Related Experiment Video
Updated: Aug 14, 2025
08:05
Design and Analysis for Fall Detection System Simplification
Published on: April 6, 2020
10.8K
Detecting pediatric wrist fractures using deep-learning-based object detection
John R Zech1, Giuseppe Carotenuto2, Zenas Igbinoba3
1Department of Radiology, Columbia University Irving Medical Center/New York Presbyterian Hospital, 622 W 168th St., New York, NY, 10032, USA. jrz2111@columbia.edu.
Pediatric Radiology
|January 17, 2023
Summary
Deep learning accurately detects pediatric wrist fractures, including subtle buckle fractures. AI assistance significantly improved trainee radiograph interpretation accuracy for these fractures.
Area of Science:
- Radiology
- Artificial Intelligence
- Medical Imaging
Background:
- Missed fractures are a leading cause of diagnostic errors in emergency departments.
- Subtle pediatric wrist fractures are particularly challenging to identify on radiographs.
Purpose of the Study:
- Evaluate a deep learning framework for classifying pediatric wrist fractures.
- Assess the AI's performance in augmenting trainee radiograph interpretation.
Main Methods:
- A Faster R-CNN deep learning model was trained on 395 pediatric wrist radiographs.
- Trainee radiologists interpreted radiographs with and without AI assistance.
Main Results:
- The AI model achieved 88% accuracy and 0.92 AUC in detecting fractures.
- AI assistance improved resident accuracy from 80% to 93% for all fractures and 69% to 92% for buckle fractures.
Conclusions:
- Deep learning effectively identifies pediatric wrist fractures, even subtle ones.
- AI significantly enhances diagnostic accuracy for trainees interpreting pediatric wrist radiographs.
