Critical Evaluation of Deep Neural Networks for Wrist Fracture Detection
This paper highlights a critical limitation of current deep learning models for medical diagnosis, specifically wrist fracture detection, by demonstrating their poor performance on difficult cases that require expert confirmation, which is important for clinicians and medical AI developers.
This study developed and evaluated DeepWrist, a deep learning pipeline for wrist fracture detection, on both a general population and a challenging test set of cases requiring CT confirmation. DeepWrist achieved near-perfect performance on the general set (average precision 0.99, AUC 0.99) but showed substantially lower performance on the challenging set (average precision 0.64, AUC 0.84).
Wrist Fracture is the most common type of fracture with a high incidence rate. Conventional radiography (i.e. X-ray imaging) is used for wrist fracture detection routinely, but occasionally fracture delineation poses issues and an additional confirmation by computed tomography (CT) is needed for diagnosis. Recent advances in the field of Deep Learning (DL), a subfield of Artificial Intelligence (AI), have shown that wrist fracture detection can be automated using Convolutional Neural Networks. However, previous studies did not pay close attention to the difficult cases which can only be confirmed via CT imaging. In this study, we have developed and analyzed a state-of-the-art DL-based pipeline for wrist (distal radius) fracture detection -- DeepWrist, and evaluated it against one general population test set, and one challenging test set comprising only cases requiring confirmation by CT. Our results reveal that a typical state-of-the-art approach, such as DeepWrist, while having a near-perfect performance on the general independent test set, has a substantially lower performance on the challenging test set -- average precision of 0.99 (0.99-0.99) vs 0.64 (0.46-0.83), respectively. Similarly, the area under the ROC curve was of 0.99 (0.98-0.99) vs 0.84 (0.72-0.93), respectively. Our findings highlight the importance of a meticulous analysis of DL-based models before clinical use, and unearth the need for more challenging settings for testing medical AI systems.