A Fully Automatic Method to Segment Choroid Plexuses in Multiple Sclerosis Using Conventional MRI Sequences

BACKGROUND

Choroid plexus (CP) volume has been recently proposed as a proxy for brain neuroinflammation in multiple sclerosis (MS).

PURPOSE

To develop and validate a fast automatic method to segment CP using routinely acquired brain T1-weighted and FLAIR MRI.

STUDY TYPE

Retrospective.

POPULATION

Fifty-five MS patients (33 relapsing-remitting, 22 progressive; mean age = 46.8 ± 10.2 years; 31 women) and 60 healthy controls (HC; mean age = 36.1 ± 12.6 years, 33 women).

FIELD STRENGTH/SEQUENCE

3D T2-weighted FLAIR and 3D T1-weighted gradient echo sequences at 3.0 T.

ASSESSMENT

Brain tissues were segmented on T1-weighted sequences and a Gaussian Mixture Model (GMM) was fitted to FLAIR image intensities obtained from the ventricle masks of the SIENAX. A second GMM was then applied on the thresholded and filtered ventricle mask. CP volumes were automatically determined and compared with those from manual segmentation by two raters (with 3 and 10 years' experience; reference standard). CP volumes from previously published automatic segmentation methods (freely available Freesurfer [FS] and FS-GMM) were also compared with reference standard. Expanded Disability Status Scale (EDSS) score was assessed within 3 days of MRI. Computational time was assessed for each automatic technique and manual segmentation.

STATISTICAL TESTS

Comparisons of CP volumes with reference standard were evaluated with Bland Altman analysis. Dice similarity coefficients (DSC) were computed to assess automatic CP segmentations. Volume differences between MS and HC for each method were assessed with t-tests and correlations of CP volumes with EDSS were assessed with Pearson's correlation coefficients (R). A P value <0.05 was considered statistically significant.

RESULTS

Compared to manual segmentation, the proposed method had the highest segmentation accuracy (mean DSC = 0.65 ± 0.06) compared to FS (mean DSC = 0.37 ± 0.08) and FS-GMM (0.58 ± 0.06). The percentage CP volume differences relative to manual segmentation were -0.1% ± 0.23, 4.6% ± 2.5, and -0.48% ± 2 for the proposed method, FS, and FS-GMM, respectively. The Pearson's correlations between automatically obtained CP volumes and the manually obtained volumes were 0.70, 0.54, and 0.56 for the proposed method, FS, and FS-GMM, respectively. A significant correlation between CP volume and EDSS was found for the proposed automatic pipeline (R = 0.2), for FS-GMM (R = 0.3) and for manual segmentation (R = 0.4). Computational time for the proposed method (32 ± 2 minutes) was similar to the manual segmentation (20 ± 5 minutes) but <25% of the FS (120 ± 15 minutes) and FS-GMM (125 ± 15 minutes) methods.

DATA CONCLUSION

This study developed an accurate and easily implementable method for automatic CP segmentation in MS using T1-weighted and FLAIR MRI.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 4.

Monoaminergic network dysfunction and development of depression in multiple sclerosis: a longitudinal investigation

BACKGROUND

Monoaminergic network dysfunction is thought to underpin depression in multiple sclerosis (MS) patients. However, longitudinal studies are lacking.

OBJECTIVES

Here, we investigated the association between development of depressive symptoms in MS and changes of resting-state functional connectivity (RS FC) within monoaminergic networks.

METHODS

Forty-nine MS patients without depression [Montgomery-Asberg Depression Scale (MADRS) ≤ 9] and 27 healthy controls underwent clinical and 3.0 T RS FC assessment at baseline and after a median follow-up of 1.6 years (interquartile range 1.0-2.1 years). Monoamine-related RS FC was derived by independent component analysis, constrained to PET atlases for dopamine, noradrenaline and serotonin transporters. Longitudinal changes of RS FC within monoaminergic networks and their correlations with MADRS scores were assessed.

RESULTS

At baseline, MS patients showed decreased RS FC vs healthy controls in all PET-guided monoaminergic networks in frontal, cingulate and cerebellar cortices, and increased RS FC in parieto-occipital regions. Fourteen (29%) MS patients developed depressive symptoms (MADRS > 9) at follow-up (D-MS) and exhibited widespread RS FC decrease over time in the PET-guided dopamine network, mainly in orbitofrontal, occipital, anterior cingulate and precuneal cortices compared to patients who did not develop depressive symptoms. In D-MS, decreased RS FC over time was also observed in parahippocampal and occipital regions of the PET-guided noradrenaline network. Decreased RS FC over time in dopamine and noradrenaline PET-guided networks correlated with concomitant increased MADRS scores (r = range - 0.65/- 0.61, p < 0.001).

CONCLUSIONS

The development of depressive symptoms in MS patients was associated with specific RS FC changes within the dopamine and noradrenaline networks.

Chronic Active Lesions and Larger Choroid Plexus Explain Cognition and Fatigue in Multiple Sclerosis

BACKGROUND AND OBJECTIVES

Chronic inflammation may contribute to cognitive dysfunction and fatigue in patients with multiple sclerosis (MS). Paramagnetic rim lesions (PRLs) and choroid plexus (CP) enlargement have been proposed as markers of chronic inflammation in MS being associated with a more severe disease course. However, their relation with cognitive impairment and fatigue has not been fully explored yet. Here, we investigated the contribution of PRL number and volume and CP enlargement to cognitive impairment and fatigue in patients with MS.

