Automated segmentation of changes in FLAIR-hyperintense white matter lesions in multiple sclerosis on serial magnetic resonance imaging.

Longitudinal analysis of white matter lesion changes on serial MRI has become an important parameter to study diseases with white-matter lesions. Here, we build on earlier work on cross-sectional lesion segmentation; we present a fully automatic pipeline for serial analysis of FLAIR-hyperintense white matter lesions. Our algorithm requires three-dimensional gradient echo T1- and FLAIR- weighted images at 3 Tesla as well as available cross-sectional lesion segmentations of both time points. Preprocessing steps include lesion filling and intrasubject registration. For segmentation of lesion changes, initial lesion maps of different time points are fused; herein changes in intensity are analyzed at the voxel level. Significance of lesion change is estimated by comparison with the difference distribution of FLAIR intensities within normal appearing white matter. The method is validated on MRI data of two time points from 40 subjects with multiple sclerosis derived from two different scanners (20 subjects per scanner). Manual segmentation of lesion increases served as gold standard. Across all lesion increases, voxel-wise Dice coefficient (0.7) as well as lesion-wise detection rate (0.8) and false-discovery rate (0.2) indicate good overall performance. Analysis of scans from a repositioning experiment in a single patient with multiple sclerosis did not yield a single false positive lesion. We also introduce the lesion change plot as a descriptive tool for the lesion change of individual patients with regard to both number and volume. An open source implementation of the algorithm is available at http://www.statistical-modeling.de/lst.html.

MRI-based prediction of conversion from clinically isolated syndrome to clinically definite multiple sclerosis using SVM and lesion geometry.

Neuroanatomical pattern classification using support vector machines (SVMs) has shown promising results in classifying Multiple Sclerosis (MS) patients based on individual structural magnetic resonance images (MRI). To determine whether pattern classification using SVMs facilitates predicting conversion to clinically definite multiple sclerosis (CDMS) from clinically isolated syndrome (CIS). We used baseline MRI data from 364 patients with CIS, randomised to interferon beta-1b or placebo. Non-linear SVMs and 10-fold cross-validation were applied to predict converters/non-converters (175/189) at two years follow-up based on clinical and demographic data, lesion-specific quantitative geometric features and grey-matter-to-whole-brain volume ratios. We applied linear SVM analysis and leave-one-out cross-validation to subgroups of converters (n = 25) and non-converters (n = 44) based on cortical grey matter segmentations. Highest prediction accuracies of 70.4% (p = 8e-5) were reached with a combination of lesion-specific geometric (image-based) and demographic/clinical features. Cortical grey matter was informative for the placebo group (acc.: 64.6%, p = 0.002) but not for the interferon group. Classification based on demographic/clinical covariates only resulted in an accuracy of 56% (p = 0.05). Overall, lesion geometry was more informative in the interferon group, EDSS and sex were more important for the placebo cohort. Alongside standard demographic and clinical measures, both lesion geometry and grey matter based information can aid prediction of conversion to CDMS.

Comparison between balanced steady-state free precession and standard spoiled gradient echo magnetization transfer ratio imaging in multiple sclerosis: methodical and clinical considerations.

Different pathological processes like demyelination and axonal loss can alter the magnetisation transfer ratio (MTR) in brain tissue. The standard method to measure this effect is to scan the respective tissue twice, one with and one without a specific saturation pulse. A major drawback of this technique based on spoiled gradient echo (GRE) sequences relates to its long acquisition time due to the saturation pulses. Recently, an alternative concept for MT imaging based on balanced steady state free precession (bSSFP) has been proposed. Modification of the duration of the radiofrequency pulses for imaging allows scanning MT sensitive and non-sensitive images. The steady-state character of bSSFP with high intrinsic signal-to-noise ratio (SNR) allows three-dimensional (3D) whole brain MTR at high spatial resolution within short and thus clinically feasible acquisition times. In the present study, both bSSFP-MT and 2D GRE-MT imaging were used in a cohort of 31 patients with multiple sclerosis (MS) to characterize different normal appearing (NA) and pathological brain structures. Under the constraint of identical SNR and scan time, a 3.4 times higher voxel size could be achieved with bSSFP. This increased resolution allowed a more accurate delineation of the different brain structures, especially of cortex, hippocampus and MS lesions. In a multiple linear regression model, we found an association between MTR of cortical lesions and a clinical measure of disability (r= -0.407, p=0.035) in the bSSFP dataset only. The different relaxation weighting of the base images (T2/T1 in bSSFP, proton density in GRE) had no effects besides a larger spreading of the MTR values of the different NA structures. This was demonstrated by the nearly perfect linearity between the NA matter MTR of both techniques as well as in the absolute MTR differences between NA matter and the respective lesions.

