Gradient induced artifacts in simultaneous EEG-fMRI: Effect of synchronization on spiral and EPI k-space trajectories

The nature of the gradient induced electroencephalography (EEG) artifact is analyzed and compared for two functional magnetic resonance imaging (fMRI) pulse sequences with different k-space trajectories: echo planar imaging (EPI) and spiral. Furthermore, the performance of the average artifact subtraction algorithm (AAS) to remove the gradient artifact for both sequences is evaluated. The results show that the EEG gradient artifact for spiral sequences is one order of magnitude higher than for EPI sequences due to the chirping spectrum of the spiral sequence and the dB/dt of its crusher gradients. However, in the presence of accurate synchronization, the use of AAS yields the same artifact suppression efficiency for both pulse sequences below 80 Hz. The quality of EEG signal after AAS is demonstrated for phantom and human data. EEG spectrogram and visual evoked potential (VEP) are compared outside the scanner and use both EPI and spiral pulse sequences. MR related artifact residues affect the spectra over 40 Hz (less than 0.2 μV up to 120 Hz) and modify the amplitude of P1, N2 and P300 in the VEP. These modifications in the EEG signal have to be taken into account when interpreting EEG data acquired in simultaneous EEG-fMRI experiments.     

DC artifact correction for arbitrary phase-cycling sequence

In magnetic resonance imaging (MRI), a non-zero offset in the receiver baseline signal during acquisition results in a bright spot or a line artifact in the center of the image known as a direct current (DC) artifact. Several methods have been suggested in the past for the removal or correction of DC artifacts in MR images, however, these methods cannot be applied directly when a specific phase-cycling technique is used in the imaging sequence. In this work, we proposed a new, simple technique that enables correction of DC artifacts for any arbitrary phase-cycling imaging sequences. The technique is composed of phase unification, DC offset estimation and correction, and phase restoration. The feasibility of the proposed method was demonstrated via phantom and in vivo experiments with a multiple phase-cycling balanced steady-state free precession (bSSFP) imaging sequence. Results showed successful removal of the DC artifacts in images acquired using bSSFP with phase-cycling angles of 0°, 90°, 180°, and 270°, indicating potential feasibility of the proposed method to any imaging sequence with arbitrary phase-cycling angles.     

An independent component analysis-based approach on ballistocardiogram artifact removing

Interest about simultaneous electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) data acquisition has rapidly increased during the last years because of the possibility that the combined method offers to join temporal and spatial resolution, providing in this way a powerful tool to investigate spontaneous and evoked brain activities. However, several intrinsic features of MRI scanning become sources of artifacts on EEG data. Noise sources of a highly predictable nature such as those related to the pulse MRI sequence and those determined by magnetic gradient switching during scanning do not represent a major problem and can be easily removed. On the contrary, the ballistocardiogram (BCG) artifact, a large signal visible on all EEG traces and related to cardiac activity inside the magnetic field, is determined by sources that are not fully stereotyped and causing important limitations in the use of artifact-removing strategies. Recently, it has been proposed to use independent component analysis (ICA) to remove BCG artifact from EEG signals. ICA is a statistical algorithm that allows blind separation of statistically independent sources when the only available information is represented by their linear combination. An important drawback with most ICA algorithms is that they exhibit a stochastic behavior: each run yields slightly different results such that the reliability of the estimated sources is difficult to assess. In this preliminary report, we present a method based on running the FastICA algorithm many times with slightly different initial conditions. Clustering structure in the signal space of the obtained components provides us with a new way to assess the reliability of the estimated sources.     

