THE USE OF ACTIVE SHAPE MODELS FOR MAKING THICKNESS MEASUREMENTS OF ARTICULAR-CARTILAGE FROM MR-IMAGES

Previously reported studies to quantify articular cartilage have used labor-intensive manual or semi-automatic data-driven techniques, demon strating high accuracy and precision. However, none has been able to a utomate the segmentation process. This paper describes a fast, automat ic, model-based approach to segmentation and thickness measurement of the femoral cartilage in 3D T-1-weighted images using active shape mod els (ASMs). Systematic experiments were performed to assess the accura cy and precision of the technique with in vivo images of both normal a nd abnormal knees. Segmentation accuracy was determined by comparing t he results of the segmentation with the boundaries delineated by a rad iologist, The mean error in locating the boundary was 0.57 pixels. To assess the precision of the measurement technique, the mean thickness of the femoral cartilage was calculated for repeated scans of five hea lthy volunteers. A mean coefficient of variation (CV) of 2.8% was obta ined for the thickness measurements.