Removal Of Gradient Artifacts Research Articles

Direct modelling of gradient artifacts for EEG-fMRI denoising and motion tracking

Objective. Simultaneous electroencephalography and functional magnetic resonance imaging recording (EEG-fMRI) has been widely used in neuroscientific and clinical research. The artifacts in the recorded EEG resulting from rapidly switching magnetic field gradients are usually corrected by average-artifact subtraction (AAS) due to their repetitive nature. But the performance of AAS is often disrupted by altered artifact waveforms across epochs, notably due to head motion. Approach. Here, a method is proposed to make use of the known MR sequence gradient waveforms for a direct modelling of gradient artifacts. After accounting for filtering effects on the gradient artifacts, a continuous modulation of the gradient waveforms superimposed on the EEG signal is obtained. Main results. Although a moving AAS template can adjust to slow drifts in gradient artifact variation, it fails to adapt to abrupt motion, resulting in residual noise. We demonstrate how this modelling approach can reduce motion-affected gradient artifacts without distorting the underlying neuronal signals. Moreover, the method provides useful head motion information highly correlated with motion tracked by an optical camera. Significance. Our work provides a novel way to improve gradient artifact removal in EEG-fMRI, and shows a potential to detect head motion without requiring additional hardware.

WERYFIKACJA ZMIENNOŚCI MIĘDZYOSOBNICZEJ AKTYWNOŚCI WZGÓRZA PODCZAS GENERACJI RYTMU BERGERA (RYTMU ALFA)

BackgroundThe aim of the present work is to investigate the relationship between spontaneous electroencephalographic (EEG) brain activity at 8–13 Hz frequency (Berger’s rhythm) and thalamus activation. The leading theory of how Berger’s rhythm is generated suggests a thalamo–occipital circuit, but there is still much uncertainty as to the role of the thalamus.Material and MethodsWe used a Siemens Magnetom 3T Trio scanner and a 64-channel NeuroScan SynAmps2 EEG system to examine 36 healthy young male adults. The study paradigm consisted of 30-s blocks with eyes closed alternated with 30-s blocks with eyes open, both repeated six times. The EEG data was preprocessed as follows: 1) fMRI gradient artifact removal; 2) BCG reduction; 3) 1–20 Hz band-pass filtration. Alpha rhythm segments were marked in the preprocessed data. fMRI data was preprocessed using typical procedures (motion correction, normalization, smoothing), and then a general linear model (GLM) analysis was performed using alpha segments derived from the EEG as events. A modified hemodynamic response function suitable for examining thalamus physiology was applied.ResultsEEG data produced a typical spatial distribution of the alpha rhythm, mostly elicited in the parieto–occipital electrodes. Group-level analysis of the fMRI data failed to reveal any activation in the thalamus. However, further investigation revealed three subgroups of patients: 1) those who had a signal decrease in the left medial dorsal nucleus; 2) those with positive activation of thalamic structures; and 3) those where no activation was detected in the thalamus.ConclusionsThe thalamus might be involved differently in alpha rhythm generation from one subject to another. The observed intersubject variability might be caused by physiological mechanisms underlying Berger’s rhythm.

Independent Vector Analysis for Gradient Artifact Removal in Concurrent EEG-fMRI Data

We consider the problem of removing gradient artifact from electroencephalogram (EEG) signal, recorded concurrently with functional magnetic resonance imaging (fMRI) acquisition. We estimate the artifact by exploiting its quasi-periodicity over the epochs and its similarity over the different channels by using independent vector analysis, a recent extension of independent component analysis for multiple datasets. The method fully makes use of the spatio-temporal information by using spatial dependences across channels to estimate the artifact for a particular channel. Thus, it provides robustness with respect to uncontrollable changes such as head movement and fluctuations in the B0 field during the acquisition. Results using both simulated data with gradient artifact and EEG data collected concurrently with fMRI show the desirable performance of the new method.

