RARF is a task‑agnostic, region‑aware rectified flow framework designed for 3D brain MRI inpainting. It limits stochastic interpolation to the inpainting region while keeping observed voxels fixed, using a 3D neural network that processes a partially voided image, Gaussian noise, a mask, and a timestep. The model is trained with masked flow‑matching and reconstruction‑consistency objectives, and during inference it transports noise toward a plausible reconstruction that preserves anatomical consistency, achieving competitive results on the BraTS Inpainting Challenge 2026.
By Tomas Guija-Valiente, Blanca Rodriguez-Gonzalez, Norberto Malpica, Angel Torrado-Carvajal
The study introduces ART‑Net, an anatomy‑aware residual attention network designed to refine four‑fold accelerated SENSE brain MRI. In a prospective paired study of 80 participants, ART‑Net achieved the highest peak signal‑to‑noise ratio and structural similarity index among evaluated methods, and it preserved anatomical fidelity with superior Dice coefficients for medial temporal and whole‑brain structures. Radiologist assessments also indicated improved gradient fidelity, regional contrast, and overall structural quality.
By Changjing Chai, Bin Huang, Libo Xu, Jian Zhou, Boyang Pan, Kristen W Yeom, Qiyong Gong, Nan-Jie Gong
arXiv:2610.09784v1 Announce Type: new
Abstract: Reliable liver segmentation in contrast-enhanced MRI is essential for quantitative hepatic assessment, treatment planning, and longitudinal disease mon...
By Ruoshi Xu, Mingqi Gao, Shengda Luo, Jingkun Chen
arXiv:2610.24858v2 Announce Type: replace
Abstract: Reducing MRI acquisition time improves patient comfort, reduces motion artefacts, and increases scanner throughput. Clinical protocols often shorte...
By Aditya Rastogi, Ramon Leiser, Prateek Singh, Martha Foltyn-Dumitru, Martin Bendszus, Fabian Bamberg, Klaus Maier-Hein, Kai Schlamp, Alexander Radbruch, Gianluca Brugnara, Philipp Vollmuth
arXiv:2603. 05693v2 Announce Type: replace-cross Abstract: Accurate longitudinal analysis of brain MRI is often hindered by evolving lesions, which bias automated neuroimaging pipelines.
By Zahra Karimaghaloo, Dumitru Fetco, Haz-Edine Assemlal, Hassan Rivaz, Douglas L. Arnold
arXiv:2609.36837v1 Announce Type: cross
Abstract: Generative super-resolution models can turn portable 64 mT MRI into images that look like 3T scans, and the field evaluates them with PSNR, SSIM, and...
By Prathamesh Pradeep Khole, Shreya Handa, Utkarsh Gupta, Razvan Marinescu
AGGR-Mamba is a lightweight output‑stage refinement network for undersampled brain MRI reconstruction that builds on a Mamba backbone. The method first generates an initial reconstruction and then applies an Attention Refine Head to recalibrate local anatomical structures, edge textures, and residual‑error regions, followed by a gated residual refinement branch that selectively preserves informative residual features while suppressing redundant responses. Experiments on the IXI brain MRI dataset with T1‑ and T2‑weighted images at acceleration factors R=4 and R=8 show that AGGR‑Mamba achieves competitive PSNR and SSIM performance across different reconstruction settings.
By Yue Zhuo, Wenbin Deng, Lingjun Liu, Xuhang Chen, Zhonghua Xie
arXiv:2610.08983v1 Announce Type: new
Abstract: Generative inpainting of brain MRI volumes is essential for synthesizing healthy tissue in pathological regions, improving the accuracy and reliability...
By Arnela Hadzic, Franz Thaler, Simon Johannes Joham, Martin Urschler
UniField++ is a two‑agent framework that enhances MRI images across multiple field‑strength transitions, such as 64 mT‑to‑3 T and 3 T‑to‑7 T. It combines a VLM‑based planner that predicts risk regions and a flow‑matching enhancer that performs loop reasoning to refine high‑risk areas, guided by radiologist‑annotated hallucination‑prone regions. The method achieves significant improvements in PSNR, SSIM, and whole‑brain segmentation performance compared to existing approaches.
By Yiyang Lin, Chenhui Wang, Tianliang Yao, Shuaibo Li, Zijun Sun, Yiming Deng, Yixuan Yuan
arXiv:2608. 06613v1 Announce Type: cross Abstract: Self-supervised 3D medical foundation models are increasingly used as general-purpose feature extractors, yet their sensitivity to MRI artifacts remains poorly understood.
By Julia Anna Mielcarz, Daniel Klaaby, Mostafa Mehdipour Ghazi
arXiv:2609.00960v1 Announce Type: new
Abstract: Magnetic resonance images of the same subject look markedly different across field strengths, which complicates the comparison and pooling of data acro...
By Baris Imre, Aram Salehi, Levente Baljer, Andrew Webb, Marius Staring, Efe Ilicak
FLAT: FLow-Aligned Training of Unrolled Networks for MRI Reconstruction proposes a new training method for unrolled neural networks used in MRI reconstruction. By interpreting unrolled networks as discretizations of conditional probability flows, the authors derive cascade parameters from Flow Matching and align intermediate reconstructions with the ideal Flow Matching trajectory. Experiments on three MRI datasets demonstrate that FLAT stabilizes the reconstruction trajectory across sub-networks and improves the final reconstruction quality.
By Kehan Qi, Saumya Gupta, Xiaoling Hu, Qingqiao Hu, Weimin Lyu, Yicun Wang, Chao Chen