FB-Diff: Fourier Basis-guided Diffusion for Temporal Interpolation of 4D Medical Imaging

2025-07-06

AI Summary

To address the mismatch between linear motion assumptions and realistic nonlinear, quasi-periodic respiratory motion in 4D medical image temporal interpolation, this paper proposes a novel frequency-domain-driven generative interpolation paradigm. Methodologically, we introduce a Fourier motion operator that jointly incorporates physiological priors and spectral information from feature space, and explicitly models frequency dynamics during diffusion via a basis interaction mechanism. Additionally, we integrate a variational autoencoder with frequency-domain constraints and employ optical flow contrastive learning to enhance temporal consistency. Experiments demonstrate state-of-the-art performance in PSNR, SSIM, and perceptual quality metrics. Our method significantly improves anatomical fidelity, temporal continuity, and motion naturalness of interpolated frames while maintaining high reconstruction accuracy.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Deep Generative Models & Autoencoders
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The temporal interpolation task for 4D medical imaging, plays a crucial role in clinical practice of respiratory motion modeling. Following the simplified linear-motion hypothesis, existing approaches adopt optical flow-based models to interpolate intermediate frames. However, realistic respiratory motions should be nonlinear and quasi-periodic with specific frequencies. Intuited by this property, we resolve the temporal interpolation task from the frequency perspective, and propose a Fourier basis-guided Diffusion model, termed FB-Diff. Specifically, due to the regular motion discipline of respiration, physiological motion priors are introduced to describe general characteristics of temporal data distributions. Then a Fourier motion operator is elaborately devised to extract Fourier bases by incorporating physiological motion priors and case-specific spectral information in the feature space of Variational Autoencoder. Well-learned Fourier bases can better simulate respiratory motions with motion patterns of specific frequencies. Conditioned on starting and ending frames, the diffusion model further leverages well-learned Fourier bases via the basis interaction operator, which promotes the temporal interpolation task in a generative manner. Extensive results demonstrate that FB-Diff achieves state-of-the-art (SOTA) perceptual performance with better temporal consistency while maintaining promising reconstruction metrics. Codes are available.

Problem

• Interpolate nonlinear quasi-periodic 4D medical imaging
• Model respiratory motion using frequency-based Fourier bases
• Enhance temporal consistency in generative interpolation tasks

Innovation

• Fourier basis-guided diffusion for 4D interpolation
• Physiological motion priors enhance frequency modeling
• Basis interaction operator improves temporal consistency