Regional cortical thickness in relapsing remitting multiple sclerosis: A multi-center study

Detection of diffusely abnormal white matter in multiple sclerosis on multiparametric brain MRI using semi-supervised deep learning

In addition to focal lesions, diffusely abnormal white matter (DAWM) is seen on brain MRI of multiple sclerosis (MS) patients and may represent early or distinct disease processes. The role of MRI-observed DAWM is understudied due to a lack of automated assessment methods. Supervised deep learning (DL) methods are highly capable in this domain, but require large sets of labeled data. To overcome this challenge, a DL-based network (DAWM-Net) was trained using semi-supervised learning on a limited set of labeled data for segmentation of DAWM, focal lesions, and normal-appearing brain tissues on multiparametric MRI. DAWM-Net segmentation performance was compared to a previous intensity thresholding-based method on an independent test set from expert consensus (N = 25). Segmentation overlap by Dice Similarity Coefficient (DSC) and Spearman correlation of DAWM volumes were assessed. DAWM-Net showed DSC > 0.93 for normal-appearing brain tissues and DSC > 0.81 for focal lesions. For DAWM-Net, the DAWM DSC was 0.49 ± 0.12 with a moderate volume correlation (ρ = 0.52, p < 0.01). The previous method showed lower DAWM DSC of 0.26 ± 0.08 and lacked a significant volume correlation (ρ = 0.23, p = 0.27). These results demonstrate the feasibility of DL-based DAWM auto-segmentation with semi-supervised learning. This tool may facilitate future investigation of the role of DAWM in MS.

U-fiber diffusion kurtosis and susceptibility characteristics in relapsing-remitting multiple sclerosis may be related to cognitive deficits and neurodegeneration

OBJECTIVES

To evaluate the diffusion kurtosis and susceptibility change in the U-fiber region of patients with relapsing-remitting multiple sclerosis (pwRRMS) and their correlations with cognitive status and degeneration.

MATERIALS AND METHODS

Mean kurtosis (MK), axial kurtosis (AK), radial kurtosis (RK), kurtosis fractional anisotropy (KFA), and the mean relative quantitative susceptibility mapping (mrQSM) values in the U-fiber region were compared between 49 pwRRMS and 48 healthy controls (HCs). The U-fiber were divided into upper and deeper groups based on the location. The whole brain volume, gray and white matter volume, and cortical thickness were obtained. The correlations between the mrQSM values, DKI-derived metrics in the U-fiber region and clinical scale scores, brain morphologic parameters were further investigated.

RESULTS

The decreased MK, AK, RK, KFA, and increased mrQSM values in U-fiber lesions (p < 0.001, FDR corrected), decreased RK, KFA, and increased mrQSM values in U-fiber non-lesions (p = 0.034, p < 0.001, p < 0.001, FDR corrected) were found in pwRRMS. There were differences in DKI-derived metrics and susceptibility values between the upper U-fiber region and the deeper one for U-fiber non-lesion areas of pwRRMS and HCs (p < 0.05), but not for U-fiber lesions in DKI-derived metrics. The DKI-derived metrics and susceptibility values were widely related with cognitive tests and brain atrophy.

CONCLUSION

RRMS patients show abnormal diffusion kurtosis and susceptibility characteristics in the U-fiber region, and these underlying tissue abnormalities are correlated with cognitive deficits and degeneration.

CLINICAL RELEVANCE STATEMENT

The macroscopic and microscopic tissue damages of U-fiber help to identify cognitive impairment and brain atrophy in multiple sclerosis and provide underlying pathophysiological mechanism.

KEY POINTS

• Diffusion kurtosis and susceptibility changes are present in the U-fiber region of multiple sclerosis. • There are gradients in diffusion kurtosis and susceptibility characteristics in the U-fiber region. • Tissue damages in the U-fiber region are correlated with cognitive impairment and brain atrophy.

Multiple sclerosis clinical forms classification with graph convolutional networks based on brain morphological connectivity

Multiple Sclerosis (MS) is an autoimmune disease that combines chronic inflammatory and neurodegenerative processes underlying different clinical forms of evolution, such as relapsing-remitting, secondary progressive, or primary progressive MS. This identification is usually performed by clinical evaluation at the diagnosis or during the course of the disease for the secondary progressive phase. In parallel, magnetic resonance imaging (MRI) analysis is a mandatory diagnostic complement. Identifying the clinical form from MR images is therefore a helpful and challenging task. Here, we propose a new approach for the automatic classification of MS forms based on conventional MRI (i.e., T1-weighted images) that are commonly used in clinical context. For this purpose, we investigated the morphological connectome features using graph based convolutional neural network. Our results obtained from the longitudinal study of 91 MS patients highlight the performance (F1-score) of this approach that is better than state-of-the-art as 3D convolutional neural networks. These results open the way for clinical applications such as disability correlation only using T1-weighted images.

Benefits of a mosaic approach for assessing cortical atrophy in individual multiple sclerosis patients

OBJECTIVE

Cortical gray matter (GM) atrophy plays a central role in multiple sclerosis (MS) pathology. However, it is not commonly assessed in clinical routine partly because a number of methodological problems hamper the development of a robust biomarker to quantify GM atrophy. In previous work, we have demonstrated the clinical utility of the "mosaic approach" (MAP) to assess individual GM atrophy in the motor neuron disease spectrum and frontotemporal dementia. In this study, we investigated the clinical utility of MAP in MS, comparing this novel biomarker to existing methods for computing GM atrophy in single patients. We contrasted the strategies based on correlations with established biomarkers reflecting MS disease burden.

METHODS

We analyzed T1-weighted MPRAGE magnetic resonance imaging data from 465 relapsing-remitting MS patients and 89 healthy controls. We inspected how variations of existing strategies to estimate individual GM atrophy ("standard approaches") as well as variations of MAP (i.e., different parcellation schemes) impact downstream analysis results, both on a group and an individual level. We interpreted individual cortical disease burden as single metric reflecting the fraction of significantly atrophic data points with respect to the control group. In addition, we evaluated the correlations to lesion volume (LV) and Expanded Disability Status Scale (EDSS).

RESULTS

We found that the MAP method yielded highest correlations with both LV and EDSS as compared to all other strategies. Although the parcellation resolution played a minor role in terms of absolute correlations with clinical variables, higher resolutions provided more clearly defined statistical brain maps which may facilitate clinical interpretability.

CONCLUSION

This study provides evidence that MAP yields high potential for a clinically relevant biomarker in MS, outperforming existing methods to compute cortical disease burden in single patients. Of note, MAP outputs brain maps illustrating individual cortical disease burden which can be directly interpreted in daily clinical routine.

Cortical thickness and cognition in older people with multiple sclerosis

BACKGROUND

The structural changes associated with cognitive performance in older people with multiple sclerosis (PwMS; age ≥ 50 years old) remain unknown.

OBJECTIVE

To determine the relationship between whole-brain (WBV), thalamus as the largest deep gray matter nuclei, and cortex-specific volume measurements with both cognitive impairment and numerical performance in older PwMS. The main hypothesis is that cognitive impairment (CI) in older PwMS is explained by cortical thinning in addition to global and thalamic neurodegenerative changes.

METHODS

A total of 101 older PwMS underwent cognitive and neuroimaging assessment. Cognitive assessment included tests established as sensitive in MS samples (Minimal Assessment of Cognitive Function in MS; MACFIMS), as well as those tests often utilized in Alzheimer's dementia studies (Wechsler's Memory Scale, Boston Naming Test, Visual Motor Integration and language). Cognitive impairment (CI) was based on -1.5 standard deviations in at least 2 cognitive domains (executive function, learning and memory, spatial processing, processing speed and working memory and language) when compared to healthy controls. WBV and thalamic volume were calculated using SIENAX/FIRST and cortical thickness using FreeSurfer. Differences in cortical thickness between CI and cognitively preserved (CP) were determined using age, sex, education, depression and WBV-adjusted analysis of covariance (ANCOVA). The relationship between domain-specific cognitive performance and cortical thickness was analyzed by linear regression models adjusted for age, sex, education, depression, WBV and thalamic volume. Benjamini-Hochberg-adjusted p-values lower than 0.05 were considered significant.

RESULTS

The average age of the study population was 62.6 (5.9) years old. After adjustment, CI PwMS had significantly thinner left fusiform (p = 0.0003), left inferior (p = 0.0032), left transverse (p = 0.0013), and bilateral superior temporal gyri (p = 0.002 and p = 0.0011) when compared to CP PwMS. After adjusting for age, sex, education, depression WBV, and thalamic volume, CI status was additionally predicted by the thickness of the left fusiform (p = 0.001) and left cuneus gyri (p = 0.004). After the adjustment, SDMT scores were additionally associated with left fusiform gyrus (p < 0.001) whereas letter-based verbal fluency performance with left pars opercularis gyrus (p < 0.001).

