Researchers from the Universities of Liverpool and Manchester, together with Manchester Imaging Ltd, (a North West based company specialising in developing AI medical devices), have received a £1.2 million grant from the National Institute for Health and Care Research's (NIHR) 'Invention for Innovation' (i4i) programme, to build an automatic system for measuring hip displacement in cerebral palsy patients.
Lead Clinician, Professor Daniel Perry, NIHR Research Professor at the University of Liverpool and a children's orthopaedic surgeon at Alder Hey Children's NHS Foundation Trust said: "AI will revolutionise the care we provide, enhance diagnostics and care pathways and free up time for our clinicians to do what they do best: caring for our children and young people. This is a great example - a practical tool directly focused on better care for children with cerebral palsy."
Children with cerebral palsy are at high risk of developing hip problems, with the ball of the hip moving out of the socket. This movement can cause the child severe pain. The dislocation, however, can be prevented through regular X-ray measurements and prompt intervention with reliable procedures if a problem is spotted.
The system, developed in conjuncture with clinicians at Alder Hey Children's NHS Foundation Trust, is intended to be integrated into the Cerebral Palsy Integrated Pathway (CPIP), the national framework used to monitor the musculoskeletal systems of children with cerebral palsy. CPIP involves affected children receiving regular assessment, physical examination and regular hip X-rays, which are then examined by medical experts in order to identify changes and predict risks.
This process, however, is not nationally standardised, and uptake differs between regions. Due to the large amount of clinician time it consumes, and the extra costs and delays involved, levels of CPIP uptake are often limited by the resources available to a particular region. This means that the standard of care for a child with cerebral palsy may be higher in one area of the country than another.
This new tool, however, will help to change that - by automating the process of hip x-ray interpretation, data capture and monitoring, enabling more patients to benefit from early detection and prevention as a result.
The AI algorithm has been trained using thousands of X-ray images and is capable of automatically locating the outline of children's hip bones, and is able to detect cases where the hips are just beginning to dislocate, through to full dislocation. The accuracy of the tool has been thoroughly tested and was found by researchers to be similar to that of human medical experts, while taking a fraction of the time to perform the analysis.
Manchester Imaging Ltd will take the AI algorithm developed at the University of Manchester and build a Medical Device that will be integrated into hospital systems, making it easy for clinicians to use.
The medical device will be used to monitor hip movement, picking out areas of concern in hip X-rays and flagging up areas where a serious problem is likely to occur, identifying when preventative intervention is likely to be needed.
The researchers say that by using the tool, clinicians will save significant amounts of time and will improve patient outcomes by speeding up the treatment process.
By using this tool to processes thousands of images across the country, X-ray image data will be automatically entered into the national CPIP database. This will enable new research to better understand the course of the disease and the benefits of monitoring.
Dr Steve Cooke, national orthopaedic lead for CPIP, remarked: "With nearly 14,000 children on CPIP there is a huge opportunity for ground-breaking research, but we need more and better data. An accurate, streamlined tool that automates what is currently a labour-intensive task will transform the way we monitor the hip in children with cerebral palsy."
