RF-DTR: A Multi-Stage DCT Token Regression Network for Progressive Rib Fracture Mask Refinement

Chen, ShouYu; Hu, Liang; Wang, JunTao; Naseem, Usman; Lai, Zhongyuan; Zhang, Qi

doi:10.24963/ijcai.2025/1017

RF-DTR: A Multi-Stage DCT Token Regression Network for Progressive Rib Fracture Mask Refinement

ShouYu Chen, Liang Hu, JunTao Wang, Usman Naseem, Zhongyuan Lai, Qi Zhang

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. AI4Tech: AI Enabling Technologies. Pages 9149-9157.

https://doi.org/10.24963/ijcai.2025/1017

Keywords

Domain-specific AI4Tech: AI4Care and AI4Health
Advanced AI4Tech: Data-driven AI4Tech

Abstract

Rib fracture patterns are key indicators of trauma severity. Detecting and locating these fractures is a critical yet time-consuming task, especially in 3D imaging, due to their minute size and irregular geometries. Existing voxel-based spatial methods fail to capture frequency-domain variations inherent in imaging and do not replicate the progressive refinement process used by clinicians during manual annotation, leading to suboptimal results. We propose a novel regression network, RF-DTR, incorporating a gated regressor mechanism and operating entirely in the frequency domain to address these challenges. Specifically, we present an innovative spatial-frequency transform applied to volumes and corresponding masks. Furthermore, we introduce a Mahalanobis regularization technique to enhance the model and learn high-frequency DCT components relevant to clinical tasks. Finally, a hierarchical penalty is proposed to improve the confidence of the prediction. Extensive experiments confirm our method's superiority in handling complex, sparsely annotated medical imaging datasets.

ShouYu Chen, Liang Hu, JunTao Wang, Usman Naseem, Zhongyuan Lai, and Qi Zhang. 2025. RF-DTR: A Multi-Stage DCT Token Regression Network for Progressive Rib Fracture Mask Refinement. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 9149–9157. https://doi.org/10.24963/ijcai.2025/1017

@inproceedings{ijcai2025p1017,
  title     = {RF-DTR: A Multi-Stage DCT Token Regression Network for Progressive Rib Fracture Mask Refinement},
  author    = {Chen, ShouYu and Hu, Liang and Wang, JunTao and Naseem, Usman and Lai, Zhongyuan and Zhang, Qi},
  booktitle = {Proceedings of the Thirty-Fourth International Joint Conference on
               Artificial Intelligence, {IJCAI-25}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {James Kwok},
  pages     = {9149--9157},
  year      = {2025},
  month     = {8},
  note      = {AI4Tech: AI Enabling Technologies},
  doi       = {10.24963/ijcai.2025/1017},
  url       = {https://doi.org/10.24963/ijcai.2025/1017},
}

TY  - CPAPER
TI  - RF-DTR: A Multi-Stage DCT Token Regression Network for Progressive Rib Fracture Mask Refinement
AU  - Chen, ShouYu
AU  - Hu, Liang
AU  - Wang, JunTao
AU  - Naseem, Usman
AU  - Lai, Zhongyuan
AU  - Zhang, Qi
T2  - Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2025
SP  - 9149
EP  - 9157
DO  - 10.24963/ijcai.2025/1017
UR  - https://doi.org/10.24963/ijcai.2025/1017
ER  -