Berengere Aubert-Broche

A new improved version of the realistic digital brain phantom.

Image analysis methods must be tested and evaluated within a controlled environment. Simulations can be an extremely helpful tool for validation because ground truth is known. We created the digital brain phantom that is at the heart of our publicly available database of realistic simulated magnetic resonance image (MRI) volumes known as BrainWeb. Even though the digital phantom had l mm(3) isotropic voxel size and a small number of tissue classes, the BrainWeb database has been used in more than one hundred peer-reviewed publications validating different image processing methods. In this paper, we describe the next step in the natural evolution of BrainWeb: the creation of digital brain phantom II that includes three major improvements over the original phantom. First, the realism of the phantom, and the resulting simulations, was improved by modeling more tissue classes to include blood vessels, bone marrow and dura mater classes. In addition. a more realistic skull class was created. The latter is particularly useful for SPECT, PET and CT simulations for which bone attenuation has an important effect. Second, the phantom was improved by an eight-fold reduction in voxel volume to 0.125 mm(3). Third, the method used to create the new phantom was modified not only to take into account the segmentation of these new structures, but also to take advantage of many more automated procedures now available. The overall process has reduced subjectivity and manual intervention when compared to the original phantom, and the process may be easily applied to create phantoms from other subjects. MRI simulations are shown to illustrate the difference between the previous and the new improved digital brain phantom II. Example PET and SPECT simulations are also presented.

Twenty New Digital Brain Phantoms for Creation of Validation Image Data Bases

Simulations provide a way of generating data where ground truth is known, enabling quantitative testing of image processing methods. In this paper, we present the construction of 20 realistic digital brain phantoms that can be used to simulate medical imaging data. The phantoms are made from 20 normal adults to take into account intersubject anatomical variabilities. Each digital brain phantom was created by registering and averaging four T1, T2, and proton density (PD)-weighted magnetic resonance imaging (MRI) scans from each subject. A fuzzy minimum distance classification was used to classify voxel intensities from T1, T2, and PD average volumes into grey-matter, white matter, cerebro-spinal fluid, and fat. Automatically generated mask volumes were required to separate brain from nonbrain structures and ten fuzzy tissue volumes were created: grey matter, white matter, cerebro-spinal fluid, skull, marrow within the bone, dura, fat, tissue around the fat, muscles, and skin/muscles. A fuzzy vessel class was also obtained from the segmentation of the magnetic resonance angiography scan of the subject. These eleven fuzzy volumes that describe the spatial distribution of anatomical tissues define the digital phantom, where voxel intensity is proportional to the fraction of tissue within the voxel. These fuzzy volumes can be used to drive simulators for different modalities including MRI, PET, or SPECT. These phantoms were used to construct 20 simulated T1-weighted MR scans. To evaluate the realism of these simulations, we propose two approaches to compare them to real data acquired with the same acquisition parameters. The first approach consists of comparing the intensities within the segmented classes in both real and simulated data. In the second approach, a whole brain voxel-wise comparison between simulations and real T1-weighted data is performed. The first comparison underlines that segmented classes appear to properly represent the anatomy on average, and that inside these classes, the simulated and real intensity values are quite similar. The second comparison enables the study of the regional variations with no a priori class. The experiments demonstrate that these variations are small when real data are corrected for intensity nonuniformity

MRI correlates of cognitive impairment in childhood-onset multiple sclerosis.

OBJECTIVE Brain MRI measures were correlated with neuropsychological function in 35 pediatric-onset multiple sclerosis (MS) patients and 33 age- and sex-matched healthy controls. METHOD Mean age of MS patients was 16.3 ± 2.3 years with average disease duration of 4.3 ± 3.1 years. Cortical gray matter, thalamic, and global brain volumes were calculated for all participants using a scaling factor computed using normalization of atrophy method to normalize total and regional brain volumes for head size. T1- and T2-weighted lesion volumes were calculated for MS patients. RESULTS Cognitive impairment (CI) was identified in 29% of the MS cohort. Cognitive deficits predominantly involved attention and processing speed, expressive language, and visuomotor integration. Relative to controls, the MS group showed significantly lower thalamic volume (p < .001), total brain volume (p < .008), and gray matter volume (p < .015). Corpus callosum area and thalamic volume differentiated patients identified as having CI from those without CI (p < .05). Regression models controlling for disease duration and age indicated that thalamic volume accounted for significant incremental variance in predicting global IQ, processing speed, and expressive vocabulary (ΔR2 ranging from .43 to .60) and was the most robust MRI predictor of cognition relative to other MRI metrics. CONCLUSIONS The robust association between cognitive function and reduced size of thalamus and global brain volume in pediatric-onset MS patients implicate neurodegenerative processes early in the disease course, and suggest that plasticity of an immature central nervous system is not sufficient to protect patients from the deleterious consequences of MS on cognitive neural networks. (PsycINFO Database Record (c) 2011 APA, all rights reserved).

