Abstract
Category:
Ankle; Trauma
Introduction/Purpose:
Early and accurate detection of ankle fractures is crucial for reducing future complications. Radiographs are the most abundant imaging techniques for assessing fractures. We believe deep learning (DL) methods, through adequately trained deep convolutional neural networks (DCNNs), can assess radiographic images fast and accurate without human intervention. In this study, we aimed to assess the performance of two different DCNNs in detecting ankle fractures using radiographs compared to the ground truth.
Methods:
In this retrospective study, our DCNNs were trained using radiographs obtained from 1050 patients with ankle fracture and the same number of individuals with otherwise healthy ankles. Inception V3 and Renet50 pre-trained models were used in our algorithms. Danis-Weber classification method was used. Out of 1050, 72 individuals were labeled as occult fractures as they were not detected in the primary radiographic assessment. Using single-view radiographs was compared with 3-views (anteroposterior, mortise, lateral) for training the DCNNs.
Results:
Our DCNNs showed a better performance using 3-views images versus single-view based on greater values for accuracy, F-score, and area under the curve (AUC). The sensitivity and specificity in detection of ankle fractures using 3-views were 97.5% and 93.9% using Resnet50 compared to 98.7% and 98.6 using inception V3, respectively. Resnet50 missed 3 occult fractures while Inception V3 missed only one case. In cases that detected the fracture, the saliency map showed the location of the fracture (Figure 1).
Conclusion:
The performance of our DCNNs showed a promising potential that can be considered in developing the currently used image interpretation programs or as a separate assistant to the clinicians to detect ankle fractures faster and more precisely.
