Search papers, labs, and topics across Lattice.
This paper introduces an unsupervised domain adaptation (UDA) framework that addresses the challenges of 3D segmentation in cone-beam CT (CBCT) by employing redundancy-reducing feature alignment, allowing for effective segmentation without the need for target-domain annotations. The method is evaluated on two liver segmentation benchmarks, demonstrating superior performance compared to existing pretrained models and UDA strategies, highlighting the necessity of explicit feature-space bridging for generalization across different imaging modalities. By releasing the liver segmentations and associated resources, the authors facilitate reproducibility and further research in this critical area of medical imaging.
Unsupervised domain adaptation can dramatically enhance 3D CBCT segmentation accuracy without requiring any target-domain annotations.
Accurate 3D segmentation of cone-beam CT (CBCT) is critical for interventional and radiation therapy applications, yet it remains limited by two compounding challenges: the scarcity of annotated CBCT data and the large domain shift from diagnostic CT. Interventional CBCT exhibits fundamental modality differences from conventional CT, driven by acquisition and physics effects as well as contrast-specific vascular content, thereby limiting effective cross-modality model transfer. We propose a novel unsupervised domain adaptation (UDA) framework based on redundancy-reducing feature alignment, enabling 3D CBCT segmentation with no target-domain annotations or inference-time adaptation. Our framework is architecture-agnostic, seamlessly adapting both CNN-based and ViT-based foundation models. We evaluate our method on two challenging CT-CBCT liver segmentation benchmarks: one for interventional vascular procedures and one for radiation therapy, demonstrating that even large-scale pretrained segmentation networks require explicit feature-space bridging to generalize across acquisition modalities, and that our approach consistently outperforms existing pretrained foundation model and UDA strategies. To support reproducibility and benchmarking, we release the liver segmentations for a public CBCT dataset, along with the code, trained models, and weights.
