Hip fractures are associated with significant morbidity and mortality, posing a major health challenge in aging populations such as Hong Kong. Accurate and timely detection is critical to saving lives and preserving patient independence. Despite advancements in medical imaging, X-rays remain the primary diagnostic tool for evaluating hip fractures. However, misdiagnosis rates can reach up to 10%, leading to delays in treatment. The emergence of artificial intelligence (AI) offers a promising solution, with advanced algorithms such as convolutional neural networks (CNNs) demonstrating potential in improving fracture detection. Here, we share our experience of a collaboration between local academic institutions and the Hospital Authority Data Collaboration Laboratory (HADCL) in developing an AI model for hip fracture detection on pelvic X-rays. We detail its integration into the Hospital Authority's Clinical Management System (CMS) and subsequent territory-wide deployment, exploring how this tool supports frontline clinicians and enhances patient care.
