Assessing the consistency of quantitative MRI measurements is critical for inclusion in longitudinal studies and clinical trials. Intraclass coefficient correlation and coefficient of variation were used to evaluate the different consistency aspects of diffusion- and myelinbased MRI measures. Multi-shell diffusion and inhomogeneous magnetization transfer datasets were collected from twenty healthy adults at a high-frequency of five MRI sessions. The consistency was evaluated across whole bundles and the track-profile along the bundles. The impact of the fiber populations on the consistency was also evaluated using the number of fiber orientations map. For whole and profile bundles, moderate to high reliability of diffusion and myelin measures were observed. We report higher reliability of measures for multiple fiber populations than single. The overall portrait of the most consistent measurements and bundles drawn from a wide range of MRI techniques presented here will be particularly useful for identifying reliable biomarkers capable of detecting, monitoring and predicting white matter changes in clinical applications and has the potential to inform patient-specific treatment strategies. Key points Reliability and variability are excellent to good for DWI measurements, and good to moderate for MT measures for whole bundles and along the bundles. The number of fiber populations affects the reliability and variability of the MRI measurements. The reliability and variability of MRI measurements are also bundle dependent.

Highlights • qT1 and QSM showed the highest sensitivity to distinguish MS focal WM and cortical pathology from peri-plaque.• MWF and MTsat exhibited the highest sensitivity to NAWM pathology.• qT1 appeared to be the most sensitive measure to NAGM pathology.• All myelin-sensitive qMRI measures exhibited high intra-scanner reproducibility.

Distributional shift, or the mismatch between training and deployment data, is a significant obstacle to the usage of machine learning in high-stakes industrial applications, such as autonomous driving and medicine. This creates a need to be able to assess how robustly ML models generalize as well as the quality of their uncertainty estimates. Standard ML baseline datasets do not allow these properties to be assessed, as the training, validation and test data are often identically distributed. Recently, a range of dedicated benchmarks have appeared, featuring both distributionally matched and shifted data. Among these benchmarks, the Shifts dataset stands out in terms of the diversity of tasks as well as the data modalities it features. While most of the benchmarks are heavily dominated by 2D image classification tasks, Shifts contains tabular weather forecasting, machine translation, and vehicle motion prediction tasks. This enables the robustness properties of models to be assessed on a diverse set of industrial-scale tasks and either universal or directly applicable task-specific conclusions to be reached. In this paper, we extend the Shifts Dataset with two datasets sourced from industrial, high-risk applications of high societal importance. Specifically, we consider the tasks of segmentation of white matter Multiple Sclerosis lesions in 3D magnetic resonance brain images and the estimation of power consumption in marine cargo vessels. Both tasks feature ubiquitous distributional shifts and a strict safety requirement due to the high cost of errors. These new datasets will allow researchers to further explore robust generalization and uncertainty estimation in new situations. In this work, we provide a description of the dataset and baseline results for both tasks.

Neuropathological studies have shown that multiple sclerosis (MS) lesions are heterogeneous in terms of myelin/axon damage and repair as well as iron content. However, it remains a challenge to identify specific chronic lesion types, especially remyelinated lesions, in vivo in patients with MS.

This cohort study examines magnetic resonance imaging scans to determine changes in regional and total brain volume among patients with relapsing multiple sclerosis who have disease progression without relapse activity.

The choroid plexus has been shown to play a crucial role in CNS inflammation. Previous studies found larger choroid plexus in multiple sclerosis (MS) compared with healthy controls. However, it is not clear whether the choroid plexus is similarly involved in MS and in neuromyelitis optica spectrum disorder (NMOSD). Thus, the aim of this study was to compare the choroid plexus volume in MS and NMOSD.In this retrospective, cross-sectional study, patients were included by convenience sampling from 4 international MS centers. The choroid plexus of the lateral ventricles was segmented fully automatically on T1-weighted MRI sequences using a deep learning algorithm (Multi-Dimensional Gated Recurrent Units). Uni- and multivariable linear models were applied to investigate associations between the choroid plexus volume, clinically meaningful disease characteristics, and MRI parameters.We studied 180 patients with MS and 98 patients with NMOSD. In total, 94 healthy individuals and 47 patients with migraine served as controls. The choroid plexus volume was larger in MS (median 1,690 µL, interquartile range [IQR] 648 µL) than in NMOSD (median 1,403 µL, IQR 510 µL), healthy individuals (median 1,533 µL, IQR 570 µL), and patients with migraine (median 1,404 µL, IQR 524 µL; all p < 0.001), whereas there was no difference between NMOSD, migraine, and healthy controls. This was also true when adjusted for age, sex, and the intracranial volume. In contrast to NMOSD, the choroid plexus volume in MS was associated with the number of T2-weighted lesions in a linear model adjusted for age, sex, total intracranial volume, disease duration, relapses in the year before MRI, disease course, Expanded Disability Status Scale score, disease-modifying treatment, and treatment duration (beta 4.4; 95% CI 0.78–8.1; p = 0.018).This study supports an involvement of the choroid plexus in MS in contrast to NMOSD and provides clues to better understand the respective pathogenesis.

