Extra-axonal contribution to double diffusion encoding-based pore size estimates in the corticospinal tract.

OBJECTIVE: To study the origin of compartment size overestimation in double diffusion encoding MRI (DDE) in vivo experiments in the human corticospinal tract. Here, the extracellular space is hypothesized to be the origin of the DDE signal. By exploiting the DDE sensitivity to pore shape, it could be possible to identify the origin of the measured signal. The signal difference between parallel and perpendicular diffusion gradient orientation can indicate if a compartment is regular or eccentric in shape. As extracellular space can be considered an eccentric compartment, a positive difference would mean a high contribution to the compartment size estimates. MATERIALS AND METHODS: Computer simulations using MISST and in vivo experiments in eight healthy volunteers were performed. DDE experiments using a double spin-echo preparation with eight perpendicular directions were measured in vivo. The difference between parallel and perpendicular gradient orientations was analyzed using a Wilcoxon signed-rank test and a Mann-Whitney U test. RESULTS: Simulations and MR experiments showed a statistically significant difference between parallel and perpendicular diffusion gradient orientation signals ([Formula: see text]). CONCLUSION: The results suggest that the DDE-based size estimate may be considerably influenced by the extra-axonal compartment. However, the experimental results are also consistent with purely intra-axonal contributions in combination with a large fiber orientation dispersion.

Clinical applicability of quantitative atrophy measures on MRI in patients suspected of Alzheimer's disease.

OBJECTIVES: Neurodegeneration in suspected Alzheimer's disease can be determined using visual rating or quantitative volumetric assessments. We examined the feasibility of volumetric measurements of gray matter (GMV) and hippocampal volume (HCV) and compared their diagnostic performance with visual rating scales in academic and non-academic memory clinics. MATERIALS AND METHODS: We included 231 patients attending local memory clinics (LMC) in the Netherlands and 501 of the academic Amsterdam Dementia Cohort (ADC). MRI scans were acquired using local protocols, including a T1-weighted sequence. Quantification of GMV and HCV was performed using FSL and FreeSurfer. Medial temporal atrophy and global atrophy were assessed with visual rating scales. ROC curves were derived to determine which measure discriminated best between cognitively normal (CN), mild cognitive impairment (MCI), and Alzheimer's dementia (AD). RESULTS: Patients attending LMC (age 70.9 ± 8.9 years; 47% females; 19% CN; 34% MCI; 47% AD) were older, had more cerebrovascular pathology, and had lower GMV and HCV compared to those of the ADC (age 64.9 ± 8.2 years; 42% females; 35% CN, 43% MCI, 22% AD). While visual ratings were feasible in > 95% of scans in both cohorts, quantification was achieved in 94-98% of ADC, but only 68-85% of LMC scans, depending on the software. Visual ratings and volumetric outcomes performed similarly in discriminating CN vs AD in both cohorts. CONCLUSION: In clinical settings, quantification of GM and hippocampal atrophy currently fails in up to one-third of scans, probably due to lack of standardized acquisition protocols. Diagnostic accuracy is similar for volumetric measures and visual rating scales, making the latter suited for clinical practice. In a real-life clinical setting, volumetric assessment of MRI scans in dementia patients may require acquisition protocol optimization and does not outperform visual rating scales. KEY POINTS: • In a real-life clinical setting, the diagnostic performance of visual rating scales is similar to that of automatic volumetric quantification and may be sufficient to distinguish Alzheimer's disease groups. • Volumetric assessment of gray matter and hippocampal volumes from MRI scans of patients attending non-academic memory clinics fails in up to 32% of cases. • Clinical MR acquisition protocols should be optimized to improve the output of quantitative software for segmentation of Alzheimer's disease-specific outcomes.

Application of the ATN classification scheme in a population without dementia: Findings from the EPAD cohort.

BACKGROUND: We classified non-demented European Prevention of Alzheimer's Dementia (EPAD) participants through the amyloid/tau/neurodegeneration (ATN) scheme and assessed their neuropsychological and imaging profiles. MATERIALS AND METHODS: From 1500 EPAD participants, 312 were excluded. Cerebrospinal fluid cut-offs of 1000 pg/mL for amyloid beta (Aß)1-42 and 27 pg/mL for p-tau181 were validated using Gaussian mixture models. Given strong correlation of p-tau and t-tau (R2  = 0.98, P < 0.001), neurodegeneration was defined by age-adjusted hippocampal volume. Multinomial regressions were used to test whether neuropsychological tests and regional brain volumes could distinguish ATN stages. RESULTS: Age was 65 ± 7 years, with 58% females and 38% apolipoprotein E (APOE) ε4 carriers; 57.1% were A-T-N-, 32.5% were in the Alzheimer's disease (AD) continuum, and 10.4% suspected non-Alzheimer's pathology. Age and cerebrovascular burden progressed with biomarker positivity (P < 0.001). Cognitive dysfunction appeared with T+. Paradoxically higher regional gray matter volumes were observed in A+T-N- compared to A-T-N- (P < 0.001). DISCUSSION: In non-demented individuals along the AD continuum, p-tau drives cognitive dysfunction. Memory and language domains are affected in the earliest stages.

