Mar 2026· arXiv.org· Vol abs/2603.03710· 0 citations· 33 references
Computer Science
TL;DR
MPFlow is proposed, a zero-shot multi-modal reconstruction framework built on rectified flow that incorporates auxiliary MRI modalities at inference time without retraining the generative prior to improve anatomical fidelity.
Abstract
Zero-shot MRI reconstruction relies on generative priors, but single-modality unconditional priors produce hallucinations under severe ill-posedness. In many clinical workflows, complementary MRI acquisitions (e.g. high-quality structural scans) are routinely available, yet existing reconstruction methods lack mechanisms to leverage this additional information. We propose MPFlow, a zero-shot multi-modal reconstruction framework built on rectified flow that incorporates auxiliary MRI modalities at inference time without retraining the generative prior to improve anatomical fidelity. Cross-modal guidance is enabled by our proposed self-supervised pretraining strategy, Patch-level Multi-modal MR Image Pretraining (PAMRI), which learns shared representations across modalities. Sampling is jointly guided by data consistency and cross-modal feature alignment using pre-trained PAMRI, systematically suppressing intrinsic and extrinsic hallucinations. Extensive experiments on HCP and BraTS show that MPFlow matches diffusion baselines on image quality using only 20% of sampling steps while reducing tumor hallucinations by more than 15% (segmentation dice score). This demonstrates that cross-modal guidance enables more reliable and efficient zero-shot MRI reconstruction.
This approach optimizes the text-to-image diffusion priors via a rectified flow strategy and an MRI-tailored variational autoencoder, and further strengthens control over the restoration process using multimodal guidance, enabling high-fidelity, one-step reconstruction.
Jing-Wei Guan, Xing-Jian Tang, Lin-Ge Li et al.· Computerized Medical Imaging...· 0 citations
This work proposes a zero-shot brain MRI inpainting framework utilizing 2.5D unconditional flow priors to capture spatial context along the superior-inferior axis without the overhead of full 3D convolutions.
Arnela Hadzic, Franz Thaler, Simon Johannes Joham et al.· 0 citations
Objective. 3D quantitative magnetic resonance imaging (qMRI) enables noninvasive tissue characterization but often requires prolonged acquisition times for multi-contrast imaging, limiting its broader clinical adoption. Although deep learning has shown promise for accelerated MRI reconstruction, supervised methods rely...
Jing-Ran Xu, Seng Jia, Yuan-Biao Yang et al.· Physics in Medicine and Biol...· 0 citations
Experimental results across several CT inverse problem settings show that Flow Matching-based approaches consistently outperform diffusion-based methods in terms of PSNR, SSIM, and perceptual quality, while requiring fewer sampling steps.
The results suggest that anatomical priors are better understood as structural stabilizers under degraded modality information, rather than as universal full-modality accuracy boosters.
I2SB-Inversion is proposed, a multi-contrast guided reconstruction framework based on the Schrödinger Bridge that achieves a a high acceleration factor of R=11.38 and consistently outperforms existing methods in both quantitative and qualitative evaluations.
Yue Wang, Yuan-Biao Yang, Zhuo-xu Cui et al.· IEEE Transactions on Medical...· 0 citations
Known for his clear and elegant writing style, Bertsekas shaped fields from control and optimization to large-scale computation and artificial intelligence.