Limitations in the accuracy of brain pathways reconstructed by diffusion MRI (dMRI) tractography have received considerable attention. While the technical advances spearheaded by the Human Connectome Project (HCP) led to significant improvements in dMRI data quality, it remains unclear how these data should be analyzed to maximize tractography accuracy. Over a period of two years, we have engaged the dMRI community in the IronTract Challenge, which aims to answer this question by leveraging a unique dataset. Macaque brains that have received both tracer injections and ex vivo dMRI at high spatial and angular resolution allow a comprehensive, quantitative assessment of tractography accuracy on state-of-the-art dMRI acquisition schemes. We find that, when analysis methods are carefully optimized, the HCP scheme can achieve similar accuracy as a more time-consuming, Cartesian-grid scheme. Importantly, we show that simple pre- and post-processing strategies can improve the accuracy and robustness of many tractography methods. Finally, we find that fiber configurations that go beyond crossing (e.g., fanning, branching) are the most challenging for tractography. The IronTract Challenge remains open and we hope that it can serve as a valuable validation tool for both users and developers of dMRI analysis methods.

Highlights • The functional characterization of callosal connections is informed by anatomical data.• Callosal connections play a conditional driving role depending on the brain state and behavioral demands.• Callosal connections play a modulatory function, in addition to a driving role.• The corpus callosum participates in learning and interhemispheric transfer of sensorimotor habits.• The corpus callosum contributes to language processing and cognitive functions.

. The limited availability of large image datasets, mainly due to data privacy and differences in acquisition protocols or hardware, is a significant issue in the development of accurate and generalizable machine learning methods in medicine. This is especially the case for Magnetic Resonance (MR) images, where different MR scanners introduce a bias that limits the performance of a machine learning model. We present a novel method that learns to ignore the scanner-related features present in MR images, by introducing specific additional constraints on the latent space. We focus on a real-world classification scenario, where only a small dataset provides images of all classes. Our method Learn to Ignore (L2I) outperforms state-of-the-art domain adaptation methods on a multi-site MR dataset for a classification task between multiple sclerosis patients and healthy controls.

Introduction: Graph theory has been applied to study the pathophysiology of multiple sclerosis (MS) since it provides global and focal measures of brain network properties that are affected by MS. Typically, the connection strength and, consequently, the network properties are computed by counting the number of streamlines (NOS) connecting couples of gray matter regions. However, recent studies have shown that this method is not quantitative. Methods: We evaluated diffusion-based microstructural measures extracted from three different models to assess the network properties in a group of 66 MS patients and 64 healthy subjects. Besides, we assessed their correlation with patients' disability and with a biological measure of neuroaxonal damage. Results: Graph metrics extracted from connectomes weighted by intra-axonal microstructural components were the most sensitive to MS pathology and the most related to clinical disability. In contrast, measures of network segregation extracted from the connectomes weighted by maps describing extracellular diffusivity were the most related to serum concentration of neurofilament light chain. Network properties assessed with NOS were neither sensitive to MS pathology nor correlated with clinical and pathological measures of disease impact in MS patients. Conclusion: Using tractometry-derived graph measures in MS patients, we identified a set of metrics based on microstructural components that are highly sensitive to the disease and that provide sensitive correlates of clinical and biological deterioration in MS patients. Impact statement Graph theory has been widely used to study the alterations in the structural connectivity of multiple sclerosis (MS) patients. Usually, brain graphs used for the extraction of network metrics are created by counting the number of streamlines connecting gray matter regions, however, this method is not quantitative. In this study, we used tractometry to average the values of diffusion-based microstructural maps along the reconstructed streamlines. Our results show that network metrics extracted from the connectomes weighted on microstructural maps provide sensitive information to MS pathology, which correlate with clinical and biological measures of disease impact.