Progression in disability and regional grey matter atrophy in relapsing-remitting multiple sclerosis.

BACKGROUND: In multiple sclerosis (MS) regional grey matter (GM) atrophy has been associated with disability progression. OBJECTIVE: The aim of this study was to compare regional GM volume changes in relapsing-remitting MS (RRMS) patients with progressive and stable disability, using voxel-based morphometry (VBM). METHODS: We acquired baseline and 1-year follow-up 3-dimensional (3D) T1-weighted magnetic resonance imaging (MRI) data of RRMS patients, using two 1.5-Tesla scanners. Patients were matched pair-wise with respect to age, gender, disease duration, medication, scanner and baseline Expanded Disability Status Scale (EDSS) into 13 pairs, with either progressive EDSS (≥ 1 point change y(-1)) or stable EDSS, as well as into 29 pairs with either progressive Multiple Sclerosis Functional Composite (MSFC) at ≥ 0.25% decrease in y(-1) in any component, or stable MSFC. We analysed longitudinal regional differences in GM volumes in the progressive and stable EDSS and MSFC groups, respectively, using VBM. RESULTS: Significant GM volume reductions occurred in the right precuneus, in the progressive EDSS group. Differential between-group effects occurred in the right precuneus and in the postcentral gyrus. Further longitudinal GM volume reductions occurred in the right orbicular gyrus, in the progressive MSFC group, but no between-group differences were observed (non-stationary cluster-wise inference, all P(corrected) < 0.05). CONCLUSION: These results suggested a direct association of disability progression and regional GM atrophy in RRMS.

Contribution of cortical and white matter lesions to cognitive impairment in multiple sclerosis.

BACKGROUND: Cortical lesions (CLs) have been reported to be a better predictor for cognitive impairment than white matter (WM) lesions in relapsing-remitting multiple sclerosis (RRMS). OBJECTIVES: The objectives of this article are to investigate the contribution of CLs and WM lesions to cognitive impairment in 91 patients with MS and clinically isolated syndrome, and to test potential associations of CLs and WM lesions with fatigue and depression. METHODS: Lesions were scored and segmented on 3D double inversion recovery sequences, according to their location (cortical, WM). Normalised grey matter volume was also determined. Cognitive performance was assessed with the SDMT and PASAT-3, fatigue with the FSMC and depression with the German version of the CES-D. RESULTS: CL volume did not correlate with fatigue or depression, but correlated significantly with both neuropsychological outcome measures: PASAT-3 (r = -0.275, p = 0.009) and SDMT (r = -0.377, p < 0.001). Multiple regression analyses with age, WM lesions, CLs and GM volume as independent variables, however, did not reveal CL volume as a significant predictor of neuropsychological outcomes, whereas WM lesion volume significantly predicted SDMT and by trend PASAT performance. CONCLUSIONS: These findings suggest a role of WM lesions in the development of cognitive deficits, especially information-processing speed, which may be higher than previously assumed. ABBREVIATIONS CES-D: Center for Epidemiologic Studies Depression scale (ADS-L: Allgemeine Depressions Skala-L, German version of CES-D), CIS: clinically isolated syndrome, CL: cortical lesion, DIR: double inversion recovery, EDSS: Expanded Disability Status Scale, FSMC: fatigue scale for motor and cognitive functions, GM: grey matter, MRI: magnetic resonance imaging, MS: multiple sclerosis, PASAT-3: paced auditory serial addition test 3s, PPMS: primary progressive multiple sclerosis, RRMS: relapsing-remitting multiple sclerosis, SDMT: symbol digit modalities test, SPM: statistical parametric mapping, SPMS: secondary progressive multiple sclerosis, WM: white matter.

