Semi automatic detection of synovial boundaries in water-immersion ultrasound examination

Rheumatoid arthritis (RA) is a chronic multisystemic autoimmune disease. Its early diagnosis and activity assessment are essential to adjust the proper therapy. Ultrasonography (US) allows direct visualization of early inflammatory joint changes, while being rapidly performed and easily accepted by patients. We propose an algorithm to semi automatically detect synovial boundaries on US images, making minimal use of a priori information on the morphological shape or on the appearance of the joint and of the synovia. After an image denoising step, three joint landmarks are manually identified. In order to identify the synovia-bone and the synovia-soft tissues interfaces, and to tackle the morphological variability of diseased joints, a cascade of two different active contours is developed, whose composition identifies the whole synovial boundary. By comparison with a manual segmentation performed by two radiologists, we obtained an overall mean sensitivity of 86.4% ±11.6%, and a mean value of 76.8% ±7.8% for Dice's similarity index.