@article{SPITTLE2025103379,
title = {Towards universal early screening for cerebral palsy: a roadmap for automated General Movements Assessment},
journal = {eClinicalMedicine},
volume = {86},
pages = {103379},
year = {2025},
issn = {2589-5370},
doi = {https://doi.org/10.1016/j.eclinm.2025.103379},
url = {https://www.sciencedirect.com/science/article/pii/S2589537025003116},
author = {Alicia J. Spittle and Peter B. Marschik and Lars Adde and Nadia Badawi and Rachel Byrne and Arend F. Bos and Alain Chatelin and John Coughlan and Francesca Fedeli and Andrea Guzzetta and Edmond S.L. Ho and Michelle J. Johnson and Amanda Kwong and Alistair McEwan and Catherine Morgan and Anderson Mughogho and Deirdre M. Murray and Silvia Orlandi and Colleen Peyton and Laura A. Prosser and Anina Ritterband-Rosenbaum and Truyen Tran and Dajie Zhang and Elyse Passmore},
keywords = {Artificial intelligence, Machine learning, General movements, Early detection, Cerebral palsy},
abstract = {Summary
Cerebral palsy (CP) is the most common lifelong physical disability, affecting millions globally. Early detection and intervention are crucial for improving outcomes, yet many children are diagnosed late. The General Movements Assessment (GMA) is a highly accurate clinical tool for detecting infants at high probability of CP, but access to health professionals trained in the GMA limits its use. Artificial intelligence (AI) has the potential to automate the GMA, increasing accessibility worldwide. We established an interdisciplinary, international consortium for the purpose of developing a roadmap for the ongoing development and implementation of an AI-enabled GMA system for universal CP screening worldwide. The consortium included clinicians (children neurologists, paediatricians, neonatologists, allied health), researchers, engineers, computer scientists, legal experts, and individuals with lived experience, from around the globe (across Africa, Australia, Europe, and North America). The roadmap identifies the following steps and key requirements within: (1) development of standards for AI validation; (2) development of AI-GMA from large and diverse validation sets; (3) development of software tools and clinical pathways; (4) regulatory requisites; and (5) implementation. With the roadmap, AI-enabled screening for CP incorporating state-of-the-art technology can be made possible. Future work will require international collaboration to allow for scaling of data sets, refining automated solutions and translation into practice.
Funding
Cerebral Palsy Foundation, Cerebral Palsy Alliance, European Union Born to Get There, the National Health and Medical Research Council.}
}