arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2606.28787cs.CVcs.AI
收藏

BREIT:一种基于多频三维EIT的脑卒中重建框架

BREIT: A Framework for Brain Stroke Reconstruction using Multi-Frequency 3D EIT

ETIS (UMR 8051), CY Cergy Paris University, ENSEA, CNRS(ETIS(UMR 8051),CY塞尔吉-巴黎大学,ENSEA,法国国家科学研究中心);AGM (UMR 8088), CY Cergy Paris University, CNRS(AGM(UMR 8088),CY塞尔吉-巴黎大学,法国国家科学研究中心);Paris-Saclay University, CentraleSupélec, Inria, CVN(巴黎-萨克雷大学,中央理工-高等电力学院,法国国家信息与自动化研究所,CVN)

Djahid Abdelmoumene, Ishak Ayad, Maï K. Nguyen, Christian Daveau

AI总结:

提出BREIT框架,包含神经影像到EIT的转换、3D完整电极模型正问题求解器和3D D-bar实现,并基于此提出dFNO-bar方法,通过傅里叶神经算子学习散射数据到电导率的映射,在合成数据上取得更高脑SSIM。

摘要:

多频电阻抗断层成像(MF-EIT)是一种非侵入性、低成本的模态,可从边界电压重建电特性分布。对于脑卒中成像,3D深度学习重建的进展受到缺乏带有配对真实体积的大规模数据集以及数据生成、模拟和评估流程非标准化的限制。我们介绍了BREIT,一个用于3D MF-EIT脑卒中重建的模块化框架,提供:(i) 一个神经影像到EIT的流程,将CT/MRI转换为频率依赖的真实导纳体积;(ii) 一个自包含的Python 3D完整电极模型(CEM)正问题求解器,用于模拟MF-EIT电压;以及(iii) 一个支持非均匀电极布局的3D D-bar实现。基于BREIT,我们提出了dFNO-bar,它通过从散射数据$t(\xi)$到电导率$\sigma(x){=}\Re\{\gamma\}$的映射学习,将傅里叶神经算子集成到D-bar中。我们在UCLH匹配的合成数据上评估了dFNO-bar与D-bar、Deep D-bar和高斯-牛顿重建,观察到在噪声设置下具有更高的脑SSIM和相当的CC。

英文摘要:

Multi-Frequency Electrical Impedance Tomography (MF-EIT) is a non-invasive, low-cost modality that reconstructs electrical property distributions from boundary voltages. For stroke imaging, progress in 3D deep-learning reconstruction is limited by the lack of large-scale datasets with paired ground-truth (GT) volumes and by non-standardized pipelines for data generation, simulation, and evaluation. We introduce BREIT, a modular framework for 3D MF-EIT stroke reconstruction providing: (i) a neuroimaging-to-EIT pipeline that converts CT/MRI into frequency-dependent GT admittivity volumes; (ii) a self-contained Python 3D Complete Electrode Model (CEM) forward solver for simulating MF-EIT voltages; and (iii) a 3D D-bar implementation supporting non-uniform electrode layouts. Building on BREIT, we propose dFNO-bar, which integrates Fourier Neural Operators into D-bar by learning a mapping from scattering data $t(ξ)$ to conductivity $σ(x){=}\Re\{γ\}$. We evaluate dFNO-bar against D-bar, Deep D-bar, and Gauss--Newton reconstructions on UCLH-matched synthetic data, and observe higher brain SSIM with comparable CC across noise settings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