发表机构
Institute of Machine Learning in Biomedical Imaging, Helmholtz Munich; School of Computation, Information & Technology, Technical University of Munich; GE HealthCare; Biomedical Imaging Group Rotterdam, Erasmus MC; Visionlab, University of Antwerp; School of Medicine & Health, Technical University of Munich; Department of Neuroradiology, Klinikum rechts der Isar, Technical University of Munich; Munich Center for Machine Learning (MCML); School of Biomedical Engineering & Imaging Sciences, King’s College London（亥姆霍兹慕尼黑生物医学成像机器学习研究所; 慕尼黑工业大学计算、信息与技术学院; GE医疗; 伊拉斯姆斯大学医学中心鹿特丹生物医学成像组; 安特卫普大学视觉实验室; 慕尼黑工业大学医学与健康学院; 慕尼黑工业大学右伊萨尔医院神经放射科; 慕尼黑机器学习中心; 伦敦国王学院生物医学工程与成像科学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
CUPA-T2*框架通过协方差感知采样将重建不确定性传播到T2*拟合，在加速脑部MRI中提升了T2*拟合性能与不确定性对齐度，为定量T2*估计提供解释支持。
AI 中文摘要
定量T2* mapping在生物标志物发现方面具有巨大潜力，但受限于扫描时间过长，在临床场景中难以应用。通过k空间欠采样结合基于学习的重建可实现显著加速，但重建伪影和噪声会传播到下游的T2*拟合过程，降低其准确性。本文提出CUPA-T2*框架，该框架通过协方差感知采样，将随机蒙特卡洛 dropout重建的体素级回波间不确定性显式传播到下游T2*拟合。T2*拟合采用异方差MLP和基于相关性的正则化项，鼓励预测方差与重建不确定性之间的对齐。在加速脑部MRI数据上的实验显示出组织依赖性表现：CUPA-T2*实现了具有竞争力的整体T2*拟合性能，并在更高加速倍数下提升了白质性能。与异方差基线相比，所提框架大幅提高了重建不确定性与预测T2*方差之间的对齐度，同时揭示了校准（ECE）与选择性预测性能（AURC）之间存在权衡。CUPA-T2*支持感知重建不确定性的T2*拟合，并提供体素级不确定性图谱以辅助定量T2*估计的解释。
英文摘要
Quantitative T2* maps have strong potential for biomarker discovery but are limited by long scan times, rendering them impractical in clinical settings. Significant acceleration can be achieved through undersampling in k-space combined with learning-based reconstruction. However, reconstruction artifacts and noise can propagate into downstream T2* fitting, degrading its accuracy. We introduce CUPA-T2*, a framework that explicitly propagates voxel-wise inter-echo uncertainty from stochastic Monte Carlo dropout reconstructions to downstream T2* fitting via covariance-aware sampling. T2* fitting is performed with a heteroscedastic MLP and a correlation-based regularizer that encourages alignment between predicted variance and reconstruction uncertainty. Experiments on accelerated brain MRI data show tissue-dependent behavior: CUPA-T2* achieves competitive overall T2* fitting performance and improves white-matter performance at higher accelerations. Compared with a heteroscedastic baseline, the proposed framework substantially increases alignment between reconstruction uncertainty and predicted T2* variance, while also revealing a trade-off with calibration (ECE) and selective prediction performance (AURC). CUPA-T2* enables reconstruction uncertainty-aware T2* fitting and delivers voxel-wise uncertainty maps to support the interpretation of quantitative T2* estimates.
CommentsAccepted at the 2nd workshop on: Reconstruction and Imaging Motion Estimation (RIME) MICCAI. This is the submitted manuscript with added link to GitHub repo, funding acknowledgements, disclosure of interests, and authors' names and affiliations. No further post submission improvements or corrections were integrated
