Validating Nonlinear Registration to Improve Subtraction Images for Lesion Detection and Quantification in Multiple Sclerosis.

BACKGROUND AND PURPOSE: To propose and validate nonlinear registration techniques for generating subtraction images because of their ability to reduce artifacts and improve lesion detection and lesion volume quantification. METHODS: Postcontrast T1 -weighted spin echo and T2 -weighted dual echo images were acquired for 20 patients with relapsing-remitting multiple sclerosis (RRMS) on a monthly basis for a year (14 women, average age 33.6 ± 6.9). The T2 -weighted images from the first scan were used as a baseline for each patient. The images from the last scan were registered to the baseline image. Four different registration algorithms used for evaluation included; linear, halfway linear, nonlinear, and nonlinear halfway. Subtraction images were generated after brain extraction, intensity normalization, and Gaussian blurring. Lesion activity changes along with identified artifacts were scored on all four techniques by two independent observers. Additionally, quantitative analysis of the algorithms was performed by estimating the volume changes of simulated lesions and real lesions. For real lesion volume change analysis, five subjects were selected randomly. Subtraction images were generated between all the 11 time points and the baseline image using linear and nonlinear registration for the five subjects. RESULTS: Lesion activity detection resulted in similar performance among the four registration techniques. Lesion volume measurements on subtraction images using nonlinear registration were closer to lesion volume on T2 -weighted images. A statistically significant difference was observed among the four registration techniques while evaluating yin-yang artifacts. Pairwise comparisons showed that nonlinear registration results in the least amount of yin-yang artifacts, which are significantly different. CONCLUSIONS: Nonlinear registration for generation of subtraction images has been demonstrated to be a promising new technique as it shows improvement in lesion activity change detection. This approach decreases the number of artifacts in subtraction images. With improved lesion volume estimates and reduced artifacts, nonlinear registration may lead to discarding less subject data and an improvement in the statistical power of subtraction imaging studies.

Generalized min-max bound-based MRI pulse sequence design framework for wide-range T1 relaxometry: A case study on the tissue specific imaging sequence.

This paper proposes a new design strategy for optimizing MRI pulse sequences for T1 relaxometry. The design strategy optimizes the pulse sequence parameters to minimize the maximum variance of unbiased T1 estimates over a range of T1 values using the Cramér-Rao bound. In contrast to prior sequences optimized for a single nominal T1 value, the optimized sequence using our bound-based strategy achieves improved precision and accuracy for a broad range of T1 estimates within a clinically feasible scan time. The optimization combines the downhill simplex method with a simulated annealing process. To show the effectiveness of the proposed strategy, we optimize the tissue specific imaging (TSI) sequence. Preliminary Monte Carlo simulations demonstrate that the optimized TSI sequence yields improved precision and accuracy over the popular driven-equilibrium single-pulse observation of T1 (DESPOT1) approach for normal brain tissues (estimated T1 700-2000 ms at 3.0T). The relative mean estimation error (MSE) for T1 estimation is less than 1.7% using the optimized TSI sequence, as opposed to less than 7.0% using DESPOT1 for normal brain tissues. The optimized TSI sequence achieves good stability by keeping the MSE under 7.0% over larger T1 values corresponding to different lesion tissues and the cerebrospinal fluid (up to 5000 ms). The T1 estimation accuracy using the new pulse sequence also shows improvement, which is more pronounced in low SNR scenarios.

Heterogeneity of Multiple Sclerosis White Matter Lesions Detected With T2*-Weighted Imaging at 7.0 Tesla.

BACKGROUND AND PURPOSE: Postmortem studies in multiple sclerosis (MS) indicate that in some white matter lesions (WM-Ls), iron is detectable with T2*-weighted (T2*-w), and its reciprocal R2* relaxation rate, magnetic resonance imaging (MRI) at 7.0 Tesla (7T). This iron appears as a hyperintense rim in R2* images surrounding a hypointense core. We describe how this observation relates to clinical/radiological characteristics of patients, in vivo. METHODS: We imaged 16 MS patients using 3T and 7T scanners. WM-Ls were identified on T1-w / T2-w 3T-MRIs. Thereafter, WM-Ls with a rim of elevated R2* at 7T were counted and compared to their appearance on conventional MRIs. RESULTS: We counted 36 WM-Ls presenting a rim of elevated R2* in 10 patients. Twenty-three (64%) lesions coincided with focal WM-Ls on T2-w MRIs; 13 (36%) coincided with only portions of larger lesions on T2-w images; and 20 (56%) corresponded to a hypointense chronic black hole. WM-Ls presenting a rim of elevated R2* were seen in both relapsing-remitting patients with low disability and in those with long-standing secondary progressive MS. CONCLUSIONS: WM-Ls with a contour of high R2* are present at different MS stages, potentially representing differences in the contribution of iron in MS disease evolution.