ExploreASL: An image processing pipeline for multi-center ASL perfusion MRI studies.

Arterial spin labeling (ASL) has undergone significant development since its inception, with a focus on improving standardization and reproducibility of its acquisition and quantification. In a community-wide effort towards robust and reproducible clinical ASL image processing, we developed the software package ExploreASL, allowing standardized analyses across centers and scanners. The procedures used in ExploreASL capitalize on published image processing advancements and address the challenges of multi-center datasets with scanner-specific processing and artifact reduction to limit patient exclusion. ExploreASL is self-contained, written in MATLAB and based on Statistical Parameter Mapping (SPM) and runs on multiple operating systems. To facilitate collaboration and data-exchange, the toolbox follows several standards and recommendations for data structure, provenance, and best analysis practice. ExploreASL was iteratively refined and tested in the analysis of >10,000 ASL scans using different pulse-sequences in a variety of clinical populations, resulting in four processing modules: Import, Structural, ASL, and Population that perform tasks, respectively, for data curation, structural and ASL image processing and quality control, and finally preparing the results for statistical analyses on both single-subject and group level. We illustrate ExploreASL processing results from three cohorts: perinatally HIV-infected children, healthy adults, and elderly at risk for neurodegenerative disease. We show the reproducibility for each cohort when processed at different centers with different operating systems and MATLAB versions, and its effects on the quantification of gray matter cerebral blood flow. ExploreASL facilitates the standardization of image processing and quality control, allowing the pooling of cohorts which may increase statistical power and discover between-group perfusion differences. Ultimately, this workflow may advance ASL for wider adoption in clinical studies, trials, and practice.

Progression of regional grey matter atrophy in multiple sclerosis.

See Stankoff and Louapre (doi:10.1093/brain/awy114) for a scientific commentary on this article.Grey matter atrophy is present from the earliest stages of multiple sclerosis, but its temporal ordering is poorly understood. We aimed to determine the sequence in which grey matter regions become atrophic in multiple sclerosis and its association with disability accumulation. In this longitudinal study, we included 1417 subjects: 253 with clinically isolated syndrome, 708 with relapsing-remitting multiple sclerosis, 128 with secondary-progressive multiple sclerosis, 125 with primary-progressive multiple sclerosis, and 203 healthy control subjects from seven European centres. Subjects underwent repeated MRI (total number of scans 3604); the mean follow-up for patients was 2.41 years (standard deviation = 1.97). Disability was scored using the Expanded Disability Status Scale. We calculated the volume of brain grey matter regions and brainstem using an unbiased within-subject template and used an established data-driven event-based model to determine the sequence of occurrence of atrophy and its uncertainty. We assigned each subject to a specific event-based model stage, based on the number of their atrophic regions. Linear mixed-effects models were used to explore associations between the rate of increase in event-based model stages, and T2 lesion load, disease-modifying treatments, comorbidity, disease duration and disability accumulation. The first regions to become atrophic in patients with clinically isolated syndrome and relapse-onset multiple sclerosis were the posterior cingulate cortex and precuneus, followed by the middle cingulate cortex, brainstem and thalamus. A similar sequence of atrophy was detected in primary-progressive multiple sclerosis with the involvement of the thalamus, cuneus, precuneus, and pallidum, followed by the brainstem and posterior cingulate cortex. The cerebellum, caudate and putamen showed early atrophy in relapse-onset multiple sclerosis and late atrophy in primary-progressive multiple sclerosis. Patients with secondary-progressive multiple sclerosis showed the highest event-based model stage (the highest number of atrophic regions, P < 0.001) at the study entry. All multiple sclerosis phenotypes, but clinically isolated syndrome, showed a faster rate of increase in the event-based model stage than healthy controls. T2 lesion load and disease duration in all patients were associated with increased event-based model stage, but no effects of disease-modifying treatments and comorbidity on event-based model stage were observed. The annualized rate of event-based model stage was associated with the disability accumulation in relapsing-remitting multiple sclerosis, independent of disease duration (P < 0.0001). The data-driven staging of atrophy progression in a large multiple sclerosis sample demonstrates that grey matter atrophy spreads to involve more regions over time. The sequence in which regions become atrophic is reasonably consistent across multiple sclerosis phenotypes. The spread of atrophy was associated with disease duration and with disability accumulation over time in relapsing-remitting multiple sclerosis.