Hippocampal volume in subjects at high risk of psychosis: a longitudinal MRI study.

INTRODUCTION: The hippocampal formation has been studied extensively in schizophrenic psychoses and alterations in hippocampal anatomy have been consistently reported. Chronic schizophrenia seems to be associated with bilateral hippocampal volume (HV) reduction, while in patients with an at-risk mental state (ARMS) there are contradictory results. This is the first region of interest (ROI) based follow-up MRI study of hippocampal volume comparing ARMS individuals with and without transition to psychosis. The aim was to investigate the timing of HV changes in ARMS in the early phase of psychosis. METHODS: Magnetic resonance imaging data from 18 antipsychotic-naïve individuals with an ARMS were collected within the FePsy-clinic for early detection of psychoses. During follow-up 8 subjects transitioned to psychosis (ARMS-T) and 10 did not (ARMS-NT). Subjects were re-scanned after the onset of psychosis or at the end of the follow-up if they did not develop psychosis. RESULTS: Across both groups there was a significant decrease in HV over time (p<0.05). There was no significant difference in progression between ARMS-T and ARMS-NT. Antipsychotic medication at follow up was associated with increased HV (p<0.05). CONCLUSIONS: We found a decrease of HV over time in subjects with an ARMS, independently of clinical outcome. We may speculate that the decrease of HV over time might reflect brain degeneration processes.

Spatiotemporal distribution of white matter lesions in relapsing-remitting and secondary progressive multiple sclerosis.

BACKGROUND: Multiple sclerosis (MS) is a chronic inflammatory disease of the central nervous system. MS lesions show a typical distribution pattern and primarily affect the white matter (WM) in the periventricular zone and in the centrum semiovale. OBJECTIVE: To track lesion development during disease progression, we compared the spatiotemporal distribution patterns of lesions in relapsing-remitting MS (RRMS) and secondary progressive MS (SPMS). METHODS: We used T1 and T2 weighted MR images of 209 RRMS and 62 SPMS patients acquired on two different 1.5 Tesla MR scanners in two clinical centers followed up for 25 (± 1.7) months. Both cross-sectional and longitudinal differences in lesion distribution between RRMS and SPMS patients were analyzed with lesion probability maps (LPMs) and permutation-based inference. RESULTS: MS lesions clustered around the lateral ventricles and in the centrum semiovale. Cross-sectionally, compared to RRMS patients, the SPMS patients showed a significantly higher regional probability of T1 hypointense lesions (p ≤ 0.03) in the callosal body, the corticospinal tract, and other tracts adjacent to the lateral ventricles, but not of T2 lesions (peak probabilities were RRMS: T1 9%, T2 18%; SPMS: T1 21%, T2 27%). No longitudinal changes of regional T1 and T2 lesion volumes between baseline and follow-up scan were found. CONCLUSION: The results suggest a particular vulnerability to neurodegeneration during disease progression in a number of WM tracts.

Multivariate pattern classification of gray matter pathology in multiple sclerosis.

Univariate analyses have identified gray matter (GM) alterations in different groups of MS patients. While these methods detect differences on the basis of the single voxel or cluster, multivariate methods like support vector machines (SVM) identify the complex neuroanatomical patterns of GM differences. Using multivariate linear SVM analysis and leave-one-out cross-validation, we aimed at identifying neuroanatomical GM patterns relevant for individual classification of MS patients. We used SVM to separate GM segmentations of T1-weighted three-dimensional magnetic resonance (MR) imaging scans within different age- and sex-matched groups of MS patients with either early (n=17) or late MS (n=17) (contrast I), low (n=20) or high (n=20) white matter lesion load (contrast II), and benign MS (BMS, n=13) or non-benign MS (NBMS, n=13) (contrast III) scanned on a single 1.5 T MR scanner. GM patterns most relevant for individual separation of MS patients comprised cortical areas of all the cerebral lobes as well as deep GM structures, including the thalamus and caudate. The patterns detected were sufficiently informative to separate individuals of the respective groups with high sensitivity and specificity in 85% (contrast I), 83% (contrast II) and 77% (contrast III) of cases. The study demonstrates that neuroanatomical spatial patterns of GM segmentations contain information sufficient for correct classification of MS patients at the single case level, thus making multivariate SVM analysis a promising clinical application.

