Aim
Compared to radiologists, ED physicians are at a disadvantage in reading radiographic studies due to the utilization of non-dedicated diagnostic monitors, suboptimal room lighting, and significant time and patient turnaround pressures.
The objective of this study is to evaluate the diagnostic performance of an AI radiograph fracture detection support tool (RBfracture™ version 1.5) compared to radiographic assessment by emergency department doctors based on discrepant radiographs between ED physicians and radiologists.
Key findings
- RBfracture demonstrated a high specificity of 95.5% and PPV of 93.8%, implying that patients will unlikely undergo unnecessary fracture treatment
- The study shows that RBfracture can potentially reduce 75.0% recalls due to discrepant radiograph findings
- Having a fracture detection AI solution has the potential of reducing discrepant cases by up to 73.7% in the ED setting
- An AI solution can be useful in times of reduced manpower hours for both radiologists and ED physicians
Open access
This research is open access
This original research can be read in The Medical Journal of Malaysia, published in July 2025.
Funding and conflicts of interests
There were no financial or non-financial support by the vendors for this study. This study has been approved by the Institutional Review Board (IRB) (IRB registration 2023/2159).