Simultaneous EEG and fMRI in the macaque monkey at 4.7 Tesla

Simultaneous electroencephalography (EEG)/functional magnetic resonance imaging (fMRI) acquisition can identify the brain networks involved in generating specific EEG patterns. Yet, the combination of these methodologies is hampered by strong artifacts that arise due to electromagnetic interference during magnetic resonance (MR) image acquisition. Here, we report corrections of the gradient-induced artifact in phantom measurements and in experiments with an awake behaving macaque monkey during fMRI acquisition at a magnetic field strength of 4.7 T. Ninety-one percent of the amplitude of a 10 microV, 10 Hz phantom signal could successfully be recovered without phase distortions. Using this method, we were able to extract the monkey EEG from scalp recordings obtained during MR image acquisition. Visual evoked potentials could also be reliably identified. In conclusion, simultaneous EEG/fMRI acquisition is feasible in the macaque monkey preparation at 4.7 T and holds promise for investigating the neural processes that give rise to particular EEG patterns.     

Phase gradient imaging for positive contrast generation to superparamagnetic iron oxide nanoparticle-labeled targets in magnetic resonance imaging

Positive contrast imaging methods produce enhanced signal at large magnetic field gradient in magnetic resonance imaging. Several postprocessing algorithms, such as susceptibility gradient mapping and phase gradient mapping methods, have been applied for positive contrast generation to detect the cells targeted by superparamagnetic iron oxide nanoparticles. In the phase gradient mapping methods, smoothness condition has to be satisfied to keep the phase gradient unwrapped. Moreover, there has been no discussion about the truncation artifact associated with the algorithm of differentiation that is performed in k-space by the multiplication with frequency value. In this work, phase gradient methods are discussed by considering the wrapping problem when the smoothness condition is not satisfied. A region-growing unwrapping algorithm is used in the phase gradient image to solve the problem. In order to reduce the truncation artifact, a cosine function is multiplied in the k-space to eliminate the abrupt change at the boundaries. Simulation, phantom and in vivo experimental results demonstrate that the modified phase gradient mapping methods may produce improved positive contrast effects by reducing truncation or wrapping artifacts.     

Crossing fibers, diffractions and nonhomogeneous magnetic field: correction of artifacts by bipolar gradient pulses

In recent years, diffusion tensor imaging (DTI) and its variants have been used to describe fiber orientations and q-space diffusion MR was proposed as a means to obtain structural information on a micron scale. Therefore, there is an increasing need for complex phantoms with predictable microcharacteristics to challenge different indices extracted from the different diffusion MR techniques used. The present study examines the effect of diffusion pulse sequence on the signal decay and diffraction patterns observed in q-space diffusion MR performed on micron-scale phantoms of different geometries and homogeneities. We evaluated the effect of the pulse gradient stimulated-echo, the longitudinal eddy current delay (LED) and the bipolar LED (BPLED) pulse sequences. Interestingly, in the less homogeneous samples, the expected diffraction patterns were observed only when diffusion was measured with the BPLED sequence. We demonstrated the correction ability of bipolar diffusion gradients and showed that more accurate physical parameters are obtained when such a diffusion gradient scheme is used. These results suggest that bipolar gradient pulses may result in more accurate data if incorporated into conventional diffusion-weighted imaging and DTI.     

Field camera versus phantom-based measurement of the gradient system transfer function (GSTF) with dwell time compensation

PurposeThe gradient system transfer function (GSTF) can be used to describe the dynamic gradient system and applied for trajectory correction in non-Cartesian MRI. This study compares the field camera and the phantom-based methods to measure the GSTF and implements a compensation for the difference in measurement dwell time.MethodsThe self-term GSTFs of a MR system were determined with two approaches: 1) using a dynamic field camera and 2) using a spherical phantom-based measurement with standard MR hardware. The phantom-based GSTF was convolved with a box function to compensate for the dwell time dependence of the measurement. The field camera and phantom-based GSTFs were used for trajectory prediction during retrospective image reconstruction of 3D wave-CAIPI phantom images.ResultsDifferences in the GSTF magnitude response were observed between the two measurement methods. For the wave-CAIPI sequence, this led to deviations in the GSTF predicted trajectories of 4% compared to measured trajectories, and residual distortions in the reconstructed phantom images generated with the phantom-based GSTF. Following dwell-time compensation, deviations in the GSTF magnitudes, GSTF-predicted trajectories, and resulting image artifacts were eliminated (< 0.5% deviation in trajectories).ConclusionWith dwell time compensation, both the field camera and the phantom-based GSTF self-terms show negligible deviations and lead to strong artifact reduction when they are used for trajectory correction in image reconstruction.     

