Thus, the proposed technique consists in five different steps: image simplification, detection of elements, decision, validation of objects and parameter computation. In order to be consistent, the validation step and the parameter computation rely on the same medical vertebra model.
Key wordsSpine model Segmentation Geodesic distance
Unable to display preview. Download preview PDF.
- 1.A. White III, M. M. Panjabi (1990), Clinical Biomechanics of the Spine, J. B. Lippincott Company, Philadelfia 1990.Google Scholar
- 4.P. Salembier, J. Serra (1992), “Morphologicalmultiscale image segmentation”, Proc. of Visual Communications and Image Processing’92, Boston, MA, 1992.Google Scholar
- 6.S. Beucher (1990), Segmentation d’images et morphologie mathématique, PhD Thesis, Paris School of Mines, Paris, June 1990.Google Scholar