用2D扩散模型生成高保真3D医学影像，兼顾效率与解剖一致性。
TRACE: Temporally Reliable Anatomically-Conditioned 3D CT Generation with Enhanced Efficiency
- 将序列2D切片视为视频帧对，结合分割先验与报告实现解剖对齐
- 通过重叠帧策略生成长轴3D体积，保持时空连贯性，推理效率提升40%
- 适合资源有限的临床场景，尤其适用于数据增强与隐私保护
3D医学图像生成对数据增强和患者隐私保护至关重要，亟需兼具可靠性和高效性的模型以适应临床应用。然而，现有方法存在解剖保真度低、轴向长度受限及计算开销大等问题，难以在资源有限地区部署。本文提出TRACE框架，采用2D多模态条件扩散模型生成具有时空一致性的3D CT图像。将连续2D切片建模为视频帧对，融合分割先验与放射科报告实现解剖对齐，并引入光流确保时间连贯性。推理时，通过重叠帧策略将帧对串联成可变长度序列，重建为时空与解剖对齐的3D体数据。实验表明，TRACE在保持解剖精度与时空一致性的同时显著提升计算效率。代码已开源：https://github.com/VinyehShaw/TRACE。
原文摘要 · Abstract (English)
3D medical image generation is essential for data augmentation and patient privacy, calling for reliable and efficient models suited for clinical practice. However, current methods suffer from limited anatomical fidelity, restricted axial length, and substantial computational cost, placing them beyond reach for regions with limited resources and infrastructure. We introduce TRACE, a framework that generates 3D medical images with spatiotemporal alignment using a 2D multimodal-conditioned diffusion approach. TRACE models sequential 2D slices as video frame pairs, combining segmentation priors and radiology reports for anatomical alignment, incorporating optical flow to sustain temporal coherence. During inference, an overlapping-frame strategy links frame pairs into a flexible length sequence, reconstructed into a spatiotemporally and anatomically aligned 3D volume. Experimental results demonstrate that TRACE effectively balances computational efficiency with preserving anatomical fidelity and spatiotemporal consistency. Code is available at: https://github.com/VinyehShaw/TRACE.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。