MR measurement technique of rapidly switched gradient magnetic fields in MR tomography

Gradient eddy currents, induced in the surrounding conductive structures in a magnetic resonance (MR) magnet, are a major problem in MR imaging, in localized MR spectroscopy and in many other MR experiments. We present a comparison of three methods of measuring the gradient time characteristics and the time changes of basic magnetic fieldB ₀ after the gradient is switched off. The methods are based on the selective excitation of a thin layer of the sample and on acquiring the MR signal obtained after the end of the gradient pulse and on the computation of the instantaneous frequency of the signal. At this point, the time gradient characteristic is proportional to the instantaneous frequency of the MR signal, which has a small signal-to-noise ratio. We use the characteristics measured to set the pre-emphasis parameters in a 200 MHz/200 mm MR scanner. From the results obtained by measurement it follows that all methods are convenient for simple and quick characterization of magnetic field gradient in MR tomographic magnets.     

Characteristics of geometric distortion correction with increasing field-of-view in open-configuration MRI

Open-configuration magnetic resonance imaging (MRI) systems are becoming increasingly desirable for musculoskeletal imaging and image-guided radiotherapy because of their non-claustrophobic configuration. However, geometric image distortion in large fields-of-view (FOV) due to field inhomogeneity and gradient nonlinearity hinders the practical applications of open-type MRI. We demonstrated the use of geometric distortion correction for increasing FOV in open MRI. Geometric distortion was modeled and corrected as a global polynomial function. The appropriate polynomial order was identified as the minimum difference between the coordinates of control points in the distorted MR image space and those predicted by polynomial modeling. The sixth order polynomial function was found to give the optimal value for geometric distortion correction. The area of maximum distortion was < 1 pixel with an FOV of 285 mm. The correction performance error was increased at most 1.2% and 2.9% for FOVs of 340 mm and ~ 400 mm compared with the FOV of 285 mm. In particular, unresolved distortion was generated by local deformation near the gradient coil center.     

The use of view angle tilting to reduce distortions in magnetic resonance imaging of cryosurgery

Susceptibility artifacts from magnetic resonance (MR)-compatible cryoprobes can distort MR images of iceballs. In this work, we investigate the ability of view angle tilting (VAT) to correct susceptibility induced distortions in MR images of cryosurgery. The efficacy of VAT was tested in an ex vivo bovine liver model of cryosurgery using MR-compatible cryoprobes. Artifacts on high bandwidth fast spin echo images of freezing obtained with and without VAT were compared with photographs of the actual iceball shape and size. In vivo imaging with VAT was demonstrated during percutaneous MR-guided cryosurgery of pig liver and brain. VAT was most successful in reducing probe and iceball distortions when the imaging plane was normal to the cryoprobe, and the cryoprobe was perpendicular to the main magnetic field of the scanner. VAT had the greatest benefit when used to correct MR images of freezing when the surface of the iceball was relatively near to the cryoprobe. For large iceballs, the artifact was small so the VAT correction was less important. We conclude that VAT significantly reduced distortions in the shape of the signal void corresponding to the extent of freezing visualized during MR-guided cryosurgery. This improved ability to visualize the exact location of the cryoprobe, as well as the precise shape of the iceball, particularly during initial freezing when the iceball is small, has potential to significantly improve the accuracy of MR-guided cryosurgery of small lesions, and the accuracy of MR-assisted temperature calculations that are based on precise imaging of the probe location, and boundary geometry of the iceball.     

