I am interested in algorithms developed by the machine learning and computer vision communities, their application to increasing the understanding of musculoskeletal disorders and early diagnosis of these conditions.
Osteoarthritis is a painful and debilitating disease of the joints. It is not only damaging to an individual’s quality of life but it also affects ~8.5 million people in the UK, costing the UK approximately 1% of its GNP. For my PhD, I wrote software which recognises early signs of osteoarthritis from MR (magnetic resonance) images. Through earlier diagnosis, we hope to recommend more suitable treatments to medical staff at an earlier stage in the disease process.
The software uses machine learning principles to identify important patterns from MR images which reveal the current disease status. Machine learning algorithms provide the computer with the ability to extract information without being explicitly programmed. The algorithms extract these patterns using thousands of examples of knee MR images alongside a radiologist’s diagnosis. This technique enables the algorithms to find important features which may not be apparent to human readers. When given a knee MR image for a new patient, the software predicts the severity of osteoarthritis.
Prior to working in academia I was a software engineer at an investment bank. Software sustainability principles were very important in this dynamic environment due to very large teams of engineers and a fast route to production ready code.