Longitudinal gray matter changes in multiple sclerosis--differential scanner and overall disease-related effects.

Voxel-based morphometry (VBM) has been used repeatedly in single-center studies to investigate regional gray matter (GM) atrophy in multiple sclerosis (MS). In multi-center trials, across-scanner variations might interfere with the detection of disease-specific structural abnormalities, thereby potentially limiting the use of VBM. Here we evaluated longitudinally inter-site differences and inter-site comparability of regional GM in MS using VBM. Baseline and follow up 3D T1-weighted magnetic resonance imaging (MRI) data of 248 relapsing-remitting (RR) MS patients, recruited in two clinical centers, (center1/2: n = 129/119; mean age 42.6 ± 10.7/43.3 ± 9.3; male:female 33:96/44:75; median disease duration 150 [72-222]/116 [60-156]) were acquired on two different 1.5T MR scanners. GM volume changes between baseline and year 2 while controlling for age, gender, disease duration, and global GM volume were analyzed. The main effect of time on regional GM volume was larger in data of center two as compared to center one in most of the brain regions. Differential effects of GM volume reductions occurred in a number of GM regions of both hemispheres, in particular in the fronto-temporal and limbic cortex (cluster P corrected <0.05). Overall disease-related effects were found bilaterally in the cerebellum, uncus, inferior orbital gyrus, paracentral lobule, precuneus, inferior parietal lobule, and medial frontal gyrus (cluster P corrected <0.05). The differential effects were smaller as compared to the overall effects in these regions. These results suggest that the effects of different scanners on longitudinal GM volume differences were rather small and thus allow pooling of MR data and subsequent combined image analysis.

Pituitary volume increase during emerging psychosis.

BACKGROUND: Morphologic abnormalities of the pituitary gland volume (PV) have been reported in schizophrenia, but at what point in time they occur remains unclear. This study determines PV across different stages of emerging psychotic disorders compared to healthy controls. METHODS: We compared PV of 36 individuals with an at-risk mental state (ARMS) for psychosis, 23 patients with a first episode psychosis (FEP) and 20 healthy controls (HC). Transition to psychosis was monitored using the BPRS transition criteria according to Yung et al. (Yung, A.R. et al., 1998. Prediction of psychosis. A step towards indicated prevention of schizophrenia. Br. J. Psychiatry Suppl. 172 (33), 14-20). Applying these transition criteria, 16 of the 36 ARMS individuals made the transition to psychosis (ARMS-T) and 20 did not (ARMS-NT). We traced PV manually on 1mm slices of magnetic resonance images in three dimensions (coronal, sagittal and axial) blind to group status. We used univariate analysis of covariance (ANCOVA) with PV as dependent variable, group and sex as between-subject factors and whole brain volume as covariate. RESULTS: PV increased from HC to ARMS-NT to ARMS-T/FEP. ANCOVA revealed a significant effect of group (F(3,78)=3.0; p=.036) and a sex × group interaction (F(3,78)=6.5; p=.001). Over all groups, women had considerably larger PV than men (F(1,78)=9.8; p=.003). CONCLUSIONS: Our findings provide further evidence that PV is increased in emerging psychotic disorders, and suggest that this is due to a stress-associated activation of the pituitary gland.

Effect of immunomodulatory medication on regional gray matter loss in relapsing-remitting multiple sclerosis--a longitudinal MRI study.

