Semi-automated segmentation and visualisation of outer bone cortex from medical images

Abstract

Good segmentation of the outer bone cortex from medical images is a prerequisite for applications in the field of finite element analysis, surgical planning environments and personalised, case dependent, bone reconstruction. However, current segmentation procedures are often unsatisfactory. This study presents an automated filter procedure to generate a set of adapted contours from which a surface mesh can be deduced directly. The degree of interaction is user determined. The bone contours are extracted from the patients CT data by quick grey value segmentation. An extended filter procedure then only retains contour information representing the outer cortex as more specific internal loops and shape irregularities are removed, tailoring the image for the above-mentioned applications. The developed medical image based design methodology to convert contour sets of multiple bone types, from tibia tumour to neurocranium, is reported and discussed.

Keywords:

• Medical image based segmentation
• Outer bone cortex
• Automated filter procedure
• Multiple bone types

Keywords:

• 65D18
• 68U01
• 65D17
• 68U10

Acknowledgements

We express our gratitude to Dr I Samson (Department of Orthopaedic Surgery, KU Leuven, UZ Pellenberg) for providing the CT data of the humerus bone defect and to the Materialise company, for putting at disposal the Mimics^® 3D image processing and editing software. This research is funded by a PhD grant of the Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen).

Notes

Calculation time using a Celeron^® 2.0 GHz processor, with 512 Mb of internal memory, uncompiled Matlab^® code.