CONCLUSION

In addition to global and thalamic neurodegenerative changes, the presence of CI in older PwMS is additionally explained by the thickness of multiple cortical regions.

The effects of cytomegalovirus on brain structure following sport-related concussion

The neurotrophic herpes virus cytomegalovirus is a known cause of neuropathology in utero and in immunocompromised populations. Cytomegalovirus is reactivated by stress and inflammation, possibly explaining the emerging evidence linking it to subtle brain changes in the context of more minor disturbances of immune function. Even mild forms of traumatic brain injury, including sport-related concussion, are major physiological stressors that produce neuroinflammation. In theory, concussion could predispose to the reactivation of cytomegalovirus and amplify the effects of physical injury on brain structure. However, to our knowledge this hypothesis remains untested. This study evaluated the effect of cytomegalovirus serostatus on white and grey matter structure in a prospective study of athletes with concussion and matched contact-sport controls. Athletes who sustained concussion (n = 88) completed MRI at 1, 8, 15 and 45 days post-injury; matched uninjured athletes (n = 73) completed similar visits. Cytomegalovirus serostatus was determined by measuring serum IgG antibodies (n = 30 concussed athletes and n = 21 controls were seropositive). Inverse probability of treatment weighting was used to adjust for confounding factors between athletes with and without cytomegalovirus. White matter microstructure was assessed using diffusion kurtosis imaging metrics in regions previously shown to be sensitive to concussion. T1-weighted images were used to quantify mean cortical thickness and total surface area. Concussion-related symptoms, psychological distress, and serum concentration of C-reactive protein at 1 day post-injury were included as exploratory outcomes. Planned contrasts compared the effects of cytomegalovirus seropositivity in athletes with concussion and controls, separately. There was a significant effect of cytomegalovirus on axial and radial kurtosis in athletes with concussion but not controls. Cytomegalovirus positive athletes with concussion showed greater axial (P = 0.007, d = 0.44) and radial (P = 0.010, d = 0.41) kurtosis than cytomegalovirus negative athletes with concussion. Similarly, there was a significant association of cytomegalovirus with cortical thickness in athletes with concussion but not controls. Cytomegalovirus positive athletes with concussion had reduced mean cortical thickness of the right hemisphere (P = 0.009, d = 0.42) compared with cytomegalovirus negative athletes with concussion and showed a similar trend for the left hemisphere (P = 0.036, d = 0.33). There was no significant effect of cytomegalovirus on kurtosis fractional anisotropy, surface area, symptoms and C-reactive protein. The results raise the possibility that cytomegalovirus infection contributes to structural brain abnormalities in the aftermath of concussion perhaps via an amplification of concussion-associated neuroinflammation. More work is needed to identify the biological pathways underlying this process and to clarify the clinical relevance of this putative viral effect.

Links between Neuroanatomy and Neurophysiology with Turning Performance in People with Multiple Sclerosis

Multiple sclerosis is accompanied by decreased mobility and various adaptations affecting neural structure and function. Therefore, the purpose of this project was to understand how motor cortex thickness and corticospinal excitation and inhibition contribute to turning performance in healthy controls and people with multiple sclerosis. In total, 49 participants (23 controls, 26 multiple sclerosis) were included in the final analysis of this study. All participants were instructed to complete a series of turns while wearing wireless inertial sensors. Motor cortex gray matter thickness was measured via magnetic resonance imaging. Corticospinal excitation and inhibition were assessed via transcranial magnetic stimulation and electromyography place on the tibialis anterior muscles bilaterally. People with multiple sclerosis demonstrated reduced turning performance for a variety of turning variables. Further, we observed significant cortical thinning of the motor cortex in the multiple sclerosis group. People with multiple sclerosis demonstrated no significant reductions in excitatory neurotransmission, whereas a reduction in inhibitory activity was observed. Significant correlations were primarily observed in the multiple sclerosis group, demonstrating lateralization to the left hemisphere. The results showed that both cortical thickness and inhibitory activity were associated with turning performance in people with multiple sclerosis and may indicate that people with multiple sclerosis rely on different neural resources to perform dynamic movements typically associated with fall risk.

Cortical morphological changes in multiple sclerosis patients: a study of cortical thickness, sulcal depth, and local gyrification index

PURPOSE

Multiple sclerosis (MS) is a disease that progresses not only with demyelination but also with neurodegeneration. One of the goals of drug treatment in MS is to prevent neurodegeneration. Cortical thickness (CT), sulcal depth (SD), and local gyrification index (LGI) are indicators related to neurodegeneration. The aim of this study is to investigate changes in CT, SD, and LGI in patients with relapsing-remitting MS (RRMS).

METHODS

T1 images of 74 RRMS patients and 65 healthy controls were used. T1 hypointense areas in RRMS patients were corrected using fully automated methods. CT, SD, and LGI were calculated for each patient.

RESULTS

RRMS patients showed widespread cortical thinning, especially in bilateral temporoparietal areas, decreased SD in bilateral supramarginal gyrus, superior temporal gyrus, postcentral gyrus, and transverse temporal gyrus, and decreased LGI, especially in the left posterior cingulate gyrus and insula. The decrease in cortical thickness was associated with the number of attacks and lesion volume. EDSS was related to CT in the right lingual, inferior temporal, and fusiform gyrus. The LGI was correlated with T2 lesion volume in bilateral insula, with EDSS in the right insula and transverse and superior temporal gyri, and with the number of attacks in the right paracentral gyrus and pre-cuneus. However, SD did not show any correlation with EDSS, T2 lesion volume, or the number of attacks.

CONCLUSION

Our results demonstrate widespread cortical thinning, decreased sulcal depth in local areas, and decreased gyrification in folds in RRMS patients, which are related to clinical parameters.

How far MS lesion detection and segmentation are integrated into the clinical workflow? A systematic review

INTRODUCTION

Over the past few years, the deep learning community has developed and validated a plethora of tools for lesion detection and segmentation in Multiple Sclerosis (MS). However, there is an important gap between validating models technically and clinically. To this end, a six-step framework necessary for the development, validation, and integration of quantitative tools in the clinic was recently proposed under the name of the Quantitative Neuroradiology Initiative (QNI).

AIMS

Investigate to what extent automatic tools in MS fulfill the QNI framework necessary to integrate automated detection and segmentation into the clinical neuroradiology workflow.

METHODS

Adopting the systematic Cochrane literature review methodology, we screened and summarised published scientific articles that perform automatic MS lesions detection and segmentation. We categorised the retrieved studies based on their degree of fulfillment of QNI's six-steps, which include a tool's technical assessment, clinical validation, and integration.

RESULTS

We found 156 studies; 146/156 (94%) fullfilled the first QNI step, 155/156 (99%) the second, 8/156 (5%) the third, 3/156 (2%) the fourth, 5/156 (3%) the fifth and only one the sixth.

CONCLUSIONS

To date, little has been done to evaluate the clinical performance and the integration in the clinical workflow of available methods for MS lesion detection/segmentation. In addition, the socio-economic effects and the impact on patients' management of such tools remain almost unexplored.

A review on multiple sclerosis prognostic findings from imaging, inflammation, and mental health studies

Magnetic resonance imaging (MRI) of the brain is commonly used to detect where chronic and active lesions are in multiple sclerosis (MS). MRI is also extensively used as a tool to calculate and extrapolate brain health by way of volumetric analysis or advanced imaging techniques. In MS patients, psychiatric symptoms are common comorbidities, with depression being the main one. Even though these symptoms are a major determinant of quality of life in MS, they are often overlooked and undertreated. There has been evidence of bidirectional interactions between the course of MS and comorbid psychiatric symptoms. In order to mitigate disability progression in MS, treating psychiatric comorbidities should be investigated and optimized. New research for the prediction of disease states or phenotypes of disability have advanced, primarily due to new technologies and a better understanding of the aging brain.

Quantitative magnetic resonance imaging biomarkers for cortical pathology in multiple sclerosis at 7 T

Substantial cortical gray matter tissue damage, which correlates with clinical disease severity, has been revealed in multiple sclerosis (MS) using advanced magnetic resonance imaging (MRI) methods at 3 T and the use of ultra-high field, as well as in histopathology studies. While clinical assessment mainly focuses on lesions using T 1 - and T 2 -weighted MRI, quantitative MRI (qMRI) methods are capable of uncovering subtle microstructural changes. The aim of this ultra-high field study is to extract possible future MR biomarkers for the quantitative evaluation of regional cortical pathology. Because of their sensitivity to iron, myelin, and in part specifically to cortical demyelination, T 1 , T 2 , R 2 * , and susceptibility mapping were performed including two novel susceptibility markers; in addition, cortical thickness as well as the volumes of 34 cortical regions were computed. Data were acquired in 20 patients and 16 age- and sex-matched healthy controls. In 18 cortical regions, large to very large effect sizes (Cohen's d ≥ 1) and statistically significant differences in qMRI values between patients and controls were revealed compared with only four regions when using more standard MR measures, namely, volume and cortical thickness. Moreover, a decrease in all susceptibility contrasts ( χ , χ + , χ - ) and R 2 * values indicates that the role of cortical demyelination might outweigh inflammatory processes in the form of iron accumulation in cortical MS pathology, and might also indicate iron loss. A significant association between susceptibility contrasts as well as R 2 * of the caudal middle frontal gyrus and disease duration was found (adjusted R² : 0.602, p = 0.0011). Quantitative MRI parameters might be more sensitive towards regional cortical pathology compared with the use of conventional markers only and therefore may play a role in early detection of tissue damage in MS in the future.

