X^{2}-高斯：用於連續時間斷層重建的四維輻射高斯分布擬合

X^{2}-Gaussian: 4D Radiative Gaussian Splatting for Continuous-time Tomographic Reconstruction

March 27, 2025

作者: Weihao Yu, Yuanhao Cai, Ruyi Zha, Zhiwen Fan, Chenxin Li, Yixuan Yuan

cs.AI

摘要

四維計算機斷層掃描（4D CT）重建對於捕捉動態解剖變化至關重要，但傳統的相位分箱工作流程存在固有侷限。現有方法通過呼吸門控設備將時間分辨率離散化為固定相位，導致運動對齊誤差並限制了臨床實用性。本文提出X^2-Gaussian，這是一種新穎框架，通過將動態輻射高斯潑濺與自監督呼吸運動學習相結合，實現了連續時間的4D-CT重建。我們的方法通過時空編碼-解碼架構建模解剖動態，預測時變高斯變形，從而消除相位離散化。為擺脫對外部門控設備的依賴，我們引入了一種生理驅動的週期一致性損失，通過可微分優化直接從投影中學習患者特定的呼吸週期。大量實驗證明了其最先進的性能，相較傳統方法實現了9.93 dB的PSNR提升，並比先前的高斯潑濺技術提高了2.25 dB。通過將連續運動建模與無硬件週期學習相統一，X^2-Gaussian推動了動態臨床成像中高保真4D CT重建的發展。項目網站：https://x2-gaussian.github.io/。

English

Four-dimensional computed tomography (4D CT) reconstruction is crucial for capturing dynamic anatomical changes but faces inherent limitations from conventional phase-binning workflows. Current methods discretize temporal resolution into fixed phases with respiratory gating devices, introducing motion misalignment and restricting clinical practicality. In this paper, We propose X^2-Gaussian, a novel framework that enables continuous-time 4D-CT reconstruction by integrating dynamic radiative Gaussian splatting with self-supervised respiratory motion learning. Our approach models anatomical dynamics through a spatiotemporal encoder-decoder architecture that predicts time-varying Gaussian deformations, eliminating phase discretization. To remove dependency on external gating devices, we introduce a physiology-driven periodic consistency loss that learns patient-specific breathing cycles directly from projections via differentiable optimization. Extensive experiments demonstrate state-of-the-art performance, achieving a 9.93 dB PSNR gain over traditional methods and 2.25 dB improvement against prior Gaussian splatting techniques. By unifying continuous motion modeling with hardware-free period learning, X^2-Gaussian advances high-fidelity 4D CT reconstruction for dynamic clinical imaging. Project website at: https://x2-gaussian.github.io/.