METHODS

Brain 3T MRI, neurologic evaluation, and neuropsychological assessment, including the Brief Repeatable Battery of Neuropsychological Tests and Modified Fatigue Impact Scale, were obtained from 129 patients with MS and 73 age-matched and sex-matched healthy controls (HC). PRLs were identified on phase images of susceptibility-weighted imaging, whereas CP volume was quantified using a fully automatic method on brain three-dimensional T1-weighted and fluid-attenuated inversion recovery MRI sequences. Predictors of cognitive impairment and fatigue were identified using random forest.

RESULTS

Thirty-six (27.9%) patients with MS were cognitively impaired, and 31/113 (27.4%) patients had fatigue. Fifty-nine (45.7%) patients with MS had ≥1 PRLs (median = 0, interquartile range = 0;2). Compared with HC, patients with MS showed significantly higher T2-hyperintense white matter lesion (WM) volume; lower normalized brain, thalamic, hippocampal, caudate, cortical, and WM volumes; and higher normalized CP volume (p from <0.001 to 0.040). The predictors of cognitive impairment (relative importance) (out-of-bag area under the curve [OOB-AUC] = 0.707) were normalized brain volume (100%), normalized caudate volume (89.1%), normalized CP volume (80.3%), normalized cortical volume (70.3%), number (67.3%) and volume (66.7%) of PRLs, and T2-hyperintense WM lesion volume (64.0%). Normalized CP volume was the only predictor of the presence of fatigue (OOB-AUC = 0.563).

DISCUSSION

Chronic inflammation, with higher number and volume of PRLs and enlarged CP, may contribute to cognitive impairment in MS in addition to gray matter atrophy. The contribution of enlarged CP in explaining fatigue supports the relevance of immune-related processes in determining this manifestation independently of disease severity. PRLs and CP enlargement may contribute to the pathophysiology of cognitive impairment and fatigue in MS, and they may represent clinically relevant therapeutic targets to limit the impact of these clinical manifestations in MS.

Spatial correspondence among regional gene expressions and gray matter volume loss in multiple sclerosis

In multiple sclerosis (MS), a non-random and clinically relevant pattern of gray matter (GM) volume loss has been described. Whether differences in regional gene expression might underlay distinctive pathological processes contributing to this regional variability has not been explored yet. Two hundred eighty-six MS patients and 172 healthy controls (HC) underwent a brain 3T MRI, a complete neurological evaluation and a neuropsychological assessment. Using Allen Human Brain Atlas, voxel-based morphometry and MENGA platform, we integrated brain transcriptome and neuroimaging data to explore the spatial cross-correlations between regional GM volume loss and expressions of 2710 genes involved in MS (p < 0.05, family-wise error-corrected). Enrichment analyses were performed to evaluate overrepresented molecular functions, biological processes and cellular components involving genes significantly associated with voxel-based morphometry-derived GM maps (p < 0.05, Bonferroni-corrected). A diffuse GM volume loss was found in MS patients compared to HC and it was spatially correlated with 74 genes involved in GABA neurotransmission and mitochondrial oxidoreductase activity mainly expressed in neurons and astrocytes. A more severe GM volume loss was spatially associated, in more disabled MS patients, with 44 genes involved in mitochondrial integrity of all resident cells of the central nervous system (CNS) and, in cognitively impaired MS patients, with 64 genes involved in mitochondrial protein heterodimerization and oxidoreductase activities expressed also in microglia and endothelial cells. Specific differences in the expressions of genes involved in synaptic GABA receptor activities and mitochondrial functions in resident CNS cells may influence regional susceptibility to MS-related excitatory/inhibitory imbalance and oxidative stress, and subsequently, to GM volume loss.

Quantification of Thalamic Atrophy in MS: From the Multicenter Italian Neuroimaging Network Initiative Data Set to Clinical Application

BACKGROUND AND PURPOSE

Thalamic atrophy occurs from the earliest phases of MS; however, this measure is not included in clinical practice. Our purpose was to obtain a reliable segmentation of the thalamus in MS by comparing existing automatic methods cross-sectionally and longitudinally.

MATERIALS AND METHODS

MR images of 141 patients with relapsing-remitting MS (mean age, 38 years; range, 19-58 years; 95 women) and 69 healthy controls (mean age, 36 years; range, 22-69 years; 47 women) were retrieved from the Italian Neuroimaging Network Initiative repository: T1WI, T2WI, and DWI at baseline and after 1 year (136 patients, 31 healthy controls). Three segmentation software programs (FSL-FIRST, FSL-MIST, FreeSurfer) were compared. At baseline, agreement among pipelines, correlations with age, disease duration, clinical score, and T2-hyperintense lesion volume were evaluated. Effect sizes in differentiating patients and controls were assessed cross-sectionally and longitudinally. Variability of longitudinal changes in controls and sample sizes were assessed. False discovery rate-adjusted P < .05 was considered significant.

RESULTS

At baseline, FSL-FIRST and FSL-MIST showed the highest agreement in the results of thalamic volume (R = 0.87, P < .001), with the highest effect size for FSL-MIST (Cohen d = 1.11); correlations with demographic and clinical variables were comparable for all software. Longitudinally, FSL-MIST showed the lowest variability in estimating thalamic volume changes for healthy controls (SD = 1.07%), the highest effect size (Cohen d = 0.44), and the smallest sample size at 80% power level (15 subjects per group).

