**Background** Cerebral palsy (CP) is the most common physical disability in childhood. Through abnormal muscle forces, children affected are at high risk of hip dislocation, leading to severe pain, loss of function and difficulties with personal care. Hip dislocation can be prevented through surveillance programmes and staged interventions. Pelvic radiographs can be used to monitor hip health by measuring a number of predictive parameters, such as Reimer’s Migration Percentage, which measures the amount of the hip that has ‘escaped’ (subluxed) from the hip socket. Subluxation that is rapidly progressing, or meeting a defined threshold, prompts intervention. **Objectives** We are developing a machine learning-based software system to automatically measure Reimer’s migration percentage, acetabular index, head shaft angle and neck shaft angle from hip radiographs of children with CP. The system automatically locates points around the femoral head and acetabulum on pelvic radiographs, and uses these to calculate the measurements. Visualising the point positions used to generate the measurements will enable clinicians to easily verify the automatically generated measurements. The overall goal is to integrate our system into standardised surveillance programmes which monitor musculoskeletal health in children with CP, such as the Cerebral Palsy Integrated Pathway (CPIP) programme in the UK. **Patient Involvement** We are involving affected families throughout the development of the software, with parent co-investigators and workshops. **Impact** The system has the potential to save clinicians time, and to improve patient care by enabling more comprehensive, consistent, and reliable monitoring of hip migration in children with CP. The system will also enable the collection of high-quality, population-wide hip measurement data for children with CP, informing important future research. <br> <img src="https://personalpages.manchester.ac.uk/staff/claudia.lindner/img/RMP_measurement_example.png">