ICA-based artefact removal and accelerated fMRI acquisition for improved resting state network imaging

Abstract

The identification of resting state networks (RSNs) and the quantification of their functional connectivity in resting-state fMRI (rfMRI) are seriously hindered by the presence of artefacts, many of which overlap spatially or spectrally with RSNs. Moreover, recent developments in fMRI acquisition yield data with higher spatial and temporal resolutions, but may increase artefacts both spatially and/or temporally. Hence the correct identification and removal of non-neural fluctuations is crucial, especially in accelerated acquisitions. In this paper we investigate the effectiveness of three data-driven cleaning procedures, compare standard against higher (spatial and temporal) resolution accelerated fMRI acquisitions, and investigate the combined effect of different acquisitions and different cleanup approaches. We applied single-subject independent component analysis (ICA), followed by automatic component classification with FMRIB’s ICA-based X-noiseifier (FIX) to identify artefactual components. We then compared two first-level (within-subject) cleaning approaches for removing those artefacts and motion-related fluctuations from the data. The effectiveness of the cleaning procedures were assessed using timeseries (amplitude and spectra), network matrix and spatial map analyses. For timeseries and network analyses we also tested the effect of a second-level cleaning (informed by group-level analysis). Comparing these approaches, the preferable balance between noise removal and signal loss was achieved by regressing out of the data the full space of motion-related fluctuations and only the unique variance of the artefactual ICA components. Using similar analyses, we also investigated the effects of different cleaning approaches on data from different acquisition sequences. With the optimal cleaning procedures, functional connectivity results from accelerated data were statistically comparable or significantly better than the standard (unaccelerated) acquisition, and, crucially, with higher spatial and temporal resolution. Moreover, we were able to perform higher dimensionality ICA decompositions with the accelerated data, which is very valuable for detailed network analyses.

Altered physiological brain variation in drug‐resistant epilepsy

“…The datasets were rigorously corrected with conventional FSL preprocessing and with advanced ICA‐based FIX that further de‐noises the data for motion and other noise sources (Griffanti et al., 2014; Salimi‐Khorshidi et al., 2014). Without FIX, there were no differences between the HC and first measurement of epilepsy.…”

“…MREG images were aligned to three‐dimensional (MPRAGE) anatomical images in MNI152 standard space (full‐search, 12 DOF) in 4‐mm resolution. Advanced ICA FIX method was used for secondary artifact removal from the preprocessed MREG data (Griffanti et al., 2014; Salimi‐Khorshidi et al., 2014). FIX method was trained on previously acquired control data and was used for both groups because of our null hypothesis.…”

Long‐term acclimatization to high‐altitude hypoxia modifies interhemispheric functional and structural connectivity in the adult brain

“…Finally, the time series of the structured noises (based on the unique variance related to the noise components and motion confounds from the preprocessed data sets) were regressed out of the original data (Griffanti et al. 2014). The noise‐removed functional images were spatially transformed to a common stereotaxic coordination for subsequent analysis.…”

Correlation between resting state fMRI total neuronal activity and PET metabolism in healthy controls and patients with disorders of consciousness

“…On the other hand, our technique is the very first one using a multivariate approach as ICA to build a single scalar map, taking full advantage of ICA as a recognized tool for the separation of artifacts in the BOLD signal (Griffanti et al. 2014). If our approach, like many others, might be appealing for centers that have no access to FDG‐PET, our main interest in this methodology is that it allows comparison of resting state fMRI in two different populations without the necessity of recognizing the different networks after IC selection (see (Demertzi et al.…”

“…Characterization of the neuronal contribution of the resting state signal is one of the most relevant, yet challenging, research question (Griffanti et al. 2014). Removal of the artifactual contribution to the BOLD signal represents a critical step for any resting state based analysis.…”