CONCLUSIONS

Multimodal segmentation by FSL-MIST increased the robustness of the results with better capability to detect small variations in thalamic volumes.

Methods for Brain Atrophy MR Quantification in Multiple Sclerosis: Application to the Multicenter INNI Dataset

BACKGROUND

Current therapeutic strategies in multiple sclerosis (MS) target neurodegeneration. However, the integration of atrophy measures into the clinical scenario is still an unmet need.

PURPOSE

To compare methods for whole-brain and gray matter (GM) atrophy measurements using the Italian Neuroimaging Network Initiative (INNI) dataset.

STUDY TYPE

Retrospective (data available from INNI).

POPULATION

A total of 466 patients with relapsing-remitting MS (mean age = 37.3 ± 10 years, 323 women) and 279 healthy controls (HC; mean age = 38.2 ± 13 years, 164 women).

FIELD STRENGTH/SEQUENCE

A 3.0-T, T1-weighted (spin echo and gradient echo without gadolinium injection) and T2-weighted spin echo scans at baseline and after 1 year (170 MS, 48 HC).

ASSESSMENT

Structural Image Evaluation using Normalization of Atrophy (SIENA-X/XL; version 5.0.9), Statistical Parametric Mapping (SPM-v12); and Jim-v8 (Xinapse Systems, Colchester, UK) software were applied to all subjects.

STATISTICAL TESTS

In MS and HC, we evaluated the intraclass correlation coefficient (ICC) among FSL-SIENA(XL), SPM-v12, and Jim-v8 for cross-sectional whole-brain and GM tissue volumes and their longitudinal changes, the effect size according to the Cohen's d at baseline and the sample size requirement for whole-brain and GM atrophy progression at different power levels (lowest = 0.7, 0.05 alpha level). False discovery rate (Benjamini-Hochberg procedure) correction was applied. A P value <0.05 was considered statistically significant.

RESULTS

SPM-v12 and Jim-v8 showed significant agreement for cross-sectional whole-brain (ICC = 0.93 for HC and ICC = 0.84 for MS) and GM volumes (ICC = 0.66 for HC and ICC = 0.90) and longitudinal assessment of GM atrophy (ICC = 0.35 for HC and ICC = 0.59 for MS), while no significant agreement was found in the comparisons between whole-brain and GM volumes for SIENA-X/XL and both SPM-v12 (P = 0.19 and P = 0.29, respectively) and Jim-v8 (P = 0.21 and P = 0.32, respectively). SPM-v12 and Jim-v8 showed the highest effect size for cross-sectional GM atrophy (Cohen's d = -0.63 and -0.61). Jim-v8 and SIENA(XL) showed the smallest sample size requirements for whole-brain (58) and GM atrophy (152), at 0.7 power level.

DATA CONCLUSION

The findings obtained in this study should be considered when selecting the appropriate brain atrophy pipeline for MS studies.

EVIDENCE LEVEL

4.

TECHNICAL EFFICACY

Stage 1.

Emerging Perspectives on MRI Application in Multiple Sclerosis: Moving from Pathophysiology to Clinical Practice

MRI plays a central role in the diagnosis of multiple sclerosis (MS) and in the monitoring of disease course and treatment response. Advanced MRI techniques have shed light on MS biology and facilitated the search for neuroimaging markers that may be applicable in clinical practice. MRI has led to improvements in the accuracy of MS diagnosis and a deeper understanding of disease progression. This has also resulted in a plethora of potential MRI markers, the importance and validity of which remain to be proven. Here, five recent emerging perspectives arising from the use of MRI in MS, from pathophysiology to clinical application, will be discussed. These are the feasibility of noninvasive MRI-based approaches to measure glymphatic function and its impairment; T1-weighted to T2-weighted intensity ratio to quantify myelin content; classification of MS phenotypes based on their MRI features rather than on their clinical features; clinical relevance of gray matter atrophy versus white matter atrophy; and time-varying versus static resting-state functional connectivity in evaluating brain functional organization. These topics are critically discussed, which may guide future applications in the field.

Neural stem cell transplantation in patients with progressive multiple sclerosis: an open-label, phase 1 study

Innovative pro-regenerative treatment strategies for progressive multiple sclerosis (PMS), combining neuroprotection and immunomodulation, represent an unmet need. Neural precursor cells (NPCs) transplanted in animal models of multiple sclerosis have shown preclinical efficacy by promoting neuroprotection and remyelination by releasing molecules sustaining trophic support and neural plasticity. Here we present the results of STEMS, a prospective, therapeutic exploratory, non-randomized, open-label, single-dose-finding phase 1 clinical trial ( NCT03269071 , EudraCT 2016-002020-86), performed at San Raffaele Hospital in Milan, Italy, evaluating the feasibility, safety and tolerability of intrathecally transplanted human fetal NPCs (hfNPCs) in 12 patients with PMS (with evidence of disease progression, Expanded Disability Status Scale ≥6.5, age 18-55 years, disease duration 2-20 years, without any alternative approved therapy). The safety primary outcome was reached, with no severe adverse reactions related to hfNPCs at 2-year follow-up, clearly demonstrating that hfNPC therapy in PMS is feasible, safe and tolerable. Exploratory secondary analyses showed a lower rate of brain atrophy in patients receiving the highest dosage of hfNPCs and increased cerebrospinal fluid levels of anti-inflammatory and neuroprotective molecules. Although preliminary, these results support the rationale and value of future clinical studies with the highest dose of hfNPCs in a larger cohort of patients.