Impact of chemotherapy for childhood leukemia on brain morphology and function.

OBJECTIVE: Using multidisciplinary treatment modalities the majority of children with cancer can be cured but we are increasingly faced with therapy-related toxicities. We studied brain morphology and neurocognitive functions in adolescent and young adult survivors of childhood acute, low and standard risk lymphoblastic leukemia (ALL), which was successfully treated with chemotherapy. We expected that intravenous and intrathecal chemotherapy administered in childhood will affect grey matter structures, including hippocampus and olfactory bulbs, areas where postnatal neurogenesis is ongoing. METHODS: We examined 27 ALL-survivors and 27 age-matched healthy controls, ages 15-22 years. ALL-survivors developed disease prior to their 11th birthday without central nervous system involvement, were treated with intrathecal and systemic chemotherapy and received no radiation. Volumes of grey, white matter and olfactory bulbs were measured on T1 and T2 magnetic resonance images manually, using FIRST (FMRIB's integrated Registration and Segmentation Tool) and voxel-based morphometry (VBM). Memory, executive functions, attention, intelligence and olfaction were assessed. RESULTS: Mean volumes of left hippocampus, amygdala, thalamus and nucleus accumbens were smaller in the ALL group. VBM analysis revealed significantly smaller volumes of the left calcarine gyrus, both lingual gyri and the left precuneus. DTI data analysis provided no evidence for white matter pathology. Lower scores in hippocampus-dependent memory were measured in ALL-subjects, while lower figural memory correlated with smaller hippocampal volumes. INTERPRETATION: Findings demonstrate that childhood ALL, treated with chemotherapy, is associated with smaller grey matter volumes of neocortical and subcortical grey matter and lower hippocampal memory performance in adolescence and adulthood.

Evolution of tumefactive lesions in multiple sclerosis: a 12-year study with serial imaging in a single patient.

We describe the acute presentation and the long-term evolution of recurrent tumefactive lesions (TLs) in a patient with relapsing-remitting multiple sclerosis. Five TLs occurred on three different occasions over a period of 12 years and these were followed by 73 serial magnetic resonance images (MRI). TL evolution was described by means of magnetization transfer imaging (MTI) and cerebrospinal fluid tissue specific imaging (TSI) over the follow-up period. During the study period, the patient had three clinical relapses with only minimal disability progression. MTI demonstrated that only the peripheral portion of each TL reverted to pre-lesional MT ratios within six months' post-enhancement. Recurring TLs may present a similar pattern of evolution that may be associated with a long-term favourable clinical outcome.

Cognitive impairment and its relation to imaging measures in multiple sclerosis: a study using a computerized battery.

BACKGROUND AND PURPOSE: Cognitive impairment (CI) is an important component of multiple sclerosis (MS) disability. A complex biological interplay between white matter (WM) and gray matter (GM) disease likely sustains CI. This study aims to address this issue by exploring the association between the extent of normal WM and GM disease and CI. METHODS: Cognitive function of 24 MS patients and 24 healthy volunteers (HVs) was studied using the Automated Neuropsychological Assessment Metrics (ANAM) battery. WM focal lesions and normal appearing WM (NAWM) volume in patients, cortical thickness (CTh) and deep GM structure volumes in both patients and HVs were measured by high field strength (3.0-Tesla; 3T) imaging. RESULTS: An analysis of covariance showed that patients performed worse than HVs on Code Substitution Delayed Memory (P = .04) and Procedural Reaction Time (P = .05) indicative of reduced performance in memory, cognitive flexibility, and processing speed. A summary score (Index of Cognitive Efficiency) indicating global test battery performance was also lower for the patient group (P = .04). Significant associations, as determined by the Spearman rank correlation tests, were noted between each of these 3 cognitive scores and measures of NAWM volume [CDD-TP1(r = .609; P = .0035), PRO-TP1 (r = .456; P = .029) and ICE (r = .489; P = .0129)], CTh (r = .5; P ≤ .05) and volume of subcortical normal appearing GM (NAGM) structures (r = .4; P≤ .04), but not WM lesions. CONCLUSIONS: Both NAWM and NAGM volumes are related to CI in MS. The results highlight once again the urgent need to develop pharmacological strategies protecting patients from widespread neurodegeneration as possible preventive strategies of CI development.