Image quality transfer and applications in diffusion MRI.

This paper introduces a new computational imaging technique called image quality transfer (IQT). IQT uses machine learning to transfer the rich information available from one-off experimental medical imaging devices to the abundant but lower-quality data from routine acquisitions. The procedure uses matched pairs to learn mappings from low-quality to corresponding high-quality images. Once learned, these mappings then augment unseen low quality images, for example by enhancing image resolution or information content. Here, we demonstrate IQT using a simple patch-regression implementation and the uniquely rich diffusion MRI data set from the human connectome project (HCP). Results highlight potential benefits of IQT in both brain connectivity mapping and microstructure imaging. In brain connectivity mapping, IQT reveals, from standard data sets, thin connection pathways that tractography normally requires specialised data to reconstruct. In microstructure imaging, IQT shows potential in estimating, from standard "single-shell" data (one non-zero b-value), maps of microstructural parameters that normally require specialised multi-shell data. Further experiments show strong generalisability, highlighting IQT's benefits even when the training set does not directly represent the application domain. The concept extends naturally to many other imaging modalities and reconstruction problems.

Eigenvector centrality dynamics are related to Alzheimer's disease pathological changes in non-demented individuals.

Amyloid-ß accumulation starts in highly connected brain regions and is associated with functional connectivity alterations in the early stages of Alzheimer's disease. This regional vulnerability is related to the high neuronal activity and strong fluctuations typical of these regions. Recently, dynamic functional connectivity was introduced to investigate changes in functional network organization over time. High dynamic functional connectivity variations indicate increased regional flexibility to participate in multiple subnetworks, promoting functional integration. Currently, only a limited number of studies have explored the temporal dynamics of functional connectivity in the pre-dementia stages of Alzheimer's disease. We study the associations between abnormal cerebrospinal fluid amyloid and both static and dynamic properties of functional hubs, using eigenvector centrality, and their relationship with cognitive performance, in 701 non-demented participants from the European Prevention of Alzheimer's Dementia cohort. Voxel-wise eigenvector centrality was computed for the whole functional magnetic resonance imaging time series (static), and within a sliding window (dynamic). Differences in static eigenvector centrality between amyloid positive (A+) and negative (A-) participants and amyloid-tau groups were found in a general linear model. Dynamic eigenvector centrality standard deviation and range were compared between groups within clusters of significant static eigenvector centrality differences, and within 10 canonical resting-state networks. The effect of the interaction between amyloid status and cognitive performance on dynamic eigenvector centrality variability was also evaluated with linear models. Models were corrected for age, sex, and education level. Lower static centrality was found in A+ participants in posterior brain areas including a parietal and an occipital cluster; higher static centrality was found in a medio-frontal cluster. Lower eigenvector centrality variability (standard deviation) occurred in A+ participants in the frontal cluster. The default mode network and the dorsal visual networks of A+ participants had lower dynamic eigenvector centrality variability. Centrality variability in the default mode network and dorsal visual networks were associated with cognitive performance in the A- and A+ groups, with lower variability being observed in A+ participants with good cognitive scores. Our results support the role and timing of eigenvector centrality alterations in very early stages of Alzheimer's disease and show that centrality variability over time adds relevant information on the dynamic patterns that cause static eigenvector centrality alterations. We propose that dynamic eigenvector centrality is an early biomarker of the interplay between early Alzheimer's disease pathology and cognitive decline.

Spatial-Temporal Patterns of ß-Amyloid Accumulation: A Subtype and Stage Inference Model Analysis.