Multicenter data harmonization for regional brain atrophy and application in multiple sclerosis

BACKGROUND

In multiple sclerosis (MS), determination of regional brain atrophy is clinically relevant. However, analysis of large datasets is rare because of the increased variability in multicenter data.

PURPOSE

To compare different methods to correct for center effects. To investigate regional gray matter (GM) volume in relapsing-remitting MS in a large multicenter dataset.

METHODS

MRI scans of 466 MS patients and 279 healthy controls (HC) were retrieved from the Italian Neuroimaging Network Initiative repository. Voxel-based morphometry was performed. The center effect was accounted for with different methods: (a) no correction, (b) factor in the statistical model, (c) ComBat method and (d) subsampling procedure to match single-center distributions. By applying the best correction method, GM atrophy was assessed in MS patients vs HC and according to clinical disability, disease duration and T₂ lesion volume. Results were assessed voxel-wise using general linear model.

RESULTS

The average residuals for the harmonization methods were 5.03 (a), 4.42 (b), 4.26 (c) and 2.98 (d). The comparison between MS patients and HC identified thalami and other deep GM nuclei, the cerebellum and several cortical regions. At single-center analysis, the thalami were always involved, whereas different other regions were found in each center. Cerebellar atrophy correlated with clinical disability, while deep GM nuclei atrophy correlated with T₂-lesion volume.

CONCLUSION

Harmonization based on subsampling more effectively decreased the residuals of the statistical model applied. In comparison with findings from single-center analysis, the multicenter results were more robust, highlighting the importance of data repositories from multiple centers.

The insula modulates the effects of aerobic training on cardiovascular function and ambulation in multiple sclerosis

BACKGROUND

Impairment of cardiovascular control is common in multiple sclerosis (MS), possibly due to damage of strategic brain regions such as the insula. Aerobic training (AT) targets cardiopulmonary system and may represent a neuroprotective strategy.

PURPOSE

To investigate whether insular damage (T2-hyperintense lesions and volume) is associated with cardiovascular fitness (CF) and influences AT effects in MS.

METHODS

Sixty-one MS patients were randomized to an AT intervention group (MS-AT) and a motor training control group (MS-C). At baseline and after training (24 sessions over 2-3 months), peak of oxygen consumption (VO2max), heart rate reserve (HRR), 6-min walk test (6MWT) and whole brain and insula MRI data were collected. Two healthy control (HC) groups were enrolled for CF and MRI data analysis.

RESULTS

At baseline, MS patients vs HC showed impaired VO2max, HRR and 6MWT (p < 0.001) and widespread gray matter atrophy, including bilateral insula. In MS patients, left insula T2-lesion volume correlated with HRR (r = 0.27, p = 0.042). After training, MS-AT, especially those without insular T2-hyperintense lesions, showed 6MWT improvement (p < 0.05) and a stable insular volume, whereas MS-C showed left insular volume loss (p < 0.001).

CONCLUSIONS

By increasing 6MWT performance, our results suggest that AT may improve walking capacity and submaximal measure of CF in MS patients. Such beneficial effect may be modulated by insula integrity.

Prediction of the information processing speed performance in multiple sclerosis using a machine learning approach in a large multicenter magnetic resonance imaging data set

Many patients with multiple sclerosis (MS) experience information processing speed (IPS) deficits, and the Symbol Digit Modalities Test (SDMT) has been recommended as a valid screening test. Magnetic resonance imaging (MRI) has markedly improved the understanding of the mechanisms associated with cognitive deficits in MS. However, which structural MRI markers are the most closely related to cognitive performance is still unclear. We used the multicenter 3T-MRI data set of the Italian Neuroimaging Network Initiative to extract multimodal data (i.e., demographic, clinical, neuropsychological, and structural MRIs) of 540 MS patients. We aimed to assess, through machine learning techniques, the contribution of brain MRI structural volumes in the prediction of IPS deficits when combined with demographic and clinical features. We trained and tested the eXtreme Gradient Boosting (XGBoost) model following a rigorous validation scheme to obtain reliable generalization performance. We carried out a classification and a regression task based on SDMT scores feeding each model with different combinations of features. For the classification task, the model trained with thalamus, cortical gray matter, hippocampus, and lesions volumes achieved an area under the receiver operating characteristic curve of 0.74. For the regression task, the model trained with cortical gray matter and thalamus volumes, EDSS, nucleus accumbens, lesions, and putamen volumes, and age reached a mean absolute error of 0.95. In conclusion, our results confirmed that damage to cortical gray matter and relevant deep and archaic gray matter structures, such as the thalamus and hippocampus, is among the most relevant predictors of cognitive performance in MS.

Exploring in vivo multiple sclerosis brain microstructural damage through T1w/T2w ratio: a multicentre study

OBJECTIVES

To evaluate white matter and grey matter T1-weighted (w)/T2w ratio (T1w/T2w ratio) in healthy controls and patients with multiple sclerosis, and its association with clinical disability.

METHODS

In this cross-sectional study, 270 healthy controls and 434 patients with multiple sclerosis were retrospectively selected from 7 European sites. T1w/T2w ratio was obtained from brain T2w and T1w scans after intensity calibration using eyes and temporal muscle.