Lesions by tissue specific imaging characterize multiple sclerosis patients with more advanced disease.

BACKGROUND: Cerebrospinal fluid tissue specific imaging (CSF-TSI), a newly implemented magnetic resonance imaging (MRI) technique, allows visualization of a subset of chronic black holes (cBHs) with MRI characteristics suggestive of the presence of CSF-like fluid, and representing lesions with extensive tissue destruction. OBJECTIVE: To investigate the relationship between lesions in CSF-TSI and disease measures in patients with multiple sclerosis (MS). METHODS: Twenty-six patients with MS were imaged at 3.0 T, obtaining T(1)-weighted (T(1)-w) and T(2)-w spin echo (SE), T(1) volumetric images and CSF-TSI images. We measured: (i) lesion volume (LV) in T(1)-w (cBH-LV) and T(2)-w SE images, and in CSF-TSI; (ii) brain parenchyma fraction (BPF). Differences between patients with and without CSF-TSI lesions were analyzed and association between clinical and MRI metrics were investigated. RESULTS: cBHs were seen in 92% of the patients while lesions in CSF-TSI were seen in 40%. Patients with CSF-TSI lesions were older, with longer disease duration, higher disability scores, larger cBH-LV and T(2)-LV, and lower BPF than patients without CSF-TSI lesions (≤0.047). Partial correlation analysis correcting for T(2)-LV, cBH-LV and BPF showed an association (p < 0.0001, r = 0.753) between CSF-TSI LV and disability score. CONCLUSIONS: CSF-TSI lesions characterize patients with more advanced disease and probably contribute to the progress of disability.

Translocator protein PET imaging for glial activation in multiple sclerosis.

Glial activation in the setting of central nervous system inflammation is a key feature of the multiple sclerosis (MS) pathology. Monitoring glial activation in subjects with MS, therefore, has the potential to be informative with respect to disease activity. The translocator protein 18 kDa (TSPO) is a promising biomarker of glial activation that can be imaged by positron emission tomography (PET). To characterize the in vivo TSPO expression in MS, we analyzed brain PET scans in subjects with MS and healthy volunteers in an observational study using [(11)C]PBR28, a newly developed translocator protein-specific radioligand. The [(11)C]PBR28 PET showed altered compartmental distribution of TSPO in the MS brain compared to healthy volunteers (p = 0.019). Focal increases in [(11)C]PBR28 binding corresponded to areas of active inflammation as evidenced by significantly greater binding in regions of gadolinium contrast enhancement compared to contralateral normal-appearing white matter (p = 0.0039). Furthermore, increase in [(11)C]PBR28 binding preceded the appearance of contrast enhancement on magnetic resonance imaging in some lesions, suggesting a role for early glial activation in MS lesion formation. Global [(11)C]PBR28 binding showed correlation with disease duration (p = 0.041), but not with measures of clinical disability. These results further define TSPO as an informative marker of glial activation in MS.

Quality and quantity of diffuse and focal white matter disease and cognitive disability of patients with multiple sclerosis.