The sequence of regional structural disconnectivity due to multiple sclerosis lesions

Objective

Prediction of disease progression is challenging in multiple sclerosis (MS) as the sequence of lesion development and retention of inflammation within a subset of chronic lesions is heterogeneous among patients. We investigated the sequence of lesion-related regional structural disconnectivity across the spectrum of disability and cognitive impairment in MS.

Methods

In a full cohort of 482 patients, the Expanded Disability Status Scale was used to classify patients into (i) no or mild vs (ii) moderate or severe disability groups. In 363 out of 482 patients, Quantitative Susceptibility Mapping was used to identify paramagnetic rim lesions (PRL), which are maintained by a rim of iron-laden innate immune cells. In 171 out of 482 patients, Brief International Cognitive Assessment was used to identify subjects with cognitive impairment. Network Modification Tool was used to estimate the regional structural disconnectivity due to MS lesions. Discriminative event-based modeling was applied to investigate the sequence of regional structural disconnectivity due to all representative lesions across the spectrum of disability and cognitive impairment.

Results

Structural disconnection in the ventral attention and subcortical networks was an early biomarker of moderate or severe disability. The earliest biomarkers of disability progression were structural disconnections due to PRL in the motor-related regions. Subcortical structural disconnection was an early biomarker of cognitive impairment.

Interpretation

MS lesion-related structural disconnections in the subcortex is an early biomarker for both disability and cognitive impairment in MS. PRL-related structural disconnection in the motor cortex may identify the patients at risk for moderate or severe disability in MS.

An Interpretable Machine Learning Model to Predict Cortical Atrophy in Multiple Sclerosis

To date, the relationship between central hallmarks of multiple sclerosis (MS), such as white matter (WM)/cortical demyelinated lesions and cortical gray matter atrophy, remains unclear. We investigated the interplay between cortical atrophy and individual lesion-type patterns that have recently emerged as new radiological markers of MS disease progression. We employed a machine learning model to predict mean cortical thinning in whole-brain and single hemispheres in 150 cortical regions using demographic and lesion-related characteristics, evaluated via an ultrahigh field (7 Tesla) MRI. We found that (i) volume and rimless (i.e., without a "rim" of iron-laden immune cells) WM lesions, patient age, and volume of intracortical lesions have the most predictive power; (ii) WM lesions are more important for prediction when their load is small, while cortical lesion load becomes more important as it increases; (iii) WM lesions play a greater role in the progression of atrophy during the latest stages of the disease. Our results highlight the intricacy of MS pathology across the whole brain. In turn, this calls for multivariate statistical analyses and mechanistic modeling techniques to understand the etiopathogenesis of lesions.

Brain magnetic resonance imaging surface-based analysis and cortical thickness measurement in relapsing remission multiple sclerosis

Background

Damage occurs in the brain tissue in MS which appears normal on standard conventional imaging (normal appearing brain tissue). This slow, evolving damage can be monitored by nonconventional advanced MR imaging techniques. New techniques for the measurement of cortical thickness have been validated against histological analysis and manual measurements. The aim of our study was to study the role of MRI surface-based analysis and cortical thickness measurement in the evaluation of patients with Relapsing Remitting Multiple Sclerosis and to detect if there is localized rather than generalized cortical atrophy in Multiple Sclerosis patients and correlating these findings with clinical data.

Results

30 patients and 30 healthy control were included in this study and they were subjected to cortical thickness analysis using MRI. The patients in our study showed decreased thickness of the precentral, paracentral, postcentral, posterior cingulate cortices and mean cortical thickness in both hemispheres when compared with the normal control group. Statistical analysis was significant (P value < 0.05) for the precentral, paracentral, postcentral, posterior cingulate cortices and mean cortical thickness in both hemispheres. On the other hand, statistical analysis was not significant (P value > 0.05) for other cortices. There was a significant negative correlation between the precentral, paracentral, postcentral, posterior cingulate cortices and mean cortical thickness in both hemispheres and EDSS scores with correlation coefficients ranging from − 0.9878 to − 0.7977.

Conclusions

MRI and post-processing segmentation analysis for cortical thickness is non-invasive imaging techniques that can increase the level of diagnostic confidence in diagnosis of MS patients and should be included as routine modality when evaluating patients with MS.

Classification of multiple sclerosis clinical profiles using machine learning and grey matter connectome

Purpose: The main goal of this study is to investigate the discrimination power of Grey Matter (GM) thickness connectome data between Multiple Sclerosis (MS) clinical profiles using statistical and Machine Learning (ML) methods. Materials and Methods: A dataset composed of 90 MS patients acquired at the MS clinic of Lyon Neurological Hospital was used for the analysis. Four MS profiles were considered, corresponding to Clinical Isolated Syndrome (CIS), Relapsing-Remitting MS (RRMS), Secondary Progressive MS (SPMS), and Primary Progressive MS (PPMS). Each patient was classified in one of these profiles by our neurologist and underwent longitudinal MRI examinations including T1-weighted image acquisition at each examination, from which the GM tissue was segmented and the cortical GM thickness measured. Following the GM parcellation using two different atlases (FSAverage and Glasser 2016), the morphological connectome was built and six global metrics (Betweenness Centrality (BC), Assortativity (r), Transitivity (T), Efficiency (E g ), Modularity (Q) and Density (D)) were extracted. Based on their connectivity metrics, MS profiles were first statistically compared and second, classified using four different learning machines (Logistic Regression, Random Forest, Support Vector Machine and AdaBoost), combined in a higher level ensemble model by majority voting. Finally, the impact of the GM spatial resolution on the MS clinical profiles classification was analyzed. Results: Using binary comparisons between the four MS clinical profiles, statistical differences and classification performances higher than 0.7 were observed. Good performances were obtained when comparing the two early clinical forms, RRMS and PPMS (F1 score of 0.86), and the two neurodegenerative profiles, PPMS and SPMS (F1 score of 0.72). When comparing the two atlases, slightly better performances were obtained with the Glasser 2016 atlas, especially between RRMS with PPMS (F1 score of 0.83), compared to the FSAverage atlas (F1 score of 0.69). Also, the thresholding value for graph binarization was investigated suggesting more informative graph properties in the percentile range between 0.6 and 0.8. Conclusion: An automated pipeline was proposed for the classification of MS clinical profiles using six global graph metrics extracted from the GM morphological connectome of MS patients. This work demonstrated that GM morphological connectivity data could provide good classification performances by combining four simple ML models, without the cost of long and complex MR techniques, such as MR diffusion, and/or deep learning architectures.

Machine learning to investigate superficial white matter integrity in early multiple sclerosis

BACKGROUND AND PURPOSE

This study aims todetermine the sensitivity of superficial white matter (SWM) integrity as a metric to distinguish early multiple sclerosis (MS) patients from healthy controls (HC).

METHODS

Fractional anisotropy and mean diffusivity (MD) values from SWM bundles across the cortex and major deep white matter (DWM) tracts were extracted from 29 early MS patients and 31 age- and sex-matched HC. Thickness of 68 cortical regions and resting-state functional-connectivity (RSFC) among them were calculated. The distribution of structural and functional metrics between groups were compared using Wilcoxon rank-sum test. Utilizing a machine learning method (adaptive boosting), 6 models were built based on: 1-SWM, 2-DWM, 3-SWM and DWM, 4-cortical thickness, or 5-RSFC measures. In model 6, all features from previous models were incorporated. The models were trained with nested 5-folds cross-validation. Area under the receiver operating characteristic curve (AUC_roc ) values were calculated to evaluate classification performance of each model. Permutation tests were used to compare the AUC_roc values.

RESULTS

Patients had higher MD in SWM bundles including insula, inferior frontal, orbitofrontal, superior and medial temporal, and pre- and post-central cortices (p < .05). No group differences were found for any other MRI metric. The model incorporating SWM and DWM features provided the best classification (AUC_roc = 0.75). The SWM model provided higher AUC_roc (0.74), compared to DWM (0.63), cortical thickness (0.67), RSFC (0.63), and all-features (0.68) models (p < .001 for all).

CONCLUSION

Our results reveal a non-random pattern of SWM abnormalities at early stages of MS even before pronounced structural and functional alterations emerge.