RESULTS

In healthy controls, T1w/T2w ratio increased until 50-60 years both in white and grey matter. Compared with healthy controls, T1w/T2w ratio was significantly lower in white matter lesions of all multiple sclerosis phenotypes, and in normal-appearing white matter and cortex of patients with relapsing-remitting and secondary progressive multiple sclerosis (p≤0.026), but it was significantly higher in the striatum and pallidum of patients with relapsing-remitting, secondary progressive and primary progressive multiple sclerosis (p≤0.042). In relapse-onset multiple sclerosis, T1w/T2w ratio was significantly lower in white matter lesions and normal-appearing white matter already at Expanded Disability Status Scale (EDSS) <3.0 and in the cortex only for EDSS ≥3.0 (p≤0.023). Conversely, T1w/T2w ratio was significantly higher in the striatum and pallidum for EDSS ≥4.0 (p≤0.005). In primary progressive multiple sclerosis, striatum and pallidum showed significantly higher T1w/T2w ratio beyond EDSS=6.0 (p≤0.001). In multiple sclerosis, longer disease duration, higher EDSS, higher brain lesional volume and lower normalised brain volume were associated with lower lesional and cortical T1w/T2w ratio and a higher T1w/T2w ratio in the striatum and pallidum (β from -1.168 to 0.286, p≤0.040).

CONCLUSIONS

T1w/T2w ratio may represent a clinically relevant marker sensitive to demyelination, neurodegeneration and iron accumulation occurring at the different multiple sclerosis phases.

Application of deep-learning to the seronegative side of the NMO spectrum

OBJECTIVES

To apply a deep-learning algorithm to brain MRIs of seronegative patients with neuromyelitis optica spectrum disorders (NMOSD) and NMOSD-like manifestations and assess whether their structural features are similar to aquaporin-4-seropositive NMOSD or multiple sclerosis (MS) patients.

PATIENTS AND METHODS

We analyzed 228 T2- and T1-weighted brain MRIs acquired from aquaporin-4-seropositive NMOSD (n = 85), MS (n = 95), aquaporin-4-seronegative NMOSD [n = 11, three with anti-myelin oligodendrocyte glycoprotein antibodies (MOG)], and aquaporin-4-seronegative patients with NMOSD-like manifestations (idiopathic recurrent optic neuritis and myelitis, n = 37), who were recruited from February 2010 to December 2019. Seventy-three percent of aquaporin-4-seronegative patients with NMOSD-like manifestations also had a clinical follow-up (median duration of 4 years). The deep-learning neural network architecture was based on four 3D convolutional layers. It was trained and validated on MRI scans of aquaporin-4-seropositive NMOSD and MS patients and was then applied to aquaporin-4-seronegative NMOSD and NMOSD-like manifestations. Assignment of unclassified aquaporin-4-seronegative patients was compared with their clinical follow-up.

RESULTS

The final algorithm differentiated aquaporin-4-seropositive NMOSD and MS patients with an accuracy of 0.95. All aquaporin-4-seronegative NMOSD and 36/37 aquaporin-4-seronegative patients with NMOSD-like manifestations were classified as NMOSD. Anti-MOG patients had a similar probability of being NMOSD or MS. At clinical follow-up, one unclassified aquaporin-4-seronegative patient evolved to MS, three developed NMOSD, and the others did not change phenotype.

CONCLUSIONS

Our findings support the inclusion of aquaporin4-seronegative patients into NMOSD and suggest a possible expansion to aquaporin-4-seronegative unclassified patients with NMOSD-like manifestations. Anti-MOG patients are likely to have intermediate brain features between NMOSD and MS.

In vivo gradients of thalamic damage in paediatric multiple sclerosis: a window into pathology

The thalamus represents one of the first structures affected by neurodegenerative processes in multiple sclerosis. A greater thalamic volume reduction over time, on its CSF side, has been described in paediatric multiple sclerosis patients. However, its determinants and the underlying pathological changes, likely occurring before this phenomenon becomes measurable, have never been explored. Using a multiparametric magnetic resonance approach, we quantified, in vivo, the different processes that can involve the thalamus in terms of focal lesions, microstructural damage and atrophy in paediatric multiple sclerosis patients and their distribution according to the distance from CSF/thalamus interface and thalamus/white matter interface. In 70 paediatric multiple sclerosis patients and 26 age- and sex-matched healthy controls, we tested for differences in thalamic volume and quantitative MRI metrics-including fractional anisotropy, mean diffusivity and T1/T2-weighted ratio-in the whole thalamus and in thalamic white matter, globally and within concentric bands originating from CSF/thalamus interface. In paediatric multiple sclerosis patients, the relationship of thalamic abnormalities with cortical thickness and white matter lesions was also investigated. Compared to healthy controls, patients had significantly increased fractional anisotropy in whole thalamus (f2 = 0.145; P = 0.03), reduced fractional anisotropy (f2 = 0.219; P = 0.006) and increased mean diffusivity (f2 = 0.178; P = 0.009) in thalamic white matter and a trend towards a reduced thalamic volume (f2 = 0.027; P = 0.058). By segmenting the whole thalamus and thalamic white matter into concentric bands, in paediatric multiple sclerosis we detected significant fractional anisotropy abnormalities in bands nearest to CSF (f2 = 0.208; P = 0.002) and in those closest to white matter (f2 range = 0.183-0.369; P range = 0.010-0.046), while we found significant mean diffusivity (f2 range = 0.101-0.369; P range = 0.018-0.042) and T1/T2-weighted ratio (f2 = 0.773; P = 0.001) abnormalities in thalamic bands closest to CSF. The increase in fractional anisotropy and decrease in mean diffusivity detected at the CSF/thalamus interface correlated with cortical thickness reduction (r range = -0.27-0.34; P range = 0.004-0.028), whereas the increase in fractional anisotropy detected at the thalamus/white matter interface correlated with white matter lesion volumes (r range = 0.24-0.27; P range = 0.006-0.050). Globally, our results support the hypothesis of heterogeneous pathological processes, including retrograde degeneration from white matter lesions and CSF-mediated damage, leading to thalamic microstructural abnormalities, likely preceding macroscopic tissue loss. Assessing thalamic microstructural changes using a multiparametric magnetic resonance approach may represent a target to monitor the efficacy of neuroprotective strategies early in the disease course.

