Development of a deep learning algorithm for radiographic detection of syndesmotic instability in ankle fractures with intraoperative validation

Abstract

Identifying syndesmotic instability in ankle fractures using conventional radiographs is still a major challenge. In this study we trained a convolutional neural network (CNN) to classify the fracture utilizing the AO-classification (AO-44 A/B/C) and to simultaneously detect syndesmosis instability in the conventional radiograph by leveraging the intraoperative stress testing as the gold standard. In this retrospective exploratory study we identified 700 patients with rotational ankle fractures at a university hospital from 2019 to 2024, from whom 1588 digital radiographs were extracted to train, validate, and test a CNN. Radiographs were classified based on the therapy-decisive gold standard of the intraoperative hook-test and the preoperatively determined AO-classification from the surgical report. To perform internal validation and quality control, the algorithm results were visualized using Guided Score Class activation maps (GSCAM).The AO44-classification sensitivity over all subclasses was 91%. Furthermore, the syndesmosis instability could be identified with a sensitivity of 0.84 (95% confidence interval (CI) 0.78, 0.92) and specificity 0.8 (95% CI 0.67, 0.9). Consistent visualization results were obtained from the GSCAMs. The integration of an explainable deep-learning algorithm, trained on an intraoperative gold standard showed a 0.84 sensitivity for syndesmotic stability testing. Thus, providing clinically interpretable outputs, suggesting potential for enhanced preoperative decision-making in complex ankle trauma.

Fig. 1

Flowchart of patient inclusion.

Fig. 2

Class distribution of the syndesmosis instability per AO-classified data. Each number represents an individual patient.

Fig. 3

Example network (VGG16) and custom toplayer-configuration.

Fig. 4

K-fold cross validation results showing homogeneous accuracy, except for fold 4.

Fig. 5

Confusion Matrices for the average syndesmosis instability classification as well as AO44-B & AO44-C subset classification.

Fig. 6

(a): Guided Score-CAM in higher grade fracture dislocation in an AO 44 B2 fracture of a 58 y old female with an unstable syndesmosis. (b) AO 44B1 fracture in a 40 y old female depicting an unstable syndesmosis with minimal radiologic signs in the a.p. x-ray. (c) False positively predicting an AO44 B1 Fracture in a 67 year-old male with Guided Score CAM focusing on the syndesmotic region.