Prevention of global gray matter (GM) volume changes in multiple sclerosis (MS) are an objective in clinical trials, but the effect of immunomodulatory medication on regional GM atrophy progression is unclear. MRIs from 86 patients with relapsing-remitting MS (RRMS) followed up for 24 months were analyzed using voxel-based morphometry. An analysis of covariance model (cluster threshold, corrected p<0.05) was used to compare GM volumes between baseline and follow-up while stratified by immunomodulatory medication (IM): Interferone INF-beta-1a (n=34), INF-beta-1b (n=16), glatiramer acetate (GA) (n=15), and no-immunomodulatory treatment (n=21). In the INF-beta-1a/1b group (n=50), significant GM volume reductions were observed during follow-up in fronto-temporal, cingulate and cerebellar cortical brain regions, without significant differences between the INF-beta-1a and INF-beta-1b patients. In the GA group and in unmedicated patients, no significant regional GM volume reductions were observed. In contrast to GA, INF-beta-1a/1b treatment was associated with GM volume reductions in hippocampal/parahippocampal and anterior cingulate cortex. This is the first longitudinal study investigating the effects of IMs on GM in RRMS. Results suggest differences in the dynamics of regional GM volume atrophy in differentially treated or untreated RRMS patients.

Spatiotemporal distribution pattern of white matter lesion volumes and their association with regional grey matter volume reductions in relapsing-remitting multiple sclerosis.

The association of white matter (WM) lesions and grey matter (GM) atrophy is a feature in relapsing-remitting multiple sclerosis (RRMS). The spatiotemporal distribution pattern of WM lesions, their relations to regional GM changes and the underlying dynamics are unclear. Here we combined parametric and non-parametric voxel-based morphometry (VBM) to clarify these issues. MRI data from RRMS patients with progressive (PLV, n = 45) and non-progressive WM lesion volumes (NPLV, n = 44) followed up for 12 months were analysed. Cross-sectionally, the spatial WM lesion distribution was compared using lesion probability maps (LPMs). Longitudinally, WM lesions and GM volumes were studied using FSL-VBM and SPM5-VBM, respectively. WM lesions clustered around the lateral ventricles and in the centrum semiovale with a more widespread pattern in the PLV than in the NPLV group. The maximum local probabilities were similar in both groups and higher for T2 lesions (PLV: 27%, NPLV: 25%) than for T1 lesions (PLV: 15%, NPLV 14%). Significant WM lesion changes accompanied by cortical GM volume reductions occurred in the corpus callosum and optic radiations (P = 0.01 corrected), and more liberally tested (uncorrected P < 0.01) in the inferior fronto-occipital and longitudinal fasciculi, and corona radiata in the PLV group. Not any WM or GM changes were found in the NPLV group. In the PLV group, WM lesion distribution and development in fibres, was associated with regional GM volume loss. The different spatiotemporal distribution patterns of patients with progressive compared to patients with non-progressive WM lesions suggest differences in the dynamics of pathogenesis.

Association of regional gray matter volume loss and progression of white matter lesions in multiple sclerosis - A longitudinal voxel-based morphometry study.

Previous studies have established regional gray matter (GM) volume loss in multiple sclerosis (MS) but the relationship between development of white matter (WM) lesions and changes of regional GM volumes is unclear. The present study addresses this issue by means of voxel-based morphometry (VBM). T1-weighted three-dimensional magnetic resonance imaging (MRI) data from MS patients followed up for 12 months were analyzed using VBM. An analysis of covariance model assessed with cluster size inference (all corrected for multiple comparisons, p<0.01) was used to compare GM volumes between baseline and follow-up while controlling for age, gender, and disease duration. Lesion burden, i.e. volumes of T1 hypointense and T2 hyperintense lesions and the number of new T2 lesions at year one, was also determined. Comparing all MS patients (n=211) longitudinally, GM volume remained unchanged during one year-follow-up. Focusing on patients with relapsing remitting MS (RRMS) (n=151), significant cortical GM volume reductions between baseline and follow-up scans were found in the anterior and posterior cingulate, the temporal cortex, and cerebellum. Within the RRMS group, those patients with increasing T2 and T1 lesion burden (n=45) showed additional GM volume loss during follow-up in the frontal and parietal cortex, and precuneus. In contrast, patients lacking an increase in WM lesion burden (n=44) did not show any significant GM changes. The present study suggests that the progression of regional GM volume reductions is associated with WM lesion progression and occurs predominantly in fronto-temporal cortical areas.