Simultaneously recorded EEG–fMRI: Removal of gradient artifacts by subtraction of head movement related average artifact waveforms

Electroencephalograms (EEGs) recorded simultaneously with functional magnetic resonance imaging (fMRI) are corrupted by large repetitive artifacts generated by the switched MR gradients. Several methods have been proposed to remove these distortions by subtraction of averaged artifact templates from the ongoing EEG. Here, we present a modification of this approach which accounts for head movements to improve the extracted template. Using the fMRI analysis package statistical parametric mapping (SPM; FIL London) the head displacement is determined at each half fMRI-volume. The basic idea is to apply a moving average algorithm for template extraction but to include only epochs that were obtained at the same head position as the artefact to be removed. This approach was derived from phantom EEG measurements demonstrating substantial variations of the artefact waveform in response to movements of the phantom in the MRI magnet. To further reduce the residual noise, we applied a resampling algorithm which aligns the EEG samples in a strict adaptive manner to the fMRI timing. Finally, we propose a new algorithm to suppress residual artifacts such as those occasionally observed in case of brief strong movements, which are not reflected by the movement indicator because of the limited temporal resolution of the fMRI sequence. On the basis of EEG recordings of six subjects these measures combined reduce the residual artefact activity quantified in terms of the spectral power at the gradient repetition rate and its harmonics by roughly 20 to 50% (depending on the amount of movement) predominantly in frequencies beyond 30 Hz.

Evaluating gradient artifact correction of EEG data acquired simultaneously with fMRI

Simultaneous electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) has become a widely used application in spite of EEG perturbations due to electromagnetic interference in the MR environment. The most prominent and disturbing artifacts in the EEG are caused by the alternating magnetic fields (gradients) of the MR scanner. Different methods for gradient artifact correction have been developed. Here we propose an approach for the systematic evaluation and comparison of these gradient artifact correction methods. Exemplarily, we evaluate different algorithms all based on artifact template subtraction--the currently most established means of gradient artifact removal. We introduce indices for the degree of gradient artifact reduction and physiological signal preservation. The combination of both indices was used as a measure for the overall performance of gradient artifact removal and was shown to be useful in identifying problems during artifact removal. We demonstrate that the evaluation as proposed here allows to reveal frequency-band specific performance differences among the algorithms. This emphasizes the importance of carefully selecting the artifact correction method appropriate for the respective case.

Simultaneous EEG-fMRI acquisition: benefit and limitation

The combined investigation with EEG and fMRI makes use of the advantages of both methods, high spatial (fMRI) and time (EEG) resolution. The feasibility of simultaneous EEG recordings during acquisition of EPI-scanning has been demonstrated. However, EEG obtained simultaneously with fMRI is highly distorted by technical and biological artifacts. Artifacts include MR-gradient switching artifacts and artifacts caused by cardioballistic movements of the head. Although several artifact removal approaches coexist and partly are implemented in commercially available MR-compatible EEG-systems, methods are currently still under development. Recent work has shown that the extraction of both event-related potentials and frequency composition from reconstructed EEG matches those from EEG recorded outside an MR-environment. Multimodal analysis of brain processes including EEG and fMRI is beginning to be developed as well. Frequently, EEG-parameters are introduced as regressors in a GLM-analysis of the fMRI-data. This method allows the identification of brain regions, where EEG and hemodynamic response are in correlation. FMRI analysis benefits from this multimodal approach, because localized hemodynamic brain activation can be related to specific aspects of the EEG, thus facilitating the interpretation of the functional specificity of brain regions. EEG analysis can benefit from fMRI-analysis by anatomically constraining source reconstruction of EEG voltage distribution. The limitations depend on the quality of EEG reconstruction after application of artifact reduction procedures. Currently, the usable bandwidth is limited to frequencies below the gamma band and above the delta band. The high frequency limitation reflects limitations in gradient artifact removal, the low frequency limitation is due to interference of EEG signal with cardioballistic artifact. In conclusion, multimodal imaging using simultaneously recorded EEG and fMRI has the potential to advance the investigation of brain function. Limiting factors are related to the quality of EEG-signal reconstruction. However, recent developments indicate that these limitations will not play a role in the future.