Type:
Educational Exhibit
Keywords:
Musculoskeletal joint, MR, Computer Applications-Detection, diagnosis, Education, Quality assurance
Authors:
J. Gajera, A. Ahluwalia, A. Howie, B. Lurie, M. Roberts, D. Hunter; Sydney, NSW/AU
DOI:
10.26044/ranzcr2021/R-0162
Background
Knee osteoarthritis (KOA) is an extremely common musculoskeletal disease and the most common form of arthritis causing pain, mobility limitation, affecting independence and quality of life.1 In current practice, there remains an inability to systematically diagnose the disease at an early stage, whereby it may be possible to slow down its progression. KOA is a big data problem in terms of data complexity, heterogeneity and size as it has been commonly considered in the literature and one of the most important hurdles in KOA management is identifying and classifying patients who will benefit most from treatment.2 Medical imaging data is a crucial source of information and many studies focus on developing prediction models for KOA based on medical imaging (Magnetic Resonance Imaging (MRI), X-ray)3-7, clinical information, self-reported and biomechanical data. Deep Learning (DL) could play a key role towards these directions, extracting valuable knowledge from various types of clinical data (biomechanical parameters, images, kinematics) and finding new solutions that utilise data from the greatest possible variety of sources.