Regional brain atrophy in children with multiple sclerosis.

We used cross-sectional tensor-based morphometry to visualize reduced volume in the whole brains of pediatric patients with multiple sclerosis, relative to healthy controls. As a marker of local volume difference, we used the Jacobian determinant of the deformation field that maps each subject to a standard space. To properly assess abnormal differences in volume in this age group, it is necessary to account for the normal, age-related differences in brain volume. This was accomplished by computing normalized z-score Jacobian determinant values at each voxel to represent the local volume difference (in standard deviations) between an individual subject and an age- and sex-matched healthy normal population. Compared with healthy controls, pediatric patients with multiple sclerosis exhibited significantly reduced volumes within the thalamus and the splenium of the corpus callosum and significant expansions in the ventricles. While T2-weighted lesion volume was correlated with reduced splenium volume, no correlation was found between T2-weighted lesion volume and reduced thalamic volume. Reduced volumes of the optic pathways, including that of the optic tracts and optic radiations, correlated with disease duration. Our results suggest that focal inflammatory lesions may play an important role in tract degeneration, including transsynaptic degeneration.

Reduced head and brain size for age and disproportionately smaller thalami in child-onset MS

Objective: Whole brain and regional volume measurement methods were used to quantify white matter, gray matter, and deep gray matter structure volumes in a population of patients with pediatric-onset multiple sclerosis (MS). Methods: Subjects included 38 patients (mean age 15.2 ± 2.4 years) and 33 age- and sex-matched healthy control (HC) participants. MRI measures included intracranial volume, normalized brain volume, normalized white and gray matter volume, and volumes of the thalamus, globus pallidus, putamen, and caudate. Because these volumes vary across age and sex in children, we normalized the volume measurements for MS and control groups by computing z scores using normative values obtained from healthy children enrolled in the MRI Study of Normal Brain Development. Results: The intracranial volume z score was significantly lower in the patients with MS (−0.45 ± 1.16; mean ± SD) compared with the HC participants (+0.25 ± 0.98; p = 0.01). Patients with MS also demonstrated significant decreases in normalized brain volume z scores (−1.09 ± 1.49 vs −0.05 ± 1.22; p = 0.002). After correction for global brain volume, thalamic volumes in the MS population remained lower than those of HCs (−0.68 ± 1.72 vs 0.15 ± 1.35; p = 0.02), indicating an even greater loss of thalamic tissue relative to more global brain measures. Moderate correlations were found between T2-weighted lesion load and normalized thalamic volumes (r = −0.44, p < 0.01) and normalized brain volume (r = −0.47, p < 0.01) and between disease duration and normalized thalamic volume (r = −0.58, p < 0.001) and normalized brain volume (r = −0.46, p < 0.01). Conclusions: When compared with age- and sex-matched control subjects, the onset of MS during childhood is associated with a smaller overall head size, brain volume, and an even smaller thalamic volume.

Onset of multiple sclerosis before adulthood leads to failure of age-expected brain growth

Objective: To determine the impact of pediatric-onset multiple sclerosis (MS) on age-expected brain growth. Methods: Whole brain and regional volumes of 36 patients with relapsing-remitting MS onset prior to 18 years of age were segmented in 185 longitudinal MRI scans (2–11 scans per participant, 3-month to 2-year scan intervals). MRI scans of 25 age- and sex-matched healthy normal controls (NC) were also acquired at baseline and 2 years later on the same scanner as the MS group. A total of 874 scans from 339 participants from the NIH-funded MRI study of normal brain development acquired at 2-year intervals were used as an age-expected healthy growth reference. All data were analyzed with an automatic image processing pipeline to estimate the volume of brain and brain substructures. Mixed-effect models were built using age, sex, and group as fixed effects. Results: Significant group and age interactions were found with the adjusted models fitting brain volumes and normalized thalamus volumes (p < 10−4). These findings indicate a failure of age-normative brain growth for the MS group, and an even greater failure of thalamic growth. In patients with MS, T2 lesion volume correlated with a greater reduction in age-expected thalamic volume. To exclude any scanner-related influence on our data, we confirmed no significant interaction of group in the adjusted models between the NC and NIH MRI Study of Normal Brain Development groups. Conclusions: Our results provide evidence that the onset of MS during childhood and adolescence limits age-expected primary brain growth and leads to subsequent brain atrophy, implicating an early onset of the neurodegenerative aspect of MS.

