TY - JOUR
T1 - Physiological contribution in spontaneous oscillations
T2 - An approximate quality-assurance index for resting-state fMRI signals
AU - Hsu, Ai Ling
AU - Chou, Kun Hsien
AU - Chao, Yi Ping
AU - Fan, Hsin Ya
AU - Wu, Chang-Wei
AU - Chen, Jyh Horng
N1 - Publisher Copyright:
© 2016 Hsu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
PY - 2016/2/1
Y1 - 2016/2/1
N2 - Resting-state fMRI (rs-fMRI) is receiving substantial attention for its sensitivity to functional abnormality in the brain networks of people with psychiatric and neurological disorders. However, because of the variety of rs-fMRI processing methods, the necessity of rs-fMRI quality assurance is increasing. Conventionally, the temporal signal-to-noise ratio (tSNR) is generally adopted for quality examination, but the tSNR does not guarantee reliable functional connectivity (FC) outcomes. Theoretically, intrinsic FC is supposed to reflect the spontaneous synchronization of neuronal basis, rather than that from thermal noise or nonneuronal physiological noise. Therefore, we proposed a new quality-assurance index for rsfMRI to estimate the physiological contributions in spontaneous oscillations (PICSO). The PICSO index was designed as a voxel-wise measure for facilitating practical applications to all existing rs-fMRI data sets on the basis of two assumptions: Gaussian distributions in temporal fluctuations and ultra-slow changes of neural-based physiological fluctuations. To thoroughly validate the sensitivity of the proposed PICSO index to FC, we calibrated the preprocessing steps according to phantom data and verified the relationship between the PICSO and factors that are considered to affect FC in healthy participants (n = 12). Our results demonstrated that FC showed a significantly positive correlation with the PICSO. Moreover, for generating robust FC outcomes, directly acquiring data at a relatively large voxel size was more effective than performing smoothness on high-resolution data sets. In conclusion, compared with tSNR, the PICSO index is more sensitive to the resulting FC, providing a practical quality-assurance indicator for all existing rs-fMRI data sets.
AB - Resting-state fMRI (rs-fMRI) is receiving substantial attention for its sensitivity to functional abnormality in the brain networks of people with psychiatric and neurological disorders. However, because of the variety of rs-fMRI processing methods, the necessity of rs-fMRI quality assurance is increasing. Conventionally, the temporal signal-to-noise ratio (tSNR) is generally adopted for quality examination, but the tSNR does not guarantee reliable functional connectivity (FC) outcomes. Theoretically, intrinsic FC is supposed to reflect the spontaneous synchronization of neuronal basis, rather than that from thermal noise or nonneuronal physiological noise. Therefore, we proposed a new quality-assurance index for rsfMRI to estimate the physiological contributions in spontaneous oscillations (PICSO). The PICSO index was designed as a voxel-wise measure for facilitating practical applications to all existing rs-fMRI data sets on the basis of two assumptions: Gaussian distributions in temporal fluctuations and ultra-slow changes of neural-based physiological fluctuations. To thoroughly validate the sensitivity of the proposed PICSO index to FC, we calibrated the preprocessing steps according to phantom data and verified the relationship between the PICSO and factors that are considered to affect FC in healthy participants (n = 12). Our results demonstrated that FC showed a significantly positive correlation with the PICSO. Moreover, for generating robust FC outcomes, directly acquiring data at a relatively large voxel size was more effective than performing smoothness on high-resolution data sets. In conclusion, compared with tSNR, the PICSO index is more sensitive to the resulting FC, providing a practical quality-assurance indicator for all existing rs-fMRI data sets.
UR - http://www.scopus.com/inward/record.url?scp=84960539408&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84960539408&partnerID=8YFLogxK
U2 - 10.1371/journal.pone.0148393
DO - 10.1371/journal.pone.0148393
M3 - Article
AN - SCOPUS:84960539408
SN - 1932-6203
VL - 11
JO - PLoS One
JF - PLoS One
IS - 2
M1 - e0148393
ER -