Zhao Yinghe, Yang Qinqin, Qian Shiting, Dong Jiyang, Cai Shuhui, Chen Zhong, Cai Congbo
Department of Electronic Science, Xiamen University, Xiamen 361005, China.
Brain Sci. 2024 Aug 17;14(8):828. doi: 10.3390/brainsci14080828.
(1) Background: Functional magnetic resonance imaging (fMRI) utilizing multi-echo gradient echo-planar imaging (ME-GE-EPI) has demonstrated higher sensitivity and stability compared to utilizing single-echo gradient echo-planar imaging (SE-GE-EPI). The direct derivation of * maps from fitting multi-echo data enables accurate recording of dynamic functional changes in the brain, exhibiting higher sensitivity than echo combination maps. However, the widely employed voxel-wise log-linear fitting is susceptible to inevitable noise accumulation during image acquisition. (2) Methods: This work introduced a synthetic data-driven deep learning (SD-DL) method to obtain * maps for multi-echo (ME) fMRI analysis. (3) Results: The experimental results showed the efficient enhancement of the temporal signal-to-noise ratio (tSNR), improved task-based blood oxygen level-dependent (BOLD) percentage signal change, and enhanced performance in multi-echo independent component analysis (MEICA) using the proposed method. (4) Conclusion: * maps derived from ME-fMRI data using the proposed SD-DL method exhibit enhanced BOLD sensitivity in comparison to * maps derived from the LLF method.
(1) 背景：与使用单回波梯度回波平面成像（SE-GE-EPI）相比，利用多回波梯度回波平面成像（ME-GE-EPI）的功能磁共振成像（fMRI）已显示出更高的灵敏度和稳定性。通过拟合多回波数据直接推导图能够准确记录大脑中的动态功能变化，比回波组合图具有更高的灵敏度。然而，广泛使用的逐体素对数线性拟合在图像采集过程中容易受到不可避免的噪声积累的影响。(2) 方法：这项工作引入了一种合成数据驱动的深度学习（SD-DL）方法来获取用于多回波（ME）fMRI分析的图。(3) 结果：实验结果表明，使用所提出的方法有效地提高了时间信噪比（tSNR），改善了基于任务的血氧水平依赖（BOLD）百分比信号变化，并提高了多回波独立成分分析（MEICA）的性能。(4) 结论：与使用LLF方法推导的图相比，使用所提出的SD-DL方法从ME-fMRI数据推导的图表现出更高的BOLD灵敏度。