Deep Learning on Conventional Magnetic Resonance Imaging Improves the Diagnosis of Multiple Sclerosis Mimics

OBJECTIVES

The aims of this study were to present a deep learning approach for the automated classification of multiple sclerosis and its mimics and compare model performance with that of 2 expert neuroradiologists.

MATERIALS AND METHODS

A total of 268 T2-weighted and T1-weighted brain magnetic resonance imagin scans were retrospectively collected from patients with migraine (n = 56), multiple sclerosis (n = 70), neuromyelitis optica spectrum disorders (n = 91), and central nervous system vasculitis (n = 51). The neural network architecture, trained on 178 scans, was based on a cascade of 4 three-dimensional convolutional layers, followed by a fully dense layer after feature extraction. The ability of the final algorithm to correctly classify the diseases in an independent test set of 90 scans was compared with that of the neuroradiologists.

RESULTS

The interrater agreement was 84.9% (Cohen κ = 0.78, P < 0.001). In the test set, deep learning and expert raters reached the highest diagnostic accuracy in multiple sclerosis (98.8% vs 72.8%, P < 0.001, for rater 1; and 81.8%, P < 0.001, for rater 2) and the lowest in neuromyelitis optica spectrum disorders (88.6% vs 4.4%, P < 0.001, for both raters), whereas they achieved intermediate values for migraine (92.2% vs 53%, P = 0.03, for rater 1; and 64.8%, P = 0.01, for rater 2) and vasculitis (92.1% vs 54.6%, P = 0.3, for rater 1; and 45.5%, P = 0.2, for rater 2). The overall performance of the automated method exceeded that of expert raters, with the worst misdiagnosis when discriminating between neuromyelitis optica spectrum disorders and vasculitis or migraine.

CONCLUSIONS

A neural network performed better than expert raters in terms of accuracy in classifying white matter disorders from magnetic resonance imaging and may help in their diagnostic work-up.

Effects of Fingolimod and Natalizumab on Brain T1-/T2-Weighted and Magnetization Transfer Ratios: a 2-Year Study

Fingolimod and natalizumab significantly reduce disease activity in relapsing-remitting multiple sclerosis (RRMS) and could promote tissue repair and neuroprotection. The ratio between conventional T1- and T2-weighted sequences (T1w/T2w-ratio) and magnetization transfer ratio (MTR) allow to quantify brain microstructural tissue abnormalities. Here, we compared fingolimod and natalizumab effects on brain T1w/T2w-ratio and MTR in RRMS over 2 years of treatment. RRMS patients starting fingolimod (n = 25) or natalizumab (n = 30) underwent 3T brain MRI scans at baseline (T0), month 6 (M6), month 12 (M12), and month 24 (M24). White matter (WM) lesions, normal-appearing (NA) WM, and gray matter (GM) T1w/T2w-ratio and MTR were estimated and compared between groups using linear mixed models. No baseline demographic, clinical, and MRI difference was found between groups. In natalizumab patients, lesion T1w/T2w-ratio and MTR significantly increased at M6 vs. T0 (p ≤ 0.035) and decreased at subsequent timepoints (p ≤ 0.037). In fingolimod patients, lesion T1w/T2w-ratio increased at M12 vs. T0 (p = 0.010), while MTR gradually increased at subsequent timepoints vs. T0 (p ≤ 0.027). Natalizumab stabilized NAWM and GM T1w/T2w-ratio and MTR. In fingolimod patients, NAWM T1w/T2w-ratio and MTR significantly increased at M24 vs. M12 (p ≤ 0.001). A significant GM T1w/T2w-ratio decrease at M6 vs. T0 (p = 0.014) and increase at M24 vs. M6 (p = 0.008) occurred, whereas GM MTR was significantly higher at M24 vs. previous timepoints (p ≤ 0.017) with significant between-group differences (p ≤ 0.034). Natalizumab may promote an early recovery of lesional damage and prevent microstructural damage accumulation in NAWM and GM during the first 2 years of treatment. Fingolimod enhances tissue damage recovery being visible after 6 months in lesions and after 2 years in NAWM and GM.

Mapping white matter damage distribution in neuromyelitis optica spectrum disorders with a multimodal MRI approach

BACKGROUND

The pathogenetic mechanisms sustaining neuroinflammatory disorders may originate from the cerebrospinal fluid.

OBJECTIVE

To evaluate white matter damage with diffusion tensor imaging and T1/T2-weighted ratio at progressive distances from the ventricular system in neuromyelitis optica spectrum disorders and multiple sclerosis.

METHODS

Fractional anisotropy, mean, axial, and radial diffusivity and T1/T2-weighted ratio maps were obtained from patients with seropositive neuromyelitis optica spectrum disorders, multiple sclerosis, and healthy controls (n = 20 each group). White matter damage was assessed as function of ventricular distance within progressive concentric bands.

