4DRGS: 4D Radiative Gaussian Splatting for Efficient 3D Vessel Reconstruction from Sparse-View Dynamic DSA Images

doi:arXiv:2412.12919

	4DRGS: 4D Radiative Gaussian Splatting for Efficient 3D Vessel Reconstruction from Sparse-View Dynamic DSA Images
	Liu, Zhentao1,2 ; Zha, Ruyi 3; Zhao, Huangxuan 4; Li, Hongdong 3; Cui, Zhiming1,2
	2024-12-17
状态	已发表
摘要	Reconstructing 3D vessel structures from sparse-view dynamic digital subtraction angiography (DSA) images enables accurate medical assessment while reducing radiation exposure. Existing methods often produce suboptimal results or require excessive computation time. In this work, we propose 4D radiative Gaussian splatting (4DRGS) to achieve high-quality reconstruction efficiently. In detail, we represent the vessels with 4D radiative Gaussian kernels. Each kernel has time- invariant geometry parameters, including position, rotation, and scale, to model static vessel structures. The time-dependent central attenuation of each kernel is predicted from a compact neural network to capture the temporal varying response of contrast agent flow. We splat these Gaussian kernels to synthesize DSA images via X-ray rasterization and optimize the model with real captured ones. The final 3D vessel volume is voxelized from the well-trained kernels. Moreover, we introduce accumulated attenuation pruning and bounded scaling activation to improve reconstruction quality. Extensive experiments on real-world patient data demonstrate that 4DRGS achieves impressive results in 5 minutes training, which is 32Ã faster than the state-of-the-art method. This underscores the potential of 4DRGS for real-world clinics.
关键词	Sparse-View DSA Reconstruction Gaussian Splatting
语种	英语
DOI	arXiv:2412.12919
相关网址	查看原文
出处	Arxiv
收录类别	PPRN.PPRN
WOS记录号	PPRN:120003306
WOS类目	Computer Science, Software Engineering ; Engineering, Electrical& Electronic
文献类型	预印本
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/483984
专题	生物医学工程学院信息科学与技术学院_博士生生物医学工程学院_PI研究组_崔智铭组
通讯作者	Cui, Zhiming
作者单位	1.ShanghaiTech Univ, Sch Biomed Engn, Shanghai, Peoples R China 2.ShanghaiTech Univ, State Key Lab Adv Med Mat & Devices, Shanghai, Peoples R China 3.Australian Natl Univ, Canberra, Australia 4.Wuhan Univ, Sch Comp Sci, Wuhan, Peoples R China
推荐引用方式 GB/T 7714	Liu, Zhentao,Zha, Ruyi,Zhao, Huangxuan,et al. 4DRGS: 4D Radiative Gaussian Splatting for Efficient 3D Vessel Reconstruction from Sparse-View Dynamic DSA Images. 2024.