Serum neurofilament light chain correlates with myelin and axonal magnetic resonance imaging markers in multiple sclerosis

BACKGROUND

Neurofilaments are cytoskeletal proteins that are detectable in the blood after neuroaxonal injury. Multiple sclerosis (MS) disease progression, greater lesion volume, and brain atrophy are associated with higher levels of serum neurofilament light chain (NfL), but few studies have examined the relationship between NfL and advanced magnetic resonance imaging (MRI) measures related to myelin and axons. We assessed the relationship between serum NfL and brain MRI measures in a diverse group of MS participants.

METHODS AND MATERIALS

103 participants (20 clinically isolated syndrome, 33 relapsing-remitting, 30 secondary progressive, 20 primary progressive) underwent 3T MRI to obtain myelin water fraction (MWF), geometric mean T₂ (GMT₂), water content, T_1; high angular resolution diffusion imaging (HARDI)-derived axial diffusivity (AD), radial diffusivity (RD), fractional anisotropy (FA); diffusion basis spectrum imaging (DBSI)-derived AD, RD, FA; restricted, hindered, water and fiber fractions; and volume measurements of normalized brain, lesion, thalamic, deep gray matter (GM), and cortical thickness. Multiple linear regressions assessed the strength of association between serum NfL (dependent variable) and each MRI measure in whole brain (WB), normal appearing white matter (NAWM) and T₂ lesions (independent variables), while controlling for age, expanded disability status scale, and disease duration.

RESULTS

Serum NfL levels were significantly associated with metrics of axonal damage (FA: R²_WB-HARDI = 0.29, R²_NAWM-HARDI = 0.31, R²_NAWM-DBSI = 0.30, R²_Lesion-DBSI = 0.31; AD: R²_WB-HARDI=0.31), myelin damage (MWF: R²_WB = 0.29, R²_NAWM = 0.30, RD: R²_WB-HARDI = 0.32, R²_NAWM-HARDI = 0.34, R²_Lesion-DBSI = 0.30), edema and inflammation (T₁: R²_Lesion = 0.32; GMT₂: R²_WB = 0.31, R²_Lesion = 0.31), and cellularity (restricted fraction R²_WB = 0.30, R²_NAWM = 0.32) across the entire MS cohort. Higher serum NfL levels were associated with significantly higher T₂ lesion volume (R² = 0.35), lower brain structure volumes (thalamus R² = 0.31; deep GM R² = 0.33; normalized brain R² = 0.31), and smaller cortical thickness R² = 0.31).

CONCLUSION

The association between NfL and myelin MRI markers suggest that elevated serum NfL is a useful biomarker that reflects not only acute axonal damage, but also damage to myelin and inflammation, likely due to the known synergistic myelin-axon coupling relationship.

The Effect of Body Mass Index on Brain Volume and Cognitive Function in Relapsing-Remitting Multiple sclerosis: A CombiRx Secondary Analysis

Background

Multiple sclerosis (MS) is an autoimmune disease leading to physical, emotional and cognitive disability. High body mass index (BMI) may impact cognitive function and brain volume in MS. Yet, there is paucity of evidence addressing the impact of BMI on cognitive function and brain volume in MS.

Objectives

The purpose of this study was to examine the effects of BMI on normal appearing brain volume and cognitive function in patients with relapsing–remitting MS.

Methods

A secondary data analysis of the NIH CombiRx study was conducted. Multivariate regression and mixed model analyses were executed to analyze the effect of BMI on brain volume and cognitive function.

Results

The mean baseline age of the 768 participants was 38.2(SD = 9.4) years. 73% were female and 88.8% were Caucasian. The mean BMI was 28.8 kg/m²(SD = 6.7). The multivariate regression and mixed model analyses failed to show a clinical effect of BMI on brain volume and cognitive function.

Conclusion

BMI did not show an effect on cognitive function and brain volume among MS patients. Although there is increased interest in the effects of modifiable factors on the course of MS, the effects of BMI on brain volume and cognitive function are debatable and warrant further research.

ClinicalTrials.gov NCT00211887

Cortical Thickness and Serum NfL Explain Cognitive Dysfunction in Newly Diagnosed Patients With Multiple Sclerosis

Background and Objectives

To determine the relative importance of global or regional MRI and blood markers of neurodegeneration and neuroaxonal injury in predicting cognitive performance for recently diagnosed patients with multiple sclerosis (MS).

Methods

Thirty-five newly diagnosed patients with relapsing-remitting MS (RRMS) and 23 healthy controls (HCs) simultaneously completed a full clinical and neuropsychological assessment, structural brain MRI, and serum neurofilament light chain (sNfL) level test. Linear regression analyses were performed to determine which global or regional measures of gray matter (GM) atrophy and cortical thickness (CT), in combination with sNfL levels and clinical scores, are most strongly related to neuropsychological impairment.

Results

Compared with HCs, patients with MS showed bilateral thalamic GM atrophy (left, p = 0.033; right, p = 0.047) and diminished CT, particularly in the right superior and transverse temporal gyri (p = 0.045; p = 0.037). Regional atrophy failed to add predictive variance, whereas anxiety symptoms, sNfL, and global CT were the best predictors (R² = 0.404; p < 0.001) of cognitive outcomes, with temporal thickness accounting for greater variance in cognitive deficits than global CT.

Discussion

Thalamic GM atrophy and thinning in temporal regions represent a distinctive MRI trait in the early stages of MS. Although sNfL levels alone do not clearly differentiate HCs and patients with RRMS, in combination with global and regional CT, sNfL levels can better explain the presence of underlying cognitive deficits. Hence, cortical thinning and sNfL increases can be considered 2 parallel neurodegenerative markers in the pathogenesis of progression in newly diagnosed patients with MS.

Ensemble Learning for Multiple Sclerosis Disability Estimation Using Brain Structural Connectivity

BACKGROUND

Multiple Sclerosis (MS) is an autoimmune inflammatory disease of the central nervous system characterized by demyelination and neurodegeneration processes. It leads to different clinical courses and degrees of disability that need to be anticipated by the neurologist for personalized therapy. Recently, machine learning (ML) techniques have reached a high level of performance in brain disease diagnosis and/or prognosis, but the decision process of a trained ML system is typically non-transparent. Using brain structural connectivity data, a fully automatic ensemble learning model, augmented with an interpretable model, is proposed for the estimation of MS patients' disability, measured by the Expanded Disability Status Scale (EDSS).

METHOD

An ensemble of four boosting-based models (GBM, XGBoost, CatBoost, LightBoost) organized following a stacking generalization scheme, was developed using DTI-based structural connectivity data. In addition, an interpretable model based on conditional logistic regression was developed to explain the best performances in terms of white matter (WM) links for three classes of EDSS (Low, Medium, High).

RESULTS

The ensemble model reached excellent level of performance (RMSE of 0.92 ± 0.28) compared to single-based models and provided a better EDSS estimation using DTI-based structural connectivity data compared to conventional MRI measures associated with patient data (age, gender and disease duration). Used for interpretation of the estimation process, the counterfactual method showed the importance of certain brain networks, corresponding mainly to left hemisphere WM links, connecting the left superior temporal with the left posterior cingulate and the right precuneus gray matter regions, and the inter-hemispheric WM links constituting the corpus callosum. Also, a better accuracy estimation was found for the high disability class.

CONCLUSION

The combination of advanced ML models and sensitive techniques such as DTI-based structural connectivity demonstrated to be useful for the estimation of MS patients' disability and to point out the most important brain WM networks involved in disability.

Cortical surface thickness, subcortical volumes and disability between races in relapsing-remitting multiple sclerosis

BACKGROUND

The interplay between cortical surface thickness (CTh), subcortical volumes (SCV) and disability in patients with relapsing remitting multiple sclerosis (RRMS) is still not clear.

OBJECTIVE

To examine the relationship between CTh, SCV, and disability and investigate differences in CTh, SCV and disability between African Americans (AA) and Caucasian Americans (CA).

METHODS

Sixty-five RRMS (33AA, 32 CA) participants underwent Expanded Disability Status Scale and Multiple Sclerosis Functional Composite (MSFC) assessments, including timed 25-foot walk (T25FW), nine-hole peg test (9HPT) on dominant (D) and non-dominant hand (ND) and paced auditory serial addition test (PASAT-3). Symbol digit modalities test (SDMT) was also administered. All participants underwent 3T brain MRI. CTh was measured in the Frontal (FA), Parietal (PA), Temporal (TA), Occipital (OA), Cingulate (CA), and Global (GA) cortical surface areas (CSA). SCV measurements included Thalamus (TV), Caudate (CV), Putamen (PV), Pallidum (PaV), Hippocampus (HV), Amygdala (AV), Accumbens (AcV), Brain Stem (BSV), and Deep Gray Matter Total Volume (DGMTV). A general linear model with multivariate analysis (MANOVA) was used to determine the differences between the two cohorts (SPSS vs 25). Spearman rank correlation analysis was performed to investigate the relationship between CTh and MSFC.