BACKGROUND AND PURPOSE: Using high-field magnetic resonance imaging (MRI), we investigated the relationships between white matter (WM) lesion volume (LV), normal-appearing WM (NAWM) normalized volume, WM-lesion and NAWM magnetization transfer ratios (MTRs), brain parenchyma fraction (BPF), and cognitive impairment (CI) in multiple sclerosis (MS). METHODS: Twenty-four patients and 24 healthy volunteers (age, sex, and years of education-matched) underwent a 3.0 Tesla (3T) scan and evaluation of depression, fatigue, and CI using the Minimal Assessment of Cognitive Function in MS (MACFIMS) battery. RESULTS: In this clinically relatively well-preserved cohort of patients (median score on the Expanded Disability Status Scale=1.5), CI was detected on Symbol Digit Modalities Test (SDMT), California Verbal Learning Test-II (CVLT-II), and Controlled Oral Word Association Test. MT data were available in 19 pairs on whom correlation analyses were performed. Associations were seen between SDMT and normalized NAWM volume (P=.034, r=.502), CVLT-II long delay and normalized NAWM volume (P=.012, r=.563), WM-LV (P=.024, r=.514), and BPF (P=.002, r=.666). CONCLUSIONS: The use of 3T MRI in a sample of clinically stable MS patients shows the importance of WM disease in hampering processing speed and word retrieval.

T1 cortical hypointensities and their association with cognitive disability in multiple sclerosis.

BACKGROUND: Neocortical lesions (NLs) largely contribute to the pathology of multiple sclerosis (MS), although their relevance in patients' disability remains unknown. OBJECTIVE: To assess the incidence of T(1) hypointense NLs by 3.0-Tesla magnetic resonance imaging (MRI) in patients with MS and examine neocortical lesion association with cognitive impairment. METHODS: In this case-control study, 21 MS patients and 21 age-, sex- and years of education-matched healthy volunteers underwent: (i) a neuropsychological examination rating cognitive impairment (Minimal Assessment of Cognitive Function in MS); (ii) a 3.0-Tesla MRI inclusive of an isotropic 1.0 mm(3) three-dimensional inversion prepared spoiled gradient-recalled-echo (3D-IRSPGR) image and T(1)- and T(2)-weighted images. Hypointensities on 3D-IRSPGR lying in the cortex, either entirely or partially were counted and association between NLs and cognitive impairment investigated. RESULTS: A total of 95 NLs were observed in 14 (66.7%) patients. NL+ patients performed poorer (p = 0.020) than NL-patients only on the delayed recall component of the California Verbal Learning Test. This difference lost statistical significance when a correction for white matter lesion volume was employed. CONCLUSIONS: Although T( 1) hypointense NLs may be present in a relatively high proportion of multiple sclerosis patients, the impact that they have in cognitive impairment is not independent from white matter disease.

Relationship of cortical atrophy to fatigue in patients with multiple sclerosis.

BACKGROUND: Fatigue is a common and disabling symptom of multiple sclerosis (MS). Previous studies reported that damage of the corticostriatothalamocortical circuit is critical in its occurrence. OBJECTIVE: To investigate the relationship between fatigue in MS and regional cortical and subcortical gray matter atrophy. DESIGN: Case-control study. SETTING: National Institutes of Health. PARTICIPANTS: Twenty-four patients with MS and 24 matched healthy volunteers who underwent 3.0-T magnetic resonance imaging and evaluations of fatigue (Modified Fatigue Impact Scale) and depression (Center for Epidemiologic Studies Depression Scale). MAIN OUTCOME MEASURES: Relationship between thalamic and basal ganglia volume, cortical thickness of frontal and parietal lobes, and, in patients, T2 lesion volume and normal-appearing white matter volume and the extent of fatigue. RESULTS: Patients were more fatigued than healthy volunteers (P = .04), while controlling for the effect of depression. Modified Fatigue Impact Scale score correlated with cortical thickness of the parietal lobe (r = -0.50, P = .01), explaining 25% of its variance. The posterior parietal cortex was the only parietal area significantly associated with the Modified Fatigue Impact Scale scores. CONCLUSIONS: Cortical atrophy of the parietal lobe had the strongest relationship with fatigue. Given the implications of the posterior parietal cortex in motor planning and integration of information from different sources, our preliminary results suggest that dysfunctions in higher-order aspects of motor control may have a role in determining fatigue in MS.

Mapping resting-state functional connectivity using perfusion MRI.