RESULTS

Compared to healthy controls, neuromyelitis optica spectrum disorders patients had similar fractional anisotropy, radial and axial diffusivity, increased mean diffusivity (p = 0.009-0.013) and reduced T1/T2-weighted ratio (p = 0.024-0.037) in all bands. In multiple sclerosis, gradient of percentage lesion volume and intra-lesional mean and axial diffusivity were higher in periventricular bands. Compared to healthy controls, multiple sclerosis patients had reduced fractional anisotropy (p = 0.001-0.043) in periventricular bands, increased mean (p < 0.001), radial (p < 0.001-0.004), and axial diffusivity (p = 0.002-0.008) and preserved T1/T2-weighted ratio in all bands.

CONCLUSION

White matter damage is higher at periventricular level in multiple sclerosis and diffuse in neuromyelitis optica spectrum disorders. Fractional anisotropy preservation, associated with increased mean diffusivity and reduced T1/T2-weighted ratio may reflect astrocyte damage.

Measurement of white matter fiber-bundle cross-section in multiple sclerosis using diffusion-weighted imaging

BACKGROUND

When investigating white matter (WM) microstructure, the axonal fiber orientation should be considered. Constrained spherical deconvolution (CSD) is a diffusion-weighted imaging (DWI) method that estimates distribution of fibers within each imaging voxel.

OBJECTIVE

To study fiber-bundle cross-section (FC) as measured by CSD in multiple sclerosis (MS) patients versus healthy controls (HCs).

METHODS

DWI and three-dimensional (3D) T₁-weighted magnetic resonance imaging (MRI) were obtained from 45 MS patients and 45 HCs. We applied fixel-based morphometry analysis to assess differences of FC in MS against HCs and voxel-based analysis of fractional anisotropy (FA).

RESULTS

We found a significant widespread reduction of WM FC in MS compared to HCs. The decrease in FA was less extensive, mainly located in regions with high lesion occurrence such as the periventricular WM and the corpus callosum. Progressive MS patients showed a significant FC reduction in the right anterior cingulum, bilateral cerebellum, and in several mesencephalic and diencephalic regions compared to relapsing-remitting MS patients.

CONCLUSION

The CSD method can be applied in MS for a fiber-specific study of WM microstructure and quantification of FC. Fixel-based findings offered greater anatomical specificity and biological interpretability by identifying tract-specific differences and allowed substantial abnormalities to be detected.

Two-year regional grey and white matter volume changes with natalizumab and fingolimod

OBJECTIVE

To compare the efficacy of fingolimod and natalizumab in preventing regional grey matter (GM) and white matter (WM) atrophy in relapsing-remitting multiple sclerosis (RRMS) over 2 years.

METHODS

Patients with RRMS starting fingolimod (n=25) or natalizumab (n=30) underwent clinical examination and 3T MRI scans at baseline (month (M) 0), M6, M12 and M24. Seventeen healthy controls were also scanned at M0 and M24. Tensor-based morphometry and SPM12 were used to assess the longitudinal regional GM/WM volume changes.

RESULTS

At M0, no clinical or GM/WM volume differences were found between treatment groups. At M24, both drugs reduced relapse rate (p<0.001 for both) and stabilised disability. At M6 vs M0, both groups experienced significant atrophy of several areas in the cortex, deep GM nuclei and supratentorial WM. Significant bilateral cerebellar GM and WM atrophy occurred in fingolimod patients only. At M12 vs M6 and M24 vs M12, further supratentorial GM and WM atrophy occurred in both groups. Bilateral GM/WM cerebellar atrophy continued to progress in fingolimod patients only. Compared with natalizumab, fingolimod-treated patients showed a significant cerebellar GM/WM atrophy, mainly at M6 vs M0, but still occurring up to M24. Compared with fingolimod, natalizumab-treated patients had a small number of areas of GM atrophy in temporo-occipital regions at the different time-points.

CONCLUSIONS

Natalizumab and fingolimod are associated with heterogeneous temporal and regional patterns of GM and WM atrophy progression. Compared with natalizumab, fingolimod-treated patients experience accelerated GM and WM atrophy in the cerebellum, while both drugs show minimal regional volumetric differences in supratentorial regions.

Effects of Natalizumab and Fingolimod on Clinical, Cognitive, and Magnetic Resonance Imaging Measures in Multiple Sclerosis

Studies comparing the effects of natalizumab and fingolimod in relapsing-remitting multiple sclerosis (RRMS) are limited. We aimed to compare natalizumab and fingolimod effects on clinical, neuropsychological, and MRI measures in RRMS patients after 2 years of treatment. RRMS patients starting natalizumab (n = 30) or fingolimod (n = 25) underwent neurologic, neuropsychological, and brain MRI assessments at baseline, month (M) 6, M12, and M24. Volumes of lesions, brain, gray matter (GM), white matter (WM), and deep GM were measured. Fifteen healthy controls (HC) were also scanned at baseline and M24. Treatment groups were matched for baseline variables. At M24 versus baseline, both drugs reduced the relapse rate (p value < 0.001), stabilized disability, and improved cognitive function (fingolimod: p value = 0.03; natalizumab: p value = 0.01), without between-group differences. The natalizumab group had a higher proportion of freedom from MRI activity (67% vs 36%, p value = 0.02) and no evidence of disease activity-3 (NEDA-3) (57% vs 28%, p value = 0.04). At M24 vs M6, brain (- 0.35%, p value = 0.002 [fingolimod]; - 0.42%, p value < 0.001 [natalizumab]), GM (- 0.62%, p value < 0.001 [fingolimod]; - 0.64%, p value < 0.001 [natalizumab]), and WM (- 0.98%, p value < 0.001 [fingolimod]; - 0.99%, p value < 0.001 [natalizumab]) atrophy progressed at higher rates than in HC, but similarly between treatment groups, whereas only the natalizumab group showed deep GM atrophy (- 0.79%, p value = 0.02) (p value vs fingolimod not significant). In both groups, atrophy progression was correlated with lesion accumulation (r from - 0.49 to - 0.36, p values from 0.013 to 0.05), whereas no correlation was found between clinical and MRI changes. Natalizumab and fingolimod reduce disease activity and improve cognition in RRMS. Natalizumab seems superior to limit lesion accumulation, whereas both drugs similarly modify atrophy progression.

