Machine learning and deep learning for ankle fracture detection in radiographs: A systematic review with narrative synthesis on diagnostic performance and future clinical integration.
Jitendra Mangwani, S P Jithin, Nimra Akram et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review evaluates the use of machine learning (ML) and deep learning (DL) for detecting ankle fractures in radiographs, analyzing 27 studies published from 2019 to June 2025. The findings indicate a growing trend towards complex tasks like fracture localization, with improved clinician accuracy and efficiency noted in recent studies. However, the lack of dataset heterogeneity and external validation limits generalizability.
Key findings
- A significant evolution towards complex tasks such as fracture localization was observed.
- Recent studies demonstrated improved clinician accuracy and efficiency with external validation of ankle-specific models.
- The review identified a clear trend of increasing research volume focused on X-ray imaging.
Why it matters
Understanding the advancements in ML and DL for ankle fracture detection can enhance diagnostic accuracy and efficiency in clinical settings, potentially reducing missed fractures and improving patient outcomes.
What to keep in mind
A significant limitation across studies was the lack of heterogeneity in datasets and external validation, which hindered generalizability.