Resting-state, low-frequency (<0.08 Hz) fluctuations of blood oxygenation level-dependent (BOLD) magnetic resonance signal have been shown to exhibit high correlation among functionally connected regions. However, correlations of cerebral blood flow (CBF) fluctuations during the resting state have not been extensively studied. The main challenges of using arterial spin labeling perfusion magnetic resonance imaging to detect CBF fluctuations are low sensitivity, low temporal resolution, and contamination from BOLD. This work demonstrates CBF-based quantitative functional connectivity mapping by combining continuous arterial spin labeling (CASL) with a neck labeling coil and a multi-channel receiver coil to achieve high perfusion sensitivity. In order to reduce BOLD contamination, the CBF signal was extracted from the CASL signal time course by high frequency filtering. This processing strategy is compatible with sinc interpolation for reducing the timing mismatch between control and label images and has the flexibility of choosing an optimal filter cutoff frequency to minimize BOLD fluctuations. Most subjects studied showed high CBF correlation in bilateral sensorimotor areas with good suppression of BOLD contamination. Root-mean-square CBF fluctuation contributing to bilateral correlation was estimated to be 29+/-19% (N=13) of the baseline perfusion, while BOLD fluctuation was 0.26+/-0.14% of the mean intensity (at 3 T and 12.5 ms echo time).

An adaptive filter for suppression of cardiac and respiratory noise in MRI time series data.

The quality of MRI time series data, which allows the study of dynamic processes, is often affected by confounding sources of signal fluctuation, including the cardiac and respiratory cycle. An adaptive filter is described, reducing these signal fluctuations as long as they are repetitive and their timing is known. The filter, applied in image domain, does not require temporal oversampling of the artifact-related fluctuations. Performance is demonstrated for suppression of cardiac and respiratory artifacts in 10-minute brain scans on 6 normal volunteers. Experimental parameters resemble a typical fMRI experiment (17 slices; 1700 ms TR). A second dataset was acquired at a rate well above the Nyquist frequency for both cardiac and respiratory cycle (single slice; 100 ms TR), allowing identification of artifacts specific to the cardiac and respiratory cycles, aiding assessment of filtering performance. Results show significant reduction in temporal standard deviation (SD(t)) in all subjects. For all 6 datasets with 1700 ms TR combined, the filtering method resulted in an average reduction in SD(t) of 9.2% in 2046 voxels substantially affected by respiratory artifacts, and 12.5% for the 864 voxels containing substantial cardiac artifacts. The maximal SD(t) reduction achieved was 52.7% for respiratory and 55.3% for cardiac filtering. Performance was found to be at least equivalent to the previously published RETROICOR method. Furthermore, the interaction between the filter and fMRI activity detection was investigated using Monte Carlo simulations, demonstrating that filtering algorithms introduce a systematic error in the detected BOLD-related signal change if applied sequentially. It is demonstrated that this can be overcome by combining physiological artifact filtering and detection of BOLD-related signal changes simultaneously. Visual fMRI data from 6 volunteers were analyzed with and without the filter proposed here. Inclusion of the cardio-respiratory regressors in the design matrix yielded a 4.6% t-score increase and 4.0% increase in the number of significantly activated voxels.

Large-amplitude, spatially correlated fluctuations in BOLD fMRI signals during extended rest and early sleep stages.

A number of recent studies of human brain activity using blood-oxygen-level-dependent (BOLD) fMRI and EEG have reported the presence of spatiotemporal patterns of correlated activity in the absence of external stimuli. Although these patterns have been hypothesized to contain important information about brain architecture, little is known about their origin or about their relationship to active cognitive processes such as conscious awareness and monitoring of the environment. In this study, we have investigated the amplitude and spatiotemporal characteristics of resting-state activity patterns and their dependence on the subjects' alertness. For this purpose, BOLD fMRI was performed at 3.0 T on 12 normal subjects using a visual stimulation protocol, followed by a 27 min rest period, during which subjects were allowed to fall asleep. In subjects who were asleep at the end of the scan, we found (a) a higher amplitude of BOLD signal fluctuation during rest compared with subjects who were awake at the end of the scan; (b) spatially independent patterns of correlated activity that involve all of gray matter, including deep brain nuclei; (c) many patterns that were consistent across subjects; (d) that average percentage levels of fluctuation in visual cortex (VC) and whole brain were higher in subjects who were asleep (up to 1.71% and 1.16%, respectively) than in those who were awake (up to 1.15% and 0.96%) at the end of the scan and were comparable with those levels evoked by intense visual stimulation (up to 1.85% and 0.76% for two subject groups); (e) no confirmation of correlation, positive or negative, between thalamus and VC found in earlier studies. These findings suggest that resting-state activity continues during sleep and does not require active cognitive processes or conscious awareness.

