Applying Independent Component Analysis to Clinical fMRI at 7 T

Abstract

Increased BOLD sensitivity at 7 T offers the possibility to increase the reliability of fMRI, but ultra-high field is also associated with an increase in artifacts related to head motion, Nyquist ghosting, and parallel imaging reconstruction errors. In this study, the ability of independent component analysis (ICA) to separate activation from these artifacts was assessed in a 7 T study of neurological patients performing chin and hand motor tasks. ICA was able to isolate primary motor activation with negligible contamination by motion effects. The results of General Linear Model (GLM) analysis of these data were, in contrast, heavily contaminated by motion. Secondary motor areas, basal ganglia, and thalamus involvement were apparent in ICA results, but there was low capability to isolate activation in the same brain regions in the GLM analysis, indicating that ICA was more sensitive as well as more specific. A method was developed to simplify the assessment of the large number of independent components. Task-related activation components could be automatically identified via these intuitive and effective features. These findings demonstrate that ICA is a practical and sensitive analysis approach in high field fMRI studies, particularly where motion is evoked. Promising applications of ICA in clinical fMRI include presurgical planning and the study of pathologies affecting subcortical brain areas.

A Hitchhiker's Guide to Functional Magnetic Resonance Imaging

“…Ghosting artifacts, which occur only in the phase-encoding direction, are triggered because odd and even lines of k-space are acquired with opposite polarity. Techniques such as implementing a multi-echo reference scan, two-dimensional phase correction or applying dual-polarity generalized autocalibrating partially parallel acquisitions (GRAPPA), can reduce the magnitude of these effects (Schmithorst et al, 2001; Chen and Wyrwicz, 2004; Robinson et al, 2013; Hoge and Polimeni, 2015). Hardware-related artifacts such as scanner and head coil heterogeneities, spiking, chemical shifts, and radiofrequency (RF) interferences all can significantly impact the fMRI image quality and compromise results (Bernstein et al, 2006; Poldrack et al, 2011).…”

Robust presurgical functional MRI at 7 T using response consistency

“…Model‐free analysis methods represent an attractive solution, as they do not require assumptions about task timing, the shape of the HRF, or linearity in the response. Independent Component Analysis (ICA) [Beckmann, ], for instance, has been shown to have considerable advantages over the GLM in the context of presurgical planning at ultra‐high field (UHF) [Robinson et al, ]. This method has not been widely embraced in the clinical context due to the need to assess and interpret the large number of components (ICs) generated.…”

ICA of fMRI studies: new approaches and cutting edge applications

“…

on resting-state connectivity and Balsters et al (2013) assess the correlation between BOLD spectral power and working memory performance.The ICA applications featured in this Research Topic range from clinical resting-state studies with patients suffering from schizophrenia (Manoliu et al, 2013;Sui et al, 2013) and neurological patients performing chin and hand motor tasks (Robinson et al, 2013) to the investigation of processing streams using chemosensory stimuli (Frasnelli et al, 2012). Combined methodological approaches are used to study belief decision making with fMRI and EEG (Douglas et al, 2013), to discriminate schizophrenia using data from fMRI, DTI, and sMRI (Sui et al, 2013), to identify amyotrophic lateral sclerosis diseased brains (Welsh et al, 2013) and to examine the microvascular specificity of the BOLD effect at 3 and 7 T using SWI (Geissler et al, 2013).

We hope this collection of original research articles illustrates the extent to which ICA is becoming an increasingly flexible and potent analysis method -particularly through innovations such as real-time ICA, temporal ICA, and parallel processing implementations -and that the capacity of ICA to isolate the underlying signal sources in fMRI data is being enhanced by multimodal and ultra-fast imaging.

…”

“…Combined methodological approaches are used to study belief decision making with fMRI and EEG (Douglas et al, 2013), to discriminate schizophrenia using data from fMRI, DTI, and sMRI (Sui et al, 2013), to identify amyotrophic lateral sclerosis diseased brains (Welsh et al, 2013) and to examine the microvascular specificity of the BOLD effect at 3 and 7 T using SWI (Geissler et al, 2013).We hope this collection of original research articles illustrates the extent to which ICA is becoming an increasingly flexible and potent analysis method -particularly through innovations such as real-time ICA, temporal ICA, and parallel processing implementations -and that the capacity of ICA to isolate the underlying signal sources in fMRI data is being enhanced by multimodal and ultra-fast imaging. These innovations are leading to an increase in the utility of ICA and the richness of information it can provide in both basic research work and clinical applications.…”