# Elbow trauma in children: development and evaluation of radiological artificial intelligence models

## TL;DR

This study developed AI models to detect elbow injuries in children using X-rays and tested their impact on radiologists' performance.

## Contribution

The study demonstrates how AI models with similar in silico metrics can have divergent real-world impacts on radiologists.

## Key findings

- Model 1 improved radiologists' sensitivity and accuracy in detecting elbow injuries.
- Model 2 decreased radiologists' specificity when used in clinical practice.
- Model 1 maintained good performance on an external test set, while Model 2's performance dropped significantly.

## Abstract

To develop a model using artificial intelligence (A.I.) able to detect post-traumatic injuries on pediatric elbow X-rays then to evaluate its performances in silico and its impact on radiologists’ interpretation in clinical practice.

A total of 1956 pediatric elbow radiographs performed following a trauma were retrospectively collected from 935 patients aged between 0 and 18 years. Deep convolutional neural networks were trained on these X-rays. The two best models were selected then evaluated on an external test set involving 120 patients, whose X-rays were performed on a different radiological equipment in another time period. Eight radiologists interpreted this external test set without then with the help of the A.I. models .

Two models stood out: model 1 had an accuracy of 95.8% and an AUROC of 0.983 and model 2 had an accuracy of 90.5% and an AUROC of 0.975. On the external test set, model 1 kept a good accuracy of 82.5% and AUROC of 0.916 while model 2 had a loss of accuracy down to 69.2% and of AUROC to 0.793. Model 1 significantly improved radiologist's sensitivity (0.82 to 0.88, P = 0.016) and accuracy (0.86 to 0.88, P = 0,047) while model 2 significantly decreased specificity of readers (0.86 to 0.83, P = 0.031).

End-to-end development of a deep learning model to assess post-traumatic injuries on elbow X-ray in children was feasible and showed that models with close metrics in silico can unpredictably lead radiologists to either improve or lower their performances in clinical settings.