RESULTS

AA have significantly decreased FA, PA, TA, GA CTh compared to CA (p = 0.004, p = 0.018, p = 0.013, p = 0.015, respectively). SCV measurements were not significantly different. Only in CA, the MSFC measures correlate significantly with regional CSA CTh. In both races and in the entire group, T25FW correlates with TV, PV, AV, AcV and DGMTV (p < 0.05). Only in AA and the entire cohort, PASAT-3 correlates with TV and AcV(p = 0.041, p = 0.006, p = 0.006, p = 0.000 respectively).

CONCLUSIONS

Differences in CSA CTh reinforce the different disease pathobiology between AA and CA. Regional CTh may represent a useful biomarker related to multi-domain disability only in CA, while in AA DGM injury might be a more important contributor to disability. Longitudinal, large-scale studies are warranted to confirm our findings.

Clinically detectable structural abnormalities in pediatric-onset multiple sclerosis: A large-scale magnetic resonance imaging analysis

BACKGROUND

Multiple Sclerosis is characterized by neural demyelination. Structural magnetic resonance imaging (MRI) provides soft tissue contrast, which forms the basis of techniques for extracting regional biomarkers across a participant's brain.

OBJECTIVES

To investigate the clinical presentation of multiple sclerosis in a large-scale MRI analysis that includes thorough consideration of extractable structural measurements (average and variability of regional cortical thicknesses, cortical surface measurements, and volumes).

METHODS

We performed a large-scale retrospective analysis of 370 T1 structural volumetric MRIs from 64 participants with multiple sclerosis and compared them with a large cohort of neurotypical participants, consisting of 993 MRIs from 988 participants. Regionally distributed measurements of cortical thickness (average and standard deviation) were extracted along with surface area, surface curvature, and volumetric measurements.

RESULTS

The largest observed finding involved regionally distributed reductions in average cortical thickness, with the parahippocampal region exhibiting the largest effect size, a finding that may be linked with known hippocampal atrophy in multiple sclerosis. Group-wise differences were also observed in terms of distributed volume, surface area, and surface curvature measurements.

CONCLUSIONS

Participants with pediatric-onset multiple sclerosis present clinically with a variety of structural abnormalities, including perirhinal cortex thickness abnormalities not previously reported in the literature.

Structural changes in the brain of patients with relapsing-remitting multiple sclerosis compared to controls: a MRI-based stereological study

BACKGROUND

Multiple sclerosis (MS) is an inflammatory autoimmune disorder of the central nervous system characterized by demyelination, inflammation, gliosis, and axonal loss. Nowadays, increasing scientific reports have focused on neurodegenerative processes and structural changes of the disease underlying pathogenesis.

AIM

The aim of this study is a structural analysis of brain magnetic resonance images (MRIs) in patients with relapsing-remitting multiple sclerosis (RRMS) comparing with normal individuals.

METHODS

This case-control study was carried out on MRIs of 20 patients with RRMS and 20 healthy controls in Zahedan, Iran. MR images with 4-mm slice thickness and 0.5-mm intervals in three anatomical planes (coronal, sagittal, axial) were acquired. Then, stereological parameters, including volume and volume density of different parts of the brain, based on Cavalries' point counting method were measured in both groups. Data analyses were performed using Mann-Whitney U and Pearson's correlation tests.

RESULTS

The results of the study showed that there were no significant differences in total brain, hemispheres, gray matter, and basal nuclei volume and volume density between the two groups (p ˃ 0.05). However, the left hemisphere, cerebellum, lateral ventricles, brainstem, corpus callosum, and white matter volume in RRMS patients were significantly lower than those in controls (p ˂ 0.05).

CONCLUSION

The findings showed that quantitative assessments based on stereological method on brain MRIs facilitate clarifying neuropathology of the disease. Also, it can be helpful as a simple index for following up the clinical situation and assessing therapeutic efficiency in MS patients. It may provide a precise treatment approach and justification of symptoms in patients with MS.

MAGNIMS consensus recommendations on the use of brain and spinal cord atrophy measures in clinical practice

Early evaluation of treatment response and prediction of disease evolution are key issues in the management of people with multiple sclerosis (MS). In the past 20 years, MRI has become the most useful paraclinical tool in both situations and is used clinically to assess the inflammatory component of the disease, particularly the presence and evolution of focal lesions - the pathological hallmark of MS. However, diffuse neurodegenerative processes that are at least partly independent of inflammatory mechanisms can develop early in people with MS and are closely related to disability. The effects of these neurodegenerative processes at a macroscopic level can be quantified by estimation of brain and spinal cord atrophy with MRI. MRI measurements of atrophy in MS have also been proposed as a complementary approach to lesion assessment to facilitate the prediction of clinical outcomes and to assess treatment responses. In this Consensus statement, the Magnetic Resonance Imaging in MS (MAGNIMS) study group critically review the application of brain and spinal cord atrophy in clinical practice in the management of MS, considering the role of atrophy measures in prognosis and treatment monitoring and the barriers to clinical use of these measures. On the basis of this review, the group makes consensus statements and recommendations for future research.

Longitudinal patterns of cortical thinning in multiple sclerosis

In multiple sclerosis (MS), cortical atrophy is correlated with clinical and neuropsychological measures. We aimed to examine the differences in the temporospatial evolution of cortical thickness (CTh) between MS‐subtypes and to study the association of CTh with T2‐weighted white matter lesions (T2LV) and clinical progression. Two hundred and forty‐three MS patients (180 relapsing–remitting [RRMS], 51 secondary‐progressive [SPMS], and 12 primary‐progressive [PPMS]) underwent annual clinical (incl. expanded disability status scale [EDSS]) and MRI‐examinations over 6 years. T2LV and CTh were measured. CTh did not differ between MS‐subgroups. Higher total T2LV was associated with extended bilateral CTh‐reduction on average, but did not correlate with CTh‐changes over time. In RRMS, CTh‐ and EDSS‐changes over time were negatively correlated in large bilateral prefrontal, frontal, parietal, temporal, and occipital areas. In SPMS, CTh was not associated with the EDSS. In PPMS, CTh‐ and EDSS‐changes over time were correlated in small clusters predominantly in left parietal areas. Increase of brain lesion load does not lead to an immediate CTh‐reduction. Although CTh did not differ between MS‐subtypes, a dissociation in the correlation between CTh‐ and EDSS‐changes over time between RRMS and progressive‐MS was shown, possibly underlining the contribution of subcortical pathology to clinical progression in progressive‐MS.

Gray matter atrophy cannot be fully explained by white matter damage in patients with MS

BACKGROUND

Source-based morphometry (SBM) was recently used for non-random "patterns" of gray matter (GM) atrophy or white matter (WM) microstructural damage.

OBJECTIVE

To assess whether and to what extent such patterns may be inter-related in MS.

METHODS

SBM was applied to images of GM concentration and fractional anisotropy (FA) in MS patients (n = 41, median EDSS = 1) and normal controls (NC, n = 28). The same procedure was repeated on an independent and similar data set (39 MS patients and 13 NC).

RESULTS

We found in MS patterns of GM atrophy and reduced FA (p < 0.05, corrected). Deep GM atrophy was mostly (70%) explained by lesion load in projection tracts and lower FA in posterior corona radiata and thalamic radiation. By contrast, sensorimotor and posterior cortex atrophy was less (50%) dependent from WM damage. All patterns correlated with EDSS (r from -0.33 to -0.56, p < 0.03) while the only cognition-related correlation was between posterior GM atrophy pattern and processing speed (r = 0.45, p = 0.014). Reliability analysis showed similar results.

CONCLUSION

In relatively early MS, we found a close link between deep GM atrophy pattern and WM damage while sensorimotor and posterior cortex patterns were partially independent from WM damage and perhaps related to primary mechanisms. Patterns were clinically relevant.

Are multi-contrast magnetic resonance images necessary for segmenting multiple sclerosis brains? A large cohort study based on deep learning

BACKGROUND

Magnetic resonance images with multiple contrasts or sequences are commonly used for segmenting brain tissues, including lesions, in multiple sclerosis (MS). However, acquisition of images with multiple contrasts increases the scan time and complexity of the analysis, possibly introducing factors that could compromise segmentation quality.

OBJECTIVE

To investigate the effect of various combinations of multi-contrast images as input on the segmented volumes of gray (GM) and white matter (WM), cerebrospinal fluid (CSF), and lesions using a deep neural network.

METHODS

U-net, a fully convolutional neural network was used to automatically segment GM, WM, CSF, and lesions in 1000 MS patients. The input to the network consisted of 15 combinations of FLAIR, T1-, T2-, and proton density-weighted images. The Dice similarity coefficient (DSC) was evaluated to assess the segmentation performance. For lesions, true positive rate (TPR) and false positive rate (FPR) were also evaluated. In addition, the effect of lesion size on lesion segmentation was investigated.