Lower physical activity is associated with higher disease burden in pediatric multiple sclerosis

Objective: To evaluate the association between physical activity (PA) and multiple sclerosis (MS) disease activity, depression, and fatigue in a cohort of children with MS and monophasic acquired demyelinating syndrome (mono-ADS). Methods: In this cross-sectional study of consecutive patients attending a specialized pediatric MS clinic, we administered the PedsQL Multidimensional Fatigue Scale, Center for Epidemiological Studies Depression Scale, and Godin Leisure-Time Exercise Questionnaire. Quantitative MRI analysis was performed to obtain whole brain and T2 lesion volume in a subset of participants (n = 60). Results: A total of 110 patients (79 mono-ADS; 31 MS; 5–18 years; M:F 1:1.2) were included. Patients with MS reported less strenuous (33.21 ± 31.88 metabolic equivalents [METs] vs 15.97 ± 22.73 METs, p = 0.002) and total (44.48 ± 39.35 METs vs 67.28 ± 59.65 METs; p = 0.0291) PA than those with mono-ADS. Patients with MS who reported greater amounts of moderate PA METs had fewer sleep/rest fatigue symptoms (r = −0.4). Participation in strenuous PA was associated with smaller T2 lesion volumes (r = −0.66) and lower annualized relapse rate (r = −0.66). No associations were found between total brain volume and participation in PA. Conclusions: Children with MS are less physically active than children with mono-ADS. Reasons for this are unclear, but may be related to ongoing disease activity, perceived limitations, or symptoms such as depression or fatigue. Children with MS reporting higher levels of strenuous PA had lower T2 lesion volumes and lower relapse rates, suggesting a potential protective effect of strenuous PA in this population. Further longitudinal studies are needed to establish the relationship of PA to MS symptoms and disease activity in this population.

Detection of inter-hemispheric asymmetries of brain perfusion in SPECT

Technetium-99m HMPAO and technetium-99m ECD single photon emission computed tomography (SPECT) imaging is commonly used to highlight brain regions with altered perfusion. It is particularly useful in the investigation of intractable partial epilepsy. However, SPECT suffers from poor spatial resolution that makes interpretation difficult. In this context, we propose an unsupervised voxel neighbourhood based method to assist the detection of significant functional inter-hemispheric asymmetries in brain SPECT, using anatomical information from MRI. For each MRI voxel, the anatomically homologous voxel in the contralateral hemisphere is identified. Both homologous voxel coordinates are then mapped into the SPECT volume using SPECT-MRI registration. Neighbourhoods are then defined around each SPECT voxel and compared to obtain a volume of inter-hemispheric differences. A volume including only the statistically significant inter-hemispheric differences is deduced from this volume using a non-parametric approach. The method was validated using realistic analytical simulated SPECT data including known asymmetries (in size and amplitude) as ground truth (gold standard). Detection performance was assessed using an ROC (receiver operating characteristic) approach based on the measures of the overlap between known and detected asymmetries. Validation with computer-simulated data demonstrates the ability to detect asymmetric zones with relatively small extension and amplitude. The registration of these detected functional asymmetries on the MRI enables good anatomical localization to be achieved.

Human Brain Myelination from Birth to 4.5 Years

The myelination of white matter from birth through the first years of life has been studied qualitatively and it is well know the myelination occurs in a orderly and predictable manner, proceeding in a caudocranial direction, from deep to superficial and from posterior to anterior. Even if the myelination is a continuous process, it is useful to characterize myelination evolution in normal brain development in order to better study demyelinating diseases. The quantification of myelination has only been studied for neonates. The original contribution of this study is to develop a method to characterize and visualize the myelination pattern using MRI data from a group of normal subjects from birth to just over 4 years of age. The method includes brain extraction and tissue classification in addition to the analysis of T2 relaxation times to attempt to separate myelinated and unmyelinated white matter. The results agree previously published qualitative observations.

Clustering of atlas-defined cortical regions based on relaxation times and proton density.

Magnetic resonance parameters, such as longitudinal (T1) and transverse (T2) relaxation times and proton density (PD) provide intrinsic information about the human brain. In vivo quantification of these parameters may enable detection of subtle regional grey matter (GM) or white matter (WM) differences and permit neurological disease detection and monitoring. The aims of the study were to quantify T1, T2 and PD values in all cortical gray matter regions for a group of healthy volunteers scanned at 1.5 T and to cluster regions showing statistically distinguishable tissue characteristics. Using a combination of spoiled gradient recalled echo (SPGR) and fast spin echo (FSE) sequences, 3D T1, T2 and PD of the brain were measured at 1.5 T for twenty healthy young volunteers. Cortical GM volumes of interest (VOIs) were identified by transforming 56 labels from an atlas onto each subject volumes: 8 frontal, 5 parietal, 6 occipital and 9 temporal on both the left and right sides. T1, T2 and PD measurements within these anatomical regions were quantified and reported here. Correspondence analysis (CA) and hierarchical clustering (HC) were combined and applied to averaged T1, T2 and PD values within each VOI in order to identify groups of anatomical structures that are related statistically. Interestingly, except for one structure, all VOIs were grouped with left-right symmetry and showed an interesting pattern: the four lobes (frontal, occipital, parietal and temporal) were roughly clustered and the precentral and postcentral gyri were merged together. Our study shows that CA and HC analysis of MRI relaxation parameters and proton density can be used for cortical clustering of atlas-defined cortical regions.