Lead researchers: Professor Michael Nilsson, Professor Rohan Walker, Professor Sarah Johnson, Associate Professor Adrian Wills, Doctor Nattai Borges, Associate Professor Michael Pollack
Associated research group: Centre for Rehab Innovations
Most people make an excellent recovery after knee replacement, thanks to advances in surgical methods. Still, there is a lack of scientific evidence to explain why some people get back into life quickly with their new knee, whilst others recover more slowly. Researchers at the University of Newcastle are developing a tool, powered by machine-learning, that will take the guess-work out of deciding which, how much and what kinds of rehabilitation will be the most useful for each patient.
More than 50,000 knee replacement procedures are performed annually in Australia, an increase of around 36% since 2006, and with an ageing population this is set to rise further. With a growing number of people requiring joint replacement surgery, there’s a significant need to have more tailored and precision-based rehabilitation interventions, just as we have for other areas in medicine like cancer and asthma.
We’ve often taken a generic approach previously, despite the fact that individuals have different capacity for recovery. Quite often it’s just not the medical situation but life circumstances that determine the success of a rehabilitation program. Some patients need to stay in hospital, others can get by with support in the community or home setting, and we need to work at all levels to ensure everyone is receiving the most appropriate treatment for their recovery.
The study involves a multidisciplinary team of clinicians, engineers and IT developers who have begun designing and refining the new precision medicine tool for clinical stratification and decision-making.
The technology will be scalable and adaptable for a range of conditions, the goal being to predict, with precision, which rehabilitation interventions individual people will need as they recover. The project will go beyond specified clinical information to include multiple parameters including determinants of health such as carer or family support, mental health factors like stress or depression, comorbidities, financial hardship and so on.