Fragment Distance-Guided Dual-Stream Learning for Automatic Pelvic Fracture Segmentation

Pelvic fracture is a complex and severe injury. Accurate diagnosis and treatment planning require the segmentation of the pelvic structure and the fractured fragments from preoperative CT scans. However, this segmentation is a challenging task, as the fragments from a pelvic fracture typically exhibit considerable variability and irregularity in the morphologies, locations, and quantities. In this study, a novel dual-stream learning framework is proposed for the automatic segmentation and category labeling of pelvic fractures. First, we develop a dual-branch architecture leveraging distance learning from bone fragments for enhanced spatial information, aiding in fracture segmentation and improving the identification of bone fragments with variable positioning. Second, we implement a multi-size feature fusion module that adaptively aggregates features from diverse receptive fields tailored to targets of different sizes and shapes, thus boosting segmentation performance. Extensive experiments on two pelvic fracture datasets from different medical centers demonstrated the accuracy and generalizability of the proposed method. It achieves a mean Dice coefficient and mean F1-score of 0.935[[EQUATION]]0.068 and 0.930[[EQUATION]]0.057 in the dataset FracCLINIC, and 0.955[[EQUATION]]0.072 and 0.912[[EQUATION]]0.129 in the dataset FracSegData, which are superior than other comparing methods. Our method optimizes the process of pelvic fracture segmentation, potentially serving as an effective tool for preoperative planning in the clinical management of pelvic fractures.