An empirical investigation of motion effects in eMRI of interictal epileptiform spikes

We recently developed a functional neuroimaging technique called encephalographic magnetic resonance imaging (eMRI). Our method acquires rapid single-shot gradient-echo echo-planar MRI (repetition time=47 ms); it attempts to measure an MR signal more directly linked to neuronal electromagnetic activity than existing methods. To increase the likelihood of detecting such an MR signal, we recorded concurrent MRI and scalp electroencephalography (EEG) during fast (20-200 ms), localized, high-amplitude (>50 μV on EEG) cortical discharges in a cohort of focal epilepsy patients. Seen on EEG as interictal spikes, these discharges occur in between seizures and induced easily detectable MR magnitude and phase changes concurrent with the spikes with a lag of milliseconds to tens of milliseconds. Due to the time scale of the responses, localized changes in blood flow or hemoglobin oxygenation are unlikely to cause the MR signal changes that we observed. While the precise underlying mechanisms are unclear, in this study, we empirically investigate one potentially important confounding variable — motion. Head motion in the scanner affects both EEG and MR recording. It can produce brief “spike-like” artifacts on EEG and induce large MR signal changes similar to our interictal spike-related signal changes. In order to explore the possibility that interictal spikes were associated with head motions (although such an association had never been reported), we had previously tracked head position in epilepsy patients during interictal spikes and explicitly demonstrated a lack of associated head motion. However, that study was performed outside the MR scanner, and the root-mean-square error in the head position measurement was 0.7 mm. The large inaccuracy in this measurement therefore did not definitively rule out motion as a possible signal generator. In this study, we instructed healthy subjects to make deliberate brief (<500 ms) head motions inside the MR scanner and imaged these head motions with concurrent EEG and MRI. We compared these artifactual MR and EEG data to genuine interictal spikes. While per-voxel MR and per-electrode EEG time courses for the motion case can mimic the corresponding time courses associated with a genuine interictal spike, head motion can be unambiguously differentiated from interictal spikes via scalp EEG potential maps. Motion induces widespread changes in scalp potential, whereas interictal spikes are localized and have a regional fall-off in amplitude. These findings make bulk head motion an unlikely generator of the large spike-related MR signal changes that we had observed. Further work is required to precisely identify the underlying mechanisms.     

基于自适应扩散模型的单帧红外条纹非均匀性校正算法

针对红外焦平面成像系统存在列向条纹非均匀性的现象,采用了一种基于自适应PM扩散模型的非均匀校正新算法。首先,综合利用图像梯度信息和局部灰度统计信息,自适应计算PM模型的扩散阈值;然后将每列像素的PM模型估计值作为该列像素的期望值;最后采用最陡下降法迭代计算得到每列像元的校正参数,并对结果进行循环校正以提高校正效果。实验结果表明:该算法可以保护图像边缘信息,与同类算法相比,能够更有效地抑制条纹非均匀性,并且能够防止图像产生鬼影。     

Special designed RF-antenna with integrated non-invasive carbon electrodes for simultaneous magnetic resonance imaging and electroencephalography acquisition at 7T

The construction of a high quality MR RF-antenna with incorporated EEG electrodes for simultaneous MRI and EEG acquisition is presented. The antenna comprises an active decoupled surface coil for receiving the MR signal and a whole body coil for transmitting the excitation RF pulses. The surface coil offers a high signal-to-noise ratio required for fMRI application and the whole body coil has a good B(1) excitation profile, which enables the application of homogeneous RF pulses. Non-invasive carbon electrodes are used in order to minimise the magnetic susceptibility artefacts that occur upon application of conductive materials. This dedicated set-up is compared to a standard set-up being a linear birdcage coil and commercially available Ag/AgCl electrodes. As the acquired EEG signals are heavily disturbed by the gradient switching, intelligent filtering is applied to obtain a clean EEG signal.     

A new perceptual difference model for diagnostically relevant quantitative image quality evaluation: A preliminary study

Purpose

Most objective image quality metrics average over a wide range of image degradations. However, human clinicians demonstrate bias toward different types of artifacts. Here, we aim to create a perceptual difference model based on Case-PDM that mimics the preference of human observers toward different artifacts.

