Abstract #1392

Automated Bone Segmentation and Bone-Cartilage Interface Extraction from MR Images of the Hip

Ying Xia¹,², Jurgen Fripp², Shekhar Chandra², Olivier Salvado², Raphael Schwarz³, Lars Lauer³, Craig Engstrom⁴, Stuart Crozier¹

¹School of Information Technology and Electrical Engineering, University of Queensland, Brisbane, Australia; ²The Australia e-Health Research Centre, CSIRO ICT Centre, Brisbane, Australia; ³Siemens Healthcare, Erlangen, Germany; ⁴School of Human Movement Studies, University of Queensland, Brisbane, Australia

Osteoarthritis (OA) is a common disease of the hip joint characterized by changes in structure and degeneration of cartilage tissue. Magnetic Resonance (MR) imaging has been shown to be an ideal modality for OA assessment, providing direct and non-invasive visualization of joint structure. Morphological measurements (volume, thickness and surface area) of the cartilage tissue have been shown to be important in characterizing and monitoring OA progression, which allows the prediction of its subsequent changes and in-time therapeutic treatment before permanent damage has been developed. In this paper, we present validation of our fully automated scheme for the bone segmentation and qualitative bone-cartilage interface (BCI) extraction and initial cartilage segmentation from MR images of the hip joint.