Enhancing White Matter fMRI Reliability through BOLD-like Signals Using Multi-Echo Techniques
Poster No:
2325
Submission Type:
Abstract Submission
Authors:
Ziyang Chen1,2, Li-Xia Yuan2, Bingchen Shao2, Qiqi Tong1, YiCheng Hsu3, Hongjian He2
Institutions:
1Research Center for Healthcare Data Science, Zhejiang Lab, Hangzhou, China, 2MIAO Lab, School of Physics, Zhejiang University, Hangzhou, China, 3MR Research Collaboration Team, Siemens Healthineers Ltd., Shanghai, China
First Author:
Ziyang Chen
Research Center for Healthcare Data Science, Zhejiang Lab|MIAO Lab, School of Physics, Zhejiang University
Hangzhou, China|Hangzhou, China
Research Center for Healthcare Data Science, Zhejiang Lab|MIAO Lab, School of Physics, Zhejiang University
Hangzhou, China|Hangzhou, China
Co-Author(s):
Introduction:
White matter functional magnetic resonance imaging (fMRI) has emerged for investigating the functional role of the brain's white matter (Ji et al. 2017, Schilling et al. 2023). Recent study has shown that the amplitude of low-frequency fluctuations (ALFF) of WM is a potential biomarker in psychiatric disorders (Ji, et al. 2023). However, signals in the white matter are more susceptible to contamination by physiological noise due to their significantly lower amplitude. Multi-echo technology offers the opportunity to separate non-T2* signals, providing enhanced BOLD signal and suppressed non-BOLD noise (Kundu et al. 2017), and allowing the measured fMRI information to more accurately reflect information related to true BOLD effect. These advantages have been proven to enhance the reliability of fMRI results in gray matter regions. Inspired by this, the study will explore whether multi-echo fMRI can enhance the reliability of resting-state functional measurements in white matter regions. The objective is to shed light on the potential benefits of using multi-echo fMRI for white matter function studies.
Methods:
Twenty-seven healthy subjects participated in the experiment and underwent MRI scanning included ME-fMRI, SE-fMRI and T1w MPRAGE on a Siemens MAGNETOM Prisma 3T scanner. Both SE-fMRI and ME-fMRI were acquired with following parameters: TR of 1300ms, isotropic resolution of 3mm and a total of 700 frames. The ME-fMRI were acquired using a four-echo EPI sequence with TEs=12.6/30.86/49.13/67.38 ms, while the SE-fMRI were acquired with TE=30ms. All the fMRI data was firstly splitted into two clips and then underwent slice time correction and motion correction. For the ME-fMRI data, optimal combination was applied to merge the four echoes into a single echo data(Kundu, Brenowitz et al. 2013). Nuisance signal, including CSF signal, global signal and ICA noise signal (ICA-AROAM for SE and ME-ICA for ME), were regressed out after filtering. All subject-level WM masks were combined to create a group level WM mask in MNI space. ALFF values were calculated in native bold space and then mapped to MNI space. The ALFF maps in MNI space were masked by group-level WM mask and subsequently z-transformed to obtain zALFF maps, which was used to calculate intraclass correlation coefficient (ICC).
Results:
ME-fMRI demonstrates a comparable group-averaged zALFF pattern (coefficient=0.89) when compared to SE-fMRI. The zALFF from SE-fMRI shows poor reliability (ICC = 0.28±0.19), whereas ME-fMRI exhibits moderate reliability (ICC = 0.41±0.2). A higher percentage of white matter regions in ME-fMRI (17.7%) surpass the substantial ICC threshold (ICC=0.6), indicating greater reliability, compared to SE-fMRI where only 5.7% of the regions achieve this threshold. Besides, the improved reliability in ME-fMRI primarily contributes to decreased within-subject variance (Vw) and slightly increased between-subject variance (Vb) compared to SE-fMRI. For the ROI-based analysis, we selected 17 regions of interest (ROIs) from the JHU white matter atlas, which have good overlap with our group-level white matter mask. All 17 ROIs show a significant improvement in ICC ranging from 0.08 to 0.22, and all these ROIs show a significant decrease in Vw. However, only 11 out of 17 ROIs showed a significant increase in Vb, with the genu of the corpus radiate showing a significant decrease with a mean value of 0.1 of Vb. The cingulum cingulate gyrus which plays an import role in different cognitive and emotional processes, benefits the most compared to SE-fMRI, with a significant mean increment of 0.22.
Conclusions:
Our findings indicate that ALFF maps in white matter, exhibit greater reliability when using BOLD-like signals from ME-fMR. The enhancement in reliability is primarily attributed to a significant reduction in Vw. The improved reliability of white matter measurements with ME-fMRI could potentially facilitate a more accurate understanding of the underlying mechanisms of white matter functions.
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
White Matter Anatomy, Fiber Pathways and Connectivity 2
Novel Imaging Acquisition Methods:
BOLD fMRI 1
Keywords:
Acquisition
FUNCTIONAL MRI
White Matter
1|2Indicates the priority used for review
Provide references using author date format
Schilling, K. G., Li, M., Rheault, F., Gao, Y., Cai, L., Zhao, Y., Xu, L., Ding, Z., Anderson, A. W., Landman, B. A., & Gore, J. C. (2023). Whole-brain, gray, and white matter time-locked functional signal changes with simple tasks and model-free analysis. Proceedings of the National Academy of Sciences of the United States of America, 120(42), e2219666120.
Ji, G. J., W. Liao, F. F. Chen, L. Zhang and K. Wang (2017). "Low-frequency blood oxygen level-dependent fluctuations in the brain white matter: more than just noise." Science Bulletin 62(9): 656-657.
Kundu, P., N. D. Brenowitz, V. Voon, Y. Worbe, P. E. Vértes, S. J. Inati, Z. S. Saad, P. A. Bandettini and E. T. Bullmore (2013). "Integrated strategy for improving functional connectivity mapping using multiecho fMRI." Proceedings of the National Academy of Sciences of the United States of America 110(40): 16187-16192.
Kundu, P., V. Voon, P. Balchandani, M. V. Lombardo, B. A. Poser and P. A. Bandettini (2017). "Multi-echo fMRI: A review of applications in fMRI denoising and analysis of BOLD signals." Neuroimage 154: 59-80.