An automated method to segment the femur for osteoarthritis research.

In this paper we develop a fully automated method for the segmentation of the femur in axial MR images and its use in the analysis of imaging biomarkers for osteoarthritis (OA). The proposed method is based on anatomical constraints implemented using morphological operations to extract the femur medulla and a level set evolution to extract the femur cortex. The average agreement of the automated segmentation algorithm with ground truth manual segmentations was 0.94 +/- 0.03 calculated using the Zijdenbos similarity index (ZSI). A pooled variance t-test analysis found significant associations between the KL grade, a clinical measure of OA severity, and both the cross-sectional area (CSA) of the femur medulla (p = 0.02) and the ratio of the femur medulla CSA to the femur cortex CSA (p = 0.04) for women. No significant association between femur measurements and KL grade was found for men.