Knee Joint Articular Cartilage Segmentation using Radial Search Method, Visualization and Quantification

摘要

Knee is a complex and highly stressed joint of the human body. Articular Cartilage is a smooth hyaline spongy material between the tibia and femur bones of knee joint. Cartilage morphology change is an important biomarker for the progression of osteoarthritis (OA). Magnetic Resonance Imaging (MRI) is the modality widely used to image the knee joint because of its hazard free and soft tissue contrast. Cartilage thickness measurement and visualization is useful for early detection and progression of the disease in case of OA affected patients. In the present work, knee joint MR images of normal and OA affected are processed for segmentation and visualization of cartilage using semiautomatic method. The radial search method is used with minor modifications in search area to reduce computation time. Cartilage thickness and volume is measured in lateral, medial and patellar regions of femur. The overall accuracy of measurements is determined by comparing the measurements with another semiautomatic method based on edge detection and interpolation. It is observed a good correlation between quantification of cartilage in two methods. The method takes less time for segmentation because of reduced manual steps. The reduced cartilage thickness and volume is observed in OA affected knee of different level of progression.