A Survey on Morphological Assessment of Knee Articular Cartilage from MR Images

Abstract

Nowadays, articular disorders and musculoskeletal diseases are one of the major health issues and attack usually the older generations. The human knee joint is commonly infected by osteoarthritis (OA) disease. OA can affect the joints of the spine, fingers, thumbs, hips, knees, and toes, but it is most prevalent in the knee and hip joints. Here we are concentrating only on Knee joints. Knee Osteoarthritis is a serious, excruciating & potentially life-threatening joint disease, may lead to permanent disability. It can be detected by measuring changes in knee internal tissues such as cartilage, meniscus and Subchondral Cartilage. For detecting knee osteoarthritis, Magnetic Resonance Imaging is performed to obtain a detailed picture of knee joint and change are measured from MR images. MRI is a test that uses powerful magnets, radio waves, and a make to comprehensive pictures inside our body. The main objective is to develop a fully automated noncontrast MRI application for detection and Segmentation of knee bones. The proposed approach is based on Convolution Neural Network to achieve a robust and perfect segmentation of even highly pathological knee structures.