Related Experiment Video
Updated: Jun 21, 2025
09:34
A Method to Estimate Cadaveric Femur Cortical Strains During Fracture Testing Using Digital Image Correlation
Published on: September 14, 2017
7.4K
Segond Fractures Can Be Identified With Excellent Accuracy Utilizing Deep Learning on Anteroposterior Knee
Jacob F Oeding1, Ayoosh Pareek2, Kyle N Kunze2
1School of Medicine, Mayo Clinic Alix School of Medicine, Rochester, Minnesota, U.S.A.
Arthroscopy, Sports Medicine, and Rehabilitation
|July 15, 2024
Summary
A deep learning model accurately detects Segond fractures on knee X-rays, outperforming human experts in speed and accuracy. This AI tool aids in diagnosing associated injuries like meniscus tears.
Area of Science:
- Orthopaedic radiology
- Artificial intelligence in medicine
- Medical imaging analysis
Background:
- Segond fractures are often missed on anteroposterior (AP) knee radiographs.
- Accurate detection is crucial for identifying associated injuries, such as lateral meniscus tears.
Purpose of the Study:
- To develop a deep learning model for Segond fracture detection on AP knee radiographs.
- To compare the model's performance against that of experienced human observers.
Main Methods:
- A dataset of 324 AP knee radiographs was curated from a specialized registry.
- Images were annotated for Segond fractures and divided into training, validation, and testing sets.
- The deep learning model's performance was evaluated against orthopaedic sports medicine experts.
Main Results:
- The deep learning model achieved perfect accuracy, sensitivity, and specificity (100%) on the test set.
- The model demonstrated a mean average precision (mAP) of 0.985 for Segond fracture detection.
- The model was significantly faster than human experts, requiring only 0.3% of their evaluation time.
Conclusions:
- A deep learning model can reliably detect Segond fractures on AP knee radiographs.
- The model exhibits superior performance compared to human experts in terms of accuracy and efficiency.
- Automated detection can improve diagnosis of concomitant injuries and facilitate large-scale research.