RESULTS

Highest DSC was observed for all the tissue volumes, including lesions, when the input was combination of all four image contrasts. All other input combinations that included FLAIR also provided high DSC for all tissue classes. However, the quality of lesion segmentation showed strong dependence on the input images. The DSC and TPR values for inputs with the four contrast combination and FLAIR alone were very similar, but FLAIR showed a moderately higher FPR for lesion size <100 μl. For lesions smaller than 20 μl all image combinations resulted in poor performance. The segmentation quality improved with lesion size.

CONCLUSIONS

Best performance for segmented tissue volumes was obtained with all four image contrasts as the input, and comparable performance was attainable with FLAIR only as the input, albeit with a moderate increase in FPR for small lesions. This implies that acquisition of only FLAIR images provides satisfactory tissue segmentation. Lesion segmentation was poor for very small lesions and improved rapidly with lesion size.

Deep Learning for Predicting Enhancing Lesions in Multiple Sclerosis from Noncontrast MRI

Background Enhancing lesions on MRI scans obtained after contrast material administration are commonly thought to represent disease activity in multiple sclerosis (MS); it is desirable to develop methods that can predict enhancing lesions without the use of contrast material. Purpose To evaluate whether deep learning can predict enhancing lesions on MRI scans obtained without the use of contrast material. Materials and Methods This study involved prospective analysis of existing MRI data. A convolutional neural network was used for classification of enhancing lesions on unenhanced MRI scans. This classification was performed for each slice, and the slice scores were combined by using a fully connected network to produce participant-wise predictions. The network input consisted of 1970 multiparametric MRI scans from 1008 patients recruited from 2005 to 2009. Enhanced lesions on postcontrast T1-weighted images served as the ground truth. The network performance was assessed by using fivefold cross-validation. Statistical analysis of the network performance included calculation of lesion detection rates and areas under the receiver operating characteristic curve (AUCs). Results MRI scans from 1008 participants (mean age, 37.7 years ± 9.7; 730 women) were analyzed. At least one enhancing lesion was observed in 519 participants. The sensitivity and specificity averaged across the five test sets were 78% ± 4.3 and 73% ± 2.7, respectively, for slice-wise prediction. The corresponding participant-wise values were 72% ± 9.0 and 70% ± 6.3. The diagnostic performances (AUCs) were 0.82 ± 0.02 and 0.75 ± 0.03 for slice-wise and participant-wise enhancement prediction, respectively. Conclusion Deep learning used with conventional MRI identified enhanced lesions in multiple sclerosis from images from unenhanced multiparametric MRI with moderate to high accuracy. © RSNA, 2019.

Effects of Optic Neuritis, T2 Lesions, and Microstructural Diffusion Integrity in the Visual Pathway on Cortical Thickness in Pediatric-Onset Multiple Sclerosis

BACKGROUND AND PURPOSE

Pediatric-onset multiple sclerosis (POMS) is associated with focal inflammatory lesions and the loss of cortical and deep gray matter. Optic neuritis (ON) and white matter (WM) lesions in the visual pathway can directly contribute to visual cortical mantle thinning. We determine the relative contributions of MS insult on anterior and posterior visual pathway integrity.

METHODS

High- and low-contrast visual acuity, optical coherence tomography (OCT), and 3T MRI scans were obtained from 20 POMS patients (10 with remote ON) and 22 age- and sex-matched healthy controls. Cortical mantle thickness was measured using FreeSurfer. Fractional anisotropy (FA) and mean diffusivity were calculated for postchiasmal optic radiations (with and without WM lesions). Groups were compared using Student's t-test (adjusted for multiple comparisons), and simple linear regression was used to investigate interrelationships between measures.

RESULTS

Mean cortical thickness of the whole brain was reduced in patients (2.49 mm) versus controls (2.58 mm, P = .0432) and in the visual cortex (2.07 mm vs. 2.17 mm, P = .0059), although the foveal confluence was spared. Mean FA of the optic radiations was reduced in POMS (.40) versus controls (.43, P = .0042) and correlated with visual cortical mantle thickness in POMS (P = .017). Visual acuity, OCT measures, and lesion volumes in the optic radiations were not associated with cortical mantle thickness.

CONCLUSIONS

POMS negatively impacts the integrity of the anterior visual pathway, but it is the loss of WM integrity that drives anterograde loss of the cortical mantle. Preserved visual acuity and foveal sparing imply some degree of functional and structural resilience.

Brain and lesion segmentation in multiple sclerosis using fully convolutional neural networks: A large-scale study

OBJECTIVE

To investigate the performance of deep learning (DL) based on fully convolutional neural network (FCNN) in segmenting brain tissues in a large cohort of multiple sclerosis (MS) patients.

METHODS

We developed a FCNN model to segment brain tissues, including T2-hyperintense MS lesions. The training, validation, and testing of FCNN were based on ~1000 magnetic resonance imaging (MRI) datasets acquired on relapsing-remitting MS patients, as a part of a phase 3 randomized clinical trial. Multimodal MRI data (dual-echo, FLAIR, and T1-weighted images) served as input to the network. Expert validated segmentation was used as the target for training the FCNN. We cross-validated our results using the leave-one-center-out approach.

RESULTS

We observed a high average (95% confidence limits) Dice similarity coefficient for all the segmented tissues: 0.95 (0.92-0.98) for white matter, 0.96 (0.93-0.98) for gray matter, 0.99 (0.98-0.99) for cerebrospinal fluid, and 0.82 (0.63-1.0) for T2 lesions. High correlations between the DL segmented tissue volumes and ground truth were observed (R² > 0.92 for all tissues). The cross validation showed consistent results across the centers for all tissues.

CONCLUSION

The results from this large-scale study suggest that deep FCNN can automatically segment MS brain tissues, including lesions, with high accuracy.

Neuroanatomical deficits shared by youth with autism spectrum disorders and psychotic disorders

Autism spectrum disorders (ASD) and early-onset psychosis (EOP) are neurodevelopmental disorders that share genetic, clinical and cognitive facets; it is unclear if these disorders also share spatially overlapping cortical thickness (CT) and surface area (SA) abnormalities. MRI scans of 30 ASD, 29 patients with early-onset first-episode psychosis (EO-FEP) and 26 typically developing controls (TD) (age range 10-18 years) were analyzed by the FreeSurfer suite to calculate vertex-wise estimates of CT, SA, and cortical volume. Two publicly available datasets of ASD and EOP (age range 7-18 years and 5-17 years, respectively) were used for replication analysis. ASD and EO-FEP had spatially overlapping areas of cortical thinning and reduced SA in the bilateral insula (all p's < .00002); 37% of all left insular vertices presenting with significant cortical thinning and 20% (left insula) and 61% (right insula) of insular vertices displaying decreased SA overlapped across both disorders. In both disorders, SA deficits contributed more to cortical volume decreases than reductions in CT did. This finding, as well as the novel finding of an absence of spatial overlap (for ASD) or marginal overlap (for EOP) of deficits in CT and SA, was replicated in the two nonoverlapping independent samples. The insula appears to be a region with transdiagnostic vulnerability for deficits in CT and SA. The finding of nonexistent or small spatial overlap between CT and SA deficits in young people with ASD and psychosis may point to the involvement of common aberrant early neurodevelopmental mechanisms in their pathophysiology.

Automated image quality evaluation of structural brain MRI using an ensemble of deep learning networks

BACKGROUND

Deep learning (DL) is a promising methodology for automatic detection of abnormalities in brain MRI.

PURPOSE

To automatically evaluate the quality of multicenter structural brain MRI images using an ensemble DL model based on deep convolutional neural networks (DCNNs).

STUDY TYPE

Retrospective.

POPULATION

The study included 1064 brain images of autism patients and healthy controls from the Autism Brain Imaging Data Exchange (ABIDE) database. MRI data from 110 multiple sclerosis patients from the CombiRx study were included for independent testing.

SEQUENCE

T₁ -weighted MR brain images acquired at 3T.

ASSESSMENT

The ABIDE data were separated into training (60%), validation (20%), and testing (20%) sets. The ensemble DL model combined the results from three cascaded networks trained separately on the three MRI image planes (axial, coronal, and sagittal). Each cascaded network consists of a DCNN followed by a fully connected network. The quality of image slices from each plane was evaluated by the DCNN and the resultant image scores were combined into a volumewise quality rating using the fully connected network. The DL predicted ratings were compared with manual quality evaluation by two experts.

STATISTICAL TESTS

Receiver operating characteristic (ROC) curve, area under ROC curve (AUC), sensitivity, specificity, accuracy, and positive (PPV) and negative (NPV) predictive values.

RESULTS

The AUC, sensitivity, specificity, accuracy, PPV, and NPV for image quality evaluation of the ABIDE test set using the ensemble model were 0.90, 0.77, 0.85, 0.84, 0.42, and 0.96, respectively. On the CombiRx set the same model achieved performance of 0.71, 0.41, 0.84, 0.73, 0.48, and 0.80.