Method

We measured artifact disturbance to observers and calibrated the novel perceptual difference model (PDM). To tune the new model, which we call Artifact-PDM, degradations were synthetically added to three healthy brain MR data sets. Four types of artifacts (noise, blur, aliasing or “oil painting” which shows up as flattened, over-smoothened regions) of standard compressed sensing (CS) reconstruction, within a reasonable range of artifact severity, as measured by both PDM and visual inspection, were considered. After the model parameters were tuned by each synthetic image, we used a functional measurement theory pair-comparison experiment to measure the disturbance of each artifact to human observers and determine the weights of each artifact's PDM score. To validate Artifact-PDM, human ratings obtained from a Double Stimulus Continuous Quality Scale experiment were compared to the model for noise, blur, aliasing, oil painting and overall qualities using a large set of CS-reconstructed MR images of varying quality. Finally, we used this new approach to compare CS to GRAPPA, a parallel MRI reconstruction algorithm.

Results

We found that, for the same Artifact-PDM score, the human observer found incoherent aliasing to be the most disturbing and noise the least. Artifact-PDM results were highly correlated to human observers in both experiments. Optimized CS reconstruction quality compared favorably to GRAPPA's for the same sampling ratio.

Conclusions

We conclude our novel metric can faithfully represent human observer artifact evaluation and can be useful in evaluating CS and GRAPPA reconstruction algorithms, especially in studying artifact trade-offs.     

Automated correction of unwanted phase jumps in reference signals which corrupt MRSI spectra after eddy current correction

A commonly applied step in the postprocessing of gradient localized proton MR spectroscopy, is correction for eddy current effects using the water signal as a reference. However, this method can degrade some of the metabolite signals, in particular if applied on proton MR spectroscopic imaging data. This artifact arises from the water reference signal in the presence of a second signal which resonates close to the main water resonance. The interference of both resonances will introduce jumps in the phase of the reference time domain signal. Using this phase for eddy current correction will result in a ringing artifact in the frequency domain of the metabolite signal over the whole frequency range. We propose a moving window correction algorithm, which screens the phase of reference signals and removes phase jumps in time domain caused by interference of signals from multiple spin systems. The phase jumps may be abrupt or gradually distributed over several time data points. Because the correction algorithm only corrects time data points which contain phase jumps, the phase is minimally disrupted. Furthermore, the algorithm is automated for large datasets, correcting only those water reference signals which are corrupted. After correction of the corrupted reference signals, normal eddy current correction may be performed. The algorithm is compared with a method which uses a low-pass filter and tested on simulated data as well as on in vivo proton spectroscopic imaging data from a healthy volunteer and from patients with a brain tumor.     

Automatic segmentation of cerebral white matter hyperintensities using only 3D FLAIR images

Magnetic Resonance (MR) white matter hyperintensities have been shown to predict an increased risk of developing cognitive decline. However, their actual role in the conversion to dementia is still not fully understood. Automatic segmentation methods can help in the screening and monitoring of Mild Cognitive Impairment patients who take part in large population-based studies. Most existing segmentation approaches use multimodal MR images. However, multiple acquisitions represent a limitation in terms of both patient comfort and computational complexity of the algorithms. In this work, we propose an automatic lesion segmentation method that uses only three-dimensional fluid-attenuation inversion recovery (FLAIR) images. We use a modified context-sensitive Gaussian mixture model to determine voxel class probabilities, followed by correction of FLAIR artifacts. We evaluate the method against the manual segmentation performed by an experienced neuroradiologist and compare the results with other unimodal segmentation approaches. Finally, we apply our method to the segmentation of multiple sclerosis lesions by using a publicly available benchmark dataset. Results show a similar performance to other state-of-the-art multimodal methods, as well as to the human rater.     

