Related Experiment Video
Updated: Aug 14, 2025
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
Dating birth-related clavicular fractures: pediatric radiologists versus artificial intelligence
Andy Tsai1, Jeannette M Pérez-Rosselló2, Kirsten Ecklund2
1Department of Radiology, Boston Children's Hospital, Harvard Medical School, 300 Longwood Ave., Boston, MA, 02115, USA. andy.tsai@childrens.harvard.edu.
Insights
A deep learning (DL) model accurately dated infant clavicle fractures, outperforming radiologists. This AI tool shows promise for improving the accuracy of fracture dating in child abuse investigations.
Area of Science:
- Medical imaging and artificial intelligence
- Pediatric radiology and forensic medicine
Background:
- Accurate fracture dating in infants is critical for diagnosing child abuse but is challenging due to subjective radiologic interpretation.
- Previous studies used birth-related clavicle fractures to assess healing patterns but did not evaluate radiologist accuracy in dating fractures.
Purpose of the Study:
- To assess the accuracy of radiologists in dating birth-related clavicle fractures.
- To compare radiologist performance against a computer algorithm for fracture age estimation.
Main Methods:
- A database of 416 anteroposterior clavicle radiographs of infants with birth-related fractures was used.
- Three blinded radiologists independently estimated fracture ages, and their results were compared to a deep learning (DL) model.
- Standard error metrics were calculated to evaluate the accuracy of both radiologists and the DL model.
Main Results:
- Radiologists demonstrated moderate to good intra- and inter-reader agreement in estimating fracture ages (Mean Absolute Error: 6.1-7.1 days).
- The DL model achieved a significantly lower Mean Absolute Error (4.2 days) compared to all radiologists (P < 0.001).
- The DL model showed superior correlation with ground truth compared to radiologists' dating estimates.
Conclusions:
- Experienced pediatric radiologists achieved moderate to good agreement in dating clavicular fractures.
- A deep learning model significantly outperformed radiologists in accurately dating infant clavicle fractures.
- AI-driven fracture dating holds potential for enhancing diagnostic accuracy in pediatric forensic evaluations.
Background:
Fracture dating from skeletal surveys is crucial in the diagnosis and investigation of infant abuse. However, this task is challenging because of the subjective nature of the radiologic interpretation and the lack of ground truth. Researchers have used birth-related clavicle fractures as a surrogate to study the radiographic pattern of healing; however, they did not elucidate the accuracy performance of the radiologists in dating fractures.
Objective:
To determine the accuracy of radiologists in dating birth-related clavicle fractures and compare their performance to that achieved by computer algorithm.
Materials And Methods:
We used a previously assembled birth-related clavicle fracture database consisting of 416 anteroposterior clavicle radiographs as the study cohort. The average and standard deviation of the fracture age within this database were 24 days and 18 days, respectively. Three blinded radiologists independently estimated the ages of the clavicle fractures depicted in the radiographs within the database. We compared these estimation results to those made by a recently published deep-learning (DL) model conducted with the identical infant cohort. We calculated standard error metrics to compare the accuracy performances of the radiologists and the computer model.
Results:
The intra- and inter-reader agreements of the fracture age estimates by the radiologists were moderate to good. The radiologists estimated the fracture ages with a mean absolute error (MAE) of 6.1-7.1 days, and standard deviation of the absolute error of 6.3-8.3 days. The accuracy performances of the three radiologists were not significantly different from one another. In comparison, the DL model estimated the age of clavicle fractures with an MAE of 4.2 days, significantly lower than all of the radiologists (P < 0.001).
Conclusion:
Three experienced pediatric radiologists dated clavicular fractures with moderate-good intra- and inter-reader agreements. The correlations between the radiologists' estimates and the ground truth were moderate to good. The fracture ages assigned by the DL model showed superior correlation with the ground truth compared to radiologists' dating estimates.