Accelerated parallel imaging for functional imaging of the human brain.

Accelerated parallel imaging (PI) techniques have recently been applied to functional imaging experiments of the human brain in order to improve the performance of commonly used single-shot techniques like echo-planar imaging (EPI). Potential benefits of PI-fMRI include the reduction of geometrical distortions due to off-resonance signals, the reduction of signal-loss in areas with substantial signal inhomogeneity, increases of the spatial and temporal resolution of the fMRI experiment and reduction of gradient acoustic noise. Although PI generally leads to a substantial decrease in image signal-to-noise ratio (SNR), its effect on the temporal stability of the signal, which ultimately determines fMRI performance, is only partially determined by image SNR. Therefore, the penalty for using PI is generally not as severe as the SNR reduction. The majority of problems related to single-shot techniques become more severe at an increased magnetic field strength, making PI an important tool in achieving the full potential of fMRI at high field.

Clinical and imaging metrics for monitoring disease progression in patients with multiple sclerosis.

Multiple sclerosis (MS) is an autoimmune disease of the CNS leading to clinical disability in 250,000-350,000 young adults in the USA and Europe. The disease affects both white matter (WM) and gray matter (GM) tissues of the brain and spinal cord. While WM disease is easily quantified using currently available magnetic resonance imaging (MRI) techniques, identification and quantification of GM disease present a daily challenge. Nonconventional brain and spinal cord MRI techniques, including magnetization transfer, MRI spectroscopy and diffusion tensor imaging, have improved our understanding of MS pathology in the deep GM. The sensitivity of high-resolution MRI obtained at a high magnetic field will improve the detection of spinal cord and brain cortical GM disease. The appropriate use of the above-mentioned techniques has the potential to more accurately explain the level of disability in MS patients.

Optimizing brain tissue contrast with EPI: a simulated annealing approach.

A new magnetization preparation and image acquisition scheme was developed to obtain high-resolution brain images with optimal tissue contrast. The pulse sequence was derived from an optimization process using simulated annealing, without prior assumptions with regard to the number of radiofrequency (RF) pulses and flip angles. The resulting scheme combined two inversion pulses with the acquisition of three images with varying contrast. The combination of the three images allowed separation of gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) based on T1, contrast. It also enabled the correction of small errors in the initial T1 estimates in postprocessing. The use of three-dimensional (3D) sensitivity-encoded (SENSE) echo-planar imaging (EPI) for image acquisition made it possible to achieve a 1.15(3) mm3 isotropic resolution within a scan time of 10 min 21 s. The cortical GM signal-to-noise ratio (SNR) in the calculated GM-only image varied between 30 and 100. The novel technique was evaluated in combination with blood oxygen level-dependent (BOLD) functional magnetic resonance imaging (fMRI) on human subjects, and provided for excellent coregistration of anatomical and functional data.

Multirate SPAMM tagging.

Many magnetic resonance tagging sequences rely on periodicity in order to produce a uniform tagging grid that covers the whole image plane. This, however, is not always desirable, since motion may be restricted to specific parts of the image, and also different motion characteristics may call for different tagging grid densities. In this paper, we present a combination of the spatial modulation of magnetization 1-1 method with selective excitation pulses that can be used in order to restrict the tagging grid only to regions of interest and produce tagging grid of different density in each region. The method is fast and easy to implement even on older or less expensive systems, since it does not require extensive gradient switching.

A rotational approach to localized SPAMM 1-1 tagging.

Magnetic resonance tagging usually relies on controlling the phase dispersion of the transverse magnetization component. Phase dispersion is, however, affected by the inherent phase of selective excitation pulses, thus limiting their combination with tagging sequences to the application of refocusable pulses, as in the localized spatial modulation of magnetization (L-SPAMM) technique. In this study, we examine the effect of selective excitation pulses on a L-SPAMM 1-1 sequence, showing that in the case of two identical pulses the phase component is canceled out, and thus preemphasis and refocus gradients are not needed, allowing us to take advantage of a constant gradient throughout the tagging sequence, and also that one might choose nonrefocusable maximum and minimum phase pulses.