Developing a software system to improve the management and outcome of knee replacement surgery by combining radiographic image information with clinical and patient data

About 100,000 knee replacements are performed each year in the UK to help people with severe knee osteoarthritis. Although most are successful, 1 in 5 patients experience chronic pain after the surgery and 1 in 23 patients will need more surgery within 10 years, due to complications such as the replacement becoming loose. I will develop a computer tool to better understand why a knee replacement may lead to chronic pain after surgery or the need for more surgery. The tool will learn to predict poor outcome based on knee X-ray images, clinical and patient-reported data. For patients and doctors to benefit from the tool more quickly, I will develop guidance and strategies on how to bring such a tool into the clinic. The tool will facilitate informed doctor-patient decision-making and enable personalised treatment, such as preventive pain management, for patients at risk of poor outcome.