Shape modelling for predicting risk of early-onset hip osteoarthritis
Lead Research Organisation:
University of Bath
Department Name: Mathematical Sciences
Abstract
Osteoarthritis of the hip is one of the most prevalent musculoskeletal disorders and gives rise to a large level of disability, carrying a significant burden for individuals' quality of life, society and healthcare provision systems. The research will develop novel image-processing techniques to identify pelvis shapes at highest risk of developing OA. Early diagnosis is essential for successful intervention, and we will provide practical techniques for automating this task using computed tomography (CT) images of the pelvis. We will achieve this by modelling the shape of the pelvis and classifying shapes at risk of early-onset osteoarthritis. To infer shapes at risk, we will exploit a set of two-hundred CT scans that have been annotated by medics.
| Description | This grant is six-months since starting and we have developed a model for the shapes of pelvises and are currently training it on data. We're facing computational challenges and are currently implementing algorithmic and programming tricks to train the model on the full data set. |
| Exploitation Route | The methodologies are broadly applicable to shape modelling and classification. When the methodology has been applied successfully to pelvis data, we hope to apply in other scenarios (arthritis in hands and knees). |
| Sectors | Healthcare |