Choosing Strategies to Deal with Artifactual EEG Data in Children with Cognitive Impairment

Rett syndrome is a disease that involves acute cognitive impairment and, consequently, a complex and varied symptomatology. This study evaluates the EEG signals of twenty-nine patients and classify them according to the level of movement artifact. The main goal is to achieve an artifact rejection strategy that performs well in all signals, regardless of the artifact level. Two different methods have been studied: one based on the data distribution and the other based on the energy function, with entropy as its main component. The method based on the data distribution shows poor performance with signals containing high amplitude outliers. On the contrary, the method based on the energy function is more robust to outliers. As it does not depend on the data distribution, it is not affected by artifactual events. A double rejection strategy has been chosen, first on a motion signal (accelerometer or EEG low-pass filtered between 1 and 10 Hz) and then on the EEG signal. The results showed a higher performance when working combining both artifact rejection methods. The energy-based method, to isolate motion artifacts, and the data-distribution-based method, to eliminate the remaining lower amplitude artifacts were used. In conclusion, a new method that proves to be robust for all types of signals is designed.     

Methods for quantitative image quality evaluation of MRI parallel reconstructions: detection and perceptual difference model

Many reconstruction algorithms are being proposed for parallel magnetic resonance imaging (MRI), which uses multiple coils and subsampled k-space data, and a quantitative method for comparison of algorithms is sorely needed. On such images, we compared three methods for quantitative image quality evaluation: human detection, computer detection model and a computer perceptual difference model (PDM). One-quarter sampling and three different reconstruction methods were investigated: a regularization method developed by Ying et al., a simplified regularization method and an iterative method proposed by Pruessmann et al. Images obtained from a full complement of k-space data were also included as reference images. Detection studies were performed using a simulated dark tumor added on MR images of fresh bovine liver. Human detection depended strongly on reconstruction methods used, with the two regularization methods achieving better performance than the iterative method. Images were also evaluated using detection by a channelized Hotelling observer model and by PDM scores. Both predicted the same trends as observed from human detection. We are encouraged that PDM gives trends similar to that for human detection studies. Its ease of use and applicability to a variety of MRI situations make it attractive for evaluating image quality in a variety of MR studies.     

An anisotropic images segmentation and bias correction method

Intensity inhomogeneities cause considerable difficulty in the quantitative analysis of magnetic resonance (MR) images. Thus, bias field correction is a necessary step before quantitative analysis of MR data can be undertaken. This paper presents an anisotropic approach to bias correction and segmentation for images with intensity inhomogeneities and noise. Intensity-based methods are usually applied to estimate the bias field; however, most of them only concern the intensity information. When the images have noise or slender topological objects, these methods cannot obtain accurate results or bias fields. We use structure information to construct an anisotropic Gibbs field and combine the anisotropic Gibbs field with the Bayesian framework to segment images while estimating the bias fields. Our method is able to capture bias of quite general profiles. Moreover, it is robust to noise and slender topological objects. The proposed method has been used for images of various modalities with promising results.     

Incorporating prior knowledge via volumetric deep residual network to optimize the reconstruction of sparsely sampled MRI

For sparse sampling that accelerates magnetic resonance (MR) image acquisition, non-linear reconstruction algorithms have been developed, which incorporated patient specific a prior information. More generic a prior information could be acquired via deep learning and utilized for image reconstruction. In this study, we developed a volumetric hierarchical deep residual convolutional neural network, referred to as T-Net, to provide a data-driven end-to-end mapping from sparsely sampled MR images to fully sampled MR images, where cartilage MR images were acquired using an Ultra-short TE sequence and retrospectively undersampled using pseudo-random Cartesian and radial acquisition schemes. The network had a hierarchical architecture that promoted the sparsity of feature maps and increased the receptive field, which were valuable for signal synthesis and artifact suppression. Relatively dense local connections and global shortcuts were established to facilitate residual learning and compensate for details lost in hierarchical processing. Additionally, volumetric processing was adopted to fully exploit spatial continuity in three-dimensional space. Data consistency was further enforced. The network was trained with 336 three-dimensional images (each consisting of 32 slices) and tested by 24 images. The incorporation of a priori information acquired via deep learning facilitated high acceleration factors (as high as 8) while maintaining high image fidelity (quantitatively evaluated using the structural similarity index measurement). The proposed T-Net had an improved performance as compared to several state-of-the-art networks.