DATA CONCLUSION

This study demonstrated the high accuracy of DL in evaluating image quality of structural brain MRI in multicenter studies.

LEVEL OF EVIDENCE

3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;50:1260-1267.

Relationships in Ongoing Structural Maintenances of the Two Cerebral Cortices of an Individual Brain

A human brain has separate left and right cerebral cortices, each of which must be continuously structurally maintained during adulthood. There is no understanding of how ongoing structural maintenances of separate parts of a mature individual brain, including the 2 cortices, are related. To explore this issue, this study used an unconventional N-of-1 magnetic resonance imaging time-series paradigm to identify relationships between maintenances of structural thicknesses of the 2 cortices in an adult human brain over week intervals for 6 months. The results suggest that maintenances of left and right cortical thicknesses were symmetrically related in some, but asymmetrically related in other, respects. For matched times, thickness magnitudes and variations on the 2 sides were positively correlated and appeared to reflect maintenance symmetry. Maintenance relationships also extended from earlier to later times with temporal continuity and apparent “if-then” contingencies which were reflected in symmetry and asymmetry dynamics spanning 1- to 2-week periods. The findings suggest concepts of individual brain cortical maintenance symmetry, asymmetry, and temporal continuity dynamics that have not been previously recognized. They have implications for defining cortical maintenance traits or states and for development of N-of-1 precision medicine paradigms that can contribute to understanding individual brain health.

Neuroimaging Studies Illustrate the Commonalities Between Ageing and Brain Diseases

The lack of specificity in neuroimaging studies of neurological and psychiatric diseases suggests that these different diseases have more in common than is generally considered. Potentially, features that are secondary effects of different pathological processes may share common neurobiological underpinnings. Intriguingly, many of these mechanisms are also observed in studies of normal (i.e., non-pathological) brain ageing. Different brain diseases may be causing premature or accelerated ageing to the brain, an idea that is supported by a line of "brain ageing" research that combines neuroimaging data with machine learning analysis. In reviewing this field, I conclude that such observations could have important implications, suggesting that we should shift experimental paradigm: away from characterizing the average case-control brain differences resulting from a disease toward methods that place individuals in their age-appropriate context. This will also lead naturally to clinical applications, whereby neuroimaging can contribute to a personalized-medicine approach to improve brain health.

Urgent challenges in quantification and interpretation of brain grey matter atrophy in individual MS patients using MRI

Atrophy of the brain grey matter (GM) is an accepted and important feature of multiple sclerosis (MS). However, its accurate measurement is hampered by various technical, pathological and physiological factors. As a consequence, it is challenging to investigate the role of GM atrophy in the disease process as well as the effect of treatments that aim to reduce neurodegeneration. In this paper we discuss the most important challenges currently hampering the measurement and interpretation of GM atrophy in MS. The focus is on measurements that are obtained in individual patients rather than on group analysis methods, because of their importance in clinical trials and ultimately in clinical care. We discuss the sources and possible solutions of the current challenges, and provide recommendations to achieve reliable measurement and interpretation of brain GM atrophy in MS.

A Fully Automated Pipeline for Normative Atrophy in Patients with Neurodegenerative Disease

Introduction

Volumetric image analysis to detect progressive brain tissue loss in patients with multiple sclerosis (MS) has recently been suggested as a promising marker for "no evidence of disease activity." Software packages for longitudinal whole-brain volume analysis in individual patients are already in clinical use; however, most of these methods have omitted region-based analysis. Here, we suggest a fully automatic analysis pipeline based on the free software packages FSL and FreeSurfer.

Materials and methods

Fifty-five T1-weighted magnetic resonance imaging (MRI) datasets of five patients with confirmed relapsing-remitting MS and mild to moderate disability were longitudinally analyzed compared to a morphometric reference database of 323 healthy controls (HCs). After lesion filling, the volumes of brain segmentations and morphometric parameters of cortical parcellations were automatically screened for global and regional abnormalities. Error margins and artifact probabilities of regional morphometric parameters were estimated. Linear models were fitted to the series of follow-up MRIs and checked for consistency with cross-sectional aging in HCs.

Results

As compared to leave-one-out cross-validation in a subset of the control dataset, anomaly detection rates were highly elevated in MRIs of two patients. We detected progressive volume changes that were stronger than expected compared to normal aging in 4/5 patients. In individual patients, we also identified stronger than expected regional decreases of subcortical gray matter, of cortical thickness, and areas of reducing gray-white contrast over time.

Conclusion

Statistical comparison with a large normative database may provide complementary and rater independent quantitative information about regional morphological changes related to disease progression or drug-related disease modification in individual patients. Regional volume loss may also be detected in clinically stable patients.

Focal cortical thinning in patients with stable relapsing-remitting multiple sclerosis: cross-sectional-based novel estimation of gray matter kinetics

PURPOSE

The aim of our study is to identify radiological patterns of cortical gray matter atrophy (CGMA) that correlate with disease duration in patients with relapsing-remitting multiple sclerosis (RRMS).

METHODS

RRMS patients were randomly selected from the Sheba Multiple Sclerosis (MS) center computerized data registry based on stratification of disease duration up to 10 years. Patients were scanned by 3.0 T (Signa, GE) MRI, using a T1 weighted 3D high resolution, FSPGR, MS protocol. Neurological disability was assessed by the Expanded Disability Status Scale (EDSS). FreeSurfer was used to obtain brain volumetric segmentation and to perform cortical thickness surface-based analysis. Clusters of change in cortical thickness with correlation to disease duration were produced.

RESULTS

Two hundred seventy-one RRMS patients, mean ± SD age 33.0 ± 7.0 years, EDSS 1.6 ± 1.2, disease duration 5.0 ± 3.4 years. Cortical thickness analysis demonstrated focal areas of cerebral thinning that correlated with disease duration. Seven clusters accounting for 11.7% of the left hemisphere surface and eight clusters accounting for 10.6% of the right hemisphere surface were identified, with cluster-wise probability of p < 0.002 and p < 0.02, respectively.The clusters included bilateral involvement of areas within the cingulate, precentral, postcentral, paracentral, superior-parietal, superior-frontal gyri and insular cortex. Mean and cluster-wise cortical thickness negatively correlated with EDSS score, p < 0.001, with stronger Spearman rho for cluster-wise measurements.

CONCLUSIONS

We identified CGMA patterns in sensitive brain regions which give insight and better understanding of the progression of cortical gray matter loss in relation to dissemination in space and time. These patterns may serve as markers to modulate therapeutic interventions to improve the management of MS patients.

Can resistance training impact MRI outcomes in relapsing-remitting multiple sclerosis?

BACKGROUND

Multiple sclerosis (MS) is characterised by accelerated brain atrophy, which relates to disease progression. Previous research shows that progressive resistance training (PRT) can counteract brain atrophy in other populations.

OBJECTIVE

To evaluate the effects of PRT by magnetic resonance imaging (MRI) and clinical measures of disease progression in people with MS.

METHODS

This study was a 24-week randomised controlled cross-over trial, including a Training ( n = 18, 24 weeks of PRT followed by self-guided physical activity) and Waitlist group ( n = 17, 24 weeks of habitual lifestyle followed by PRT). Assessments included disability measures and MRI (lesion load, global brain volume, percentage brain volume change (PBVC) and cortical thickness).

RESULTS

While the MS Functional Composite score improved, Expanded Disability Status Scale, lesion load and global brain volumes did not differ between groups. PBVC tended to differ between groups and higher absolute cortical thickness values were observed in 19 of 74 investigated cortical regions after PRT. Observed changes were confirmed and reproduced when comparing relative cortical thickness changes between groups for four areas: anterior cingulate gyrus, temporal pole, orbital sulcus and inferior temporal sulcus.

CONCLUSION

PRT seem to induce an increase in cortical thickness, indicating that PRT have a neuroprotective or even neuroregenerative effect in relapsing-remitting MS.

Critical analysis on the present methods for brain volume measurements in multiple sclerosis

Objective

The treatment of multiple sclerosis (MS) has quickly evolved from a time when controlling clinical relapses would suffice, to the present day, when complete disease control is expected. Measurement of brain volume is still at an early stage to be indicative of therapeutic decisions in MS.

Methods

This paper provides a critical review of potential biases and artifacts in brain measurement in the follow-up of patients with MS.

Results

Clinical conditions (such as hydration or ovulation), time of the day, type of magnetic resonance machine (manufacturer and potency), brain volume artifacts and different platforms for volumetric assessment of the brain can induce variations that exceed the acceptable physiological rate of annual loss of brain volume.

Conclusion

Although potentially extremely valuable, brain volume measurement still has to be regarded with caution in MS.