Measurement of Whole-Brain and Gray Matter Atrophy in Multiple Sclerosis: Assessment with MR Imaging

Purpose To compare available methods for whole-brain and gray matter (GM) atrophy estimation in multiple sclerosis (MS) in terms of repeatability (same magnetic resonance [MR] imaging unit) and reproducibility (different system/field strength) for their potential clinical applications. Materials and Methods The softwares ANTs-v1.9, CIVET-v2.1, FSL-SIENAX/SIENA-5.0.1, Icometrix-MSmetrix-1.7, and SPM-v12 were compared. This retrospective study, performed between March 2015 and March 2017, collected data from (a) eight simulated MR images and longitudinal data (2 weeks) from 10 healthy control subjects to assess the cross-sectional and longitudinal accuracy of atrophy measures, (b) test-retest MR images in 29 patients with MS acquired within the same day at different imaging unit field strengths/manufacturers to evaluate precision, and (c) longitudinal data (1 year) in 24 patients with MS for the agreement between methods. Tissue segmentation, image registration, and white matter (WM) lesion filling were also evaluated. Multiple paired t tests were used for comparisons. Results High values of accuracy (0.87-0.97) for whole-brain and GM volumes were found, with the lowest values for MSmetrix. ANTs showed the lowest mean error (0.02%) for whole-brain atrophy in healthy control subjects, with a coefficient of variation of 0.5%. SPM showed the smallest mean error (0.07%) and coefficient of variation (0.08%) for GM atrophy. Globally, good repeatability (P > .05) but poor reproducibility (P < .05) were found for all methods. WM lesion filling technique mainly affected ANTs, MSmetrix, and SPM results (P < .05). Conclusion From this comparison, it would be possible to select a software for atrophy measurement, depending on the requirements of the application (research center, clinical trial) and its goal (accuracy and repeatability or reproducibility). An improved reproducibility is required for clinical application. ^© RSNA, 2018 Online supplemental material is available for this article.

A Semiautomatic Method for Multiple Sclerosis Lesion Segmentation on Dual-Echo MR Imaging: Application in a Multicenter Context

BACKGROUND AND PURPOSE

The automatic segmentation of MS lesions could reduce time required for image processing together with inter- and intraoperator variability for research and clinical trials. A multicenter validation of a proposed semiautomatic method for hyperintense MS lesion segmentation on dual-echo MR imaging is presented.

MATERIALS AND METHODS

The classification technique used is based on a region-growing approach starting from manual lesion identification by an expert observer with a final segmentation-refinement step. The method was validated in a cohort of 52 patients with relapsing-remitting MS, with dual-echo images acquired in 6 different European centers.

RESULTS

We found a mathematic expression that made the optimization of the method independent of the need for a training dataset. The automatic segmentation was in good agreement with the manual segmentation (dice similarity coefficient = 0.62 and root mean square error = 2 mL). Assessment of the segmentation errors showed no significant differences in algorithm performance between the different MR scanner manufacturers (P > .05).

CONCLUSIONS

The method proved to be robust, and no center-specific training of the algorithm was required, offering the possibility for application in a clinical setting. Adoption of the method should lead to improved reliability and less operator time required for image analysis in research and clinical trials in MS.

Estimating Brain Lesion Volume Change in Multiple Sclerosis by Subtraction of Magnetic Resonance Images

BACKGROUND

Change in lesion volume over time, measured on brain magnetic resonance imaging (MRI) scans, is an important outcome measure for natural history studies and clinical trials in multiple sclerosis (MS).

PURPOSE

To develop and test image analysis methods for quantification of lesion volume change in order to improve reliability.

METHODS

The technique is based on registration and subtraction, and was evaluated in a cohort of 20 MS patients with dual-echo images acquired annually over a period of four years. The study protocol was approved by the local ethics review boards of participating centers, and all subjects gave written informed consent. The repeatability was compared to that obtained by the standard method for obtaining lesion volume change by evaluating the total volume at each time point, and then subtracting the volumes to obtain the difference.

RESULTS

Compared to the standard method, the subtraction method had improved intrarater correlation (0.95 and 0.72 for the subtraction method and the standard method, respectively) and interrater correlation (0.51 and 0.28, respectively). Furthermore, the mean time required to analyze the scans from one patient was 41 minutes for the subtraction method compared to 125 minutes for the standard method.

CONCLUSION

Use of the subtraction algorithm leads to improved reliability and lower operator fatigue in clinical trials and studies of the natural history of MS.