Volume versus surface-based cortical thickness measurements: A comparative study with healthy controls and multiple sclerosis patients

The cerebral cortex is a highly folded outer layer of grey matter tissue that plays a key role in cognitive functions. In part, alterations of the cortex during development and disease can be captured by measuring the cortical thickness across the whole brain. Available software tools differ with regard to labor intensity and computational demands. In this study, we compared the computational anatomy toolbox (CAT), a recently proposed volume-based tool, with the well-established surface-based tool FreeSurfer. We observed that overall thickness measures were highly inter-correlated, although thickness estimates were systematically lower in CAT than in FreeSurfer. Comparison of multiple sclerosis (MS) patients with age-matched healthy control subjects showed highly comparable clusters of MS-related thinning for both methods. Likewise, both methods yielded comparable clusters of age-related cortical thinning, although correlations between age and average cortical thickness were stronger for FreeSurfer. Our data suggest that, for the analysis of cortical thickness, the volume-based CAT tool can be regarded a considerable alternative to the well-established surface-based FreeSurfer tool.

MRI Analysis of White Matter Myelin Water Content in Multiple Sclerosis: A Novel Approach Applied to Finding Correlates of Cortical Thinning

A novel lesion-mask free method based on a gamma mixture model was applied to myelin water fraction (MWF) maps to estimate the association between cortical thickness and myelin content, and how it differs between relapsing-remitting (RRMS) and secondary-progressive multiple sclerosis (SPMS) groups (135 and 23 patients, respectively). It was compared to an approach based on lesion masks. The gamma mixture distribution of whole brain, white matter (WM) MWF was characterized with three variables: the mode (most frequent value) m₁ of the gamma component shown to relate to lesion, the mode m₂ of the component shown to be associated with normal appearing (NA) WM, and the mixing ratio (λ) between the two distributions. The lesion-mask approach relied on the mean MWF within lesion and within NAWM. A multivariate regression analysis was carried out to find the best predictors of cortical thickness for each group and for each approach. The gamma-mixture method was shown to outperform the lesion-mask approach in terms of adjusted R², both for the RRMS and SPMS groups. The predictors of the final gamma-mixture models were found to be m₁ (β = 1.56, p < 0.005), λ (β = −0.30, p < 0.0005) and age (β = −0.0031, p < 0.005) for the RRMS group (adjusted R² = 0.16), and m₂ (β = 4.72, p < 0.0005) for the SPMS group (adjusted R² = 0.45). Further, a DICE coefficient analysis demonstrated that the lesion mask had more overlap to an ROI associated with m₁, than to an ROI associated with m₂ (p < 0.00001), and vice versa for the NAWM mask (p < 0.00001). These results suggest that during the relapsing phase, focal WM damage is associated with cortical thinning, yet in SPMS patients, global WM deterioration has a much stronger influence on secondary degeneration. Through these findings, we demonstrate the potential contribution of myelin loss on neuronal degeneration at different disease stages and the usefulness of our statistical reduction technique which is not affected by the typical bias associated with approaches based on lesion masks.

Estimating multimodal brain connectivity in multiple sclerosis: an exploratory factor analysis

Graph-theoretical approaches have become a popular way to model brain data collected using magnetic resonance imaging (MRI), both from the structural and the functional perspectives. In structural networks, tract-based mapping allows to model different aspects of brain structures by means of the specific characteristics of the different MRI modalities. However, there has been little effort to join the information carried by each modality and to understand what level of common variance is shown in these data. In this paper, we proposed a combined approach based on graph theory and factor analysis to model magnetization transfer and microstructural properties in 18 relapsing remitting multiple sclerosis (RRMS) patients and 17 healthy controls. After defining the common factors and outlining their relationships with MRI data, we evaluated between-group differences using global and local graph measures. The results showed that one common factor describes brain structures in terms of myelin and global integrity, and such factor is able to highlight specific between-group differences.

Imaging as an Outcome Measure in Multiple Sclerosis

Multiple sclerosis (MS) is a chronic demyelinating disease of the central nervous system. Magnetic resonance imaging (MRI) is sensitive to lesion formation both in the brain and spinal cord. Imaging plays a prominent role in the diagnosis and monitoring of MS. Over a dozen anti-inflammatory therapies are approved for MS and the development of many of these medications was made possible through the use of contrast-enhancing lesions on MRI as a phase II outcome. A similar phase II outcome method for the neurodegeneration that underlies progressive courses of the disease is still unavailable. Although magnetic resonance is an invaluable tool for the diagnosis and monitoring of treatment effects in MS, several imaging barriers still exist. In general, MRI is less sensitive to gray matter lesions, lacks pathological specificity, and does not provide quantitative data easily. Several advanced imaging methods including diffusion tensor imaging, magnetization transfer, functional MRI, myelin water fraction imaging, ultra-high field MRI, positron emission tomography, and optical coherence tomography of the retina study promising ways of overcoming the difficulties in MS imaging.

Universality in human cortical folding in health and disease

The folding of the cortex in mammalian brains across species has recently been shown to follow a universal scaling law that can be derived from a simple physics model. However, it was yet to be determined whether this law also applies to the morphological diversity of different individuals in a single species, in particular with respect to factors, such as age, sex, and disease. To this end, we derived and investigated the cortical morphology from magnetic resonance images (MRIs) of over 1,000 healthy human subjects from three independent public databases. Our results show that all three MRI datasets follow the scaling law obtained from the comparative neuroanatomical data, which strengthens the case for the existence of a common mechanism for cortical folding. Additionally, for comparable age groups, both male and female brains scale in exactly the same way, despite systematic differences in size and folding. Furthermore, age introduces a systematic shift in the offset of the scaling law. In the model, this shift can be interpreted as changes in the mechanical forces acting on the cortex. We also applied this analysis to a dataset derived from comparable cohorts of Alzheimer's disease patients and healthy subjects of similar age. We show a systematically lower offset and a possible change in the exponent for Alzheimer's disease subjects compared with the control cohort. Finally, we discuss implications of the changes in offset and exponent in the data and relate it to existing literature. We, thus, provide a possible mechanistic link between previously independent observations.

Differential Impact of Multiple Sclerosis on Cortical and Deep Gray Matter Structures in African Americans and Caucasian Americans

BACKGROUND

African Americans with multiple sclerosis (AAwMS) have different disease phenotypes when compared to Caucasians Americans with MS (CAwMS). The pathologic basis of this difference in disease presentation is unknown.

METHODS

Fifty-Four AAwMS and 54 CAwMS were appropriately matched for age, gender, treatment duration, and disease duration. FreeSurfer was used to segment brain white matter and gray matter from T1 images and compute thalamic volume. Regional cortical thickness was calculated using QDEC.

RESULTS

The 2 matched cohorts differed in disability, with AAwMS demonstrating significantly higher EDSS scores (2.3±2.2 vs. 1.3±1.5, P < .009), yet the 2 populations had similar T2 hyperintense lesion volumes (P = .35). AAwMS had a significantly lower total global cortical thickness when compared to CAwMS (P = .03). Controlling for EDSS, AAwMS showed multiple cortical regions to be significantly thinner than CAwMS; these included areas within the temporal, parietal and occipital lobes, as well as the precentral and postcentral gyrus. Middletemporal cortex was most affected in AAwMS in the left hemisphere (P = .009), while the superiortemporal cortex was most affected in the right hemisphere (P = .0001). In contrast, thalamic volume was significantly reduced in CAwMS when compared to AAwMS (P = .01). In both groups, worse disability was associated with lower total thalamic volume percentage.

CONCLUSION

AAwMS and CAwMS patients differ with regard to global and regional cortical thickness and thalamic volume. This diverging pattern of gray matter volumetrics among otherwise matched patients suggests that racial-specific disease differences may exist.

Cortical Thickness in Mild Traumatic Brain Injury

Magnetic resonance imaging data were acquired at ∼24 h and ∼3 months post-injury on mild traumatic brain injury (mTBI; n = 75) and orthopedic injury (n = 60) cohorts. The mTBI subjects were randomly assigned to a treatment group with atorvastatin or a non-treatment mTBI group. The treatment group was further divided into drug and placebo subgroups. FreeSurfer software package was used to compute cortical thickness based on the three dimensional T1-weighted images at both time-points. Cross-sectional analysis was carried out to compare cortical thickness between the mTBI and control groups. Longitudinal unbiased templates were generated for all subjects and cortical thickness measurements were compared between baseline and follow-up scans in the mTBI group. At baseline, significant reduction in cortical thickness was observed in the left middle temporal and the right superior parietal regions in the mTBI group, relative to the control group (p = 0.01). At follow-up, significant cortical thinning was again observed in the left middle temporal cortex in the mTBI group. Further analysis revealed significant cortical thinning only in the non-treatment group relative to the control group. In the follow-up, small regions with significant but subtle cortical thinning and thickening were seen in the frontal, temporal, and parietal lobes in the left hemisphere in the non-treatment group only. Our results indicate that cortical thickness could serve as a useful measure in identifying subtle changes in mTBI patients.