BACKGROUND AND OBJECTIVES: ß-amyloid (Aß) staging models assume a single spatial-temporal progression of amyloid accumulation. We assessed evidence for Aß accumulation subtypes by applying the data-driven Subtype and Stage Inference (SuStaIn) model to amyloid-PET data. METHODS: Amyloid-PET data of 3,010 participants were pooled from 6 cohorts (ALFA+, EMIF-AD, ABIDE, OASIS, and ADNI). Standardized uptake value ratios were calculated for 17 regions. We applied the SuStaIn algorithm to identify consistent subtypes in the pooled dataset based on the cross-validation information criterion and the most probable subtype/stage classification per scan. The effects of demographics and risk factors on subtype assignment were assessed using multinomial logistic regression. RESULTS: Participants were mostly cognitively unimpaired (n = 1890 [62.8%]), had a mean age of 68.72 (SD 9.1) years, 42.1% were APOE ε4 carriers, and 51.8% were female. A 1-subtype model recovered the traditional amyloid accumulation trajectory, but SuStaIn identified 3 optimal subtypes, referred to as frontal, parietal, and occipital based on the first regions to show abnormality. Of the 788 (26.2%) with strong subtype assignment (>50% probability), the majority was assigned to frontal (n = 415 [52.5%]), followed by parietal (n = 199 [25.3%]) and occipital subtypes (n = 175 [22.2%]). Significant differences across subtypes included distinct proportions of APOE ε4 carriers (frontal 61.8%, parietal 57.1%, occipital 49.4%), participants with dementia (frontal 19.7%, parietal 19.1%, occipital 31.0%), and lower age for the parietal subtype (frontal/occipital 72.1 years, parietal 69.3 years). Higher amyloid (Centiloid) and CSF p-tau burden was observed for the frontal subtype; parietal and occipital subtypes did not differ. At follow-up, most participants (81.1%) maintained baseline subtype assignment and 25.6% progressed to a later stage. DISCUSSION: Whereas a 1-trajectory model recovers the established pattern of amyloid accumulation, SuStaIn determined that 3 subtypes were optimal, showing distinct associations with Alzheimer disease risk factors. Further analyses to determine clinical utility are warranted.

The Open-Access European Prevention of Alzheimer's Dementia (EPAD) MRI dataset and processing workflow.

The European Prevention of Alzheimer Dementia (EPAD) is a multi-center study that aims to characterize the preclinical and prodromal stages of Alzheimer's Disease. The EPAD imaging dataset includes core (3D T1w, 3D FLAIR) and advanced (ASL, diffusion MRI, and resting-state fMRI) MRI sequences. Here, we give an overview of the semi-automatic multimodal and multisite pipeline that we developed to curate, preprocess, quality control (QC), and compute image-derived phenotypes (IDPs) from the EPAD MRI dataset. This pipeline harmonizes DICOM data structure across sites and performs standardized MRI preprocessing steps. A semi-automated MRI QC procedure was implemented to visualize and flag MRI images next to site-specific distributions of QC features - i.e. metrics that represent image quality. The value of each of these QC features was evaluated through comparison with visual assessment and step-wise parameter selection based on logistic regression. IDPs were computed from 5 different MRI modalities and their sanity and potential clinical relevance were ascertained by assessing their relationship with biological markers of aging and dementia. The EPAD v1500.0 data release encompassed core structural scans from 1356 participants 842 fMRI, 831 dMRI, and 858 ASL scans. From 1356 3D T1w images, we identified 17 images with poor quality and 61 with moderate quality. Five QC features - Signal to Noise Ratio (SNR), Contrast to Noise Ratio (CNR), Coefficient of Joint Variation (CJV), Foreground-Background energy Ratio (FBER), and Image Quality Rate (IQR) - were selected as the most informative on image quality by comparison with visual assessment. The multimodal IDPs showed greater impairment in associations with age and dementia biomarkers, demonstrating the potential of the dataset for future clinical analyses.

Opportunities for Understanding MS Mechanisms and Progression With MRI Using Large-Scale Data Sharing and Artificial Intelligence.

Patients with multiple sclerosis (MS) have heterogeneous clinical presentations, symptoms, and progression over time, making MS difficult to assess and comprehend in vivo. The combination of large-scale data sharing and artificial intelligence creates new opportunities for monitoring and understanding MS using MRI. First, development of validated MS-specific image analysis methods can be boosted by verified reference, test, and benchmark imaging data. Using detailed expert annotations, artificial intelligence algorithms can be trained on such MS-specific data. Second, understanding disease processes could be greatly advanced through shared data of large MS cohorts with clinical, demographic, and treatment information. Relevant patterns in such data that may be imperceptible to a human observer could be detected through artificial intelligence techniques. This applies from image analysis (lesions, atrophy, or functional network changes) to large multidomain datasets (imaging, cognition, clinical disability, genetics). After reviewing data sharing and artificial intelligence, we highlight 3 areas that offer strong opportunities for making advances in the next few years: crowdsourcing, personal data protection, and organized analysis challenges. Difficulties as well as specific recommendations to overcome them are discussed, in order to best leverage data sharing and artificial intelligence to improve image analysis, imaging, and the understanding of MS.

The sequence of structural, functional and cognitive changes in multiple sclerosis.

BACKGROUND: As disease progression remains poorly understood in multiple sclerosis (MS), we aim to investigate the sequence in which different disease milestones occur using a novel data-driven approach. METHODS: We analysed a cohort of 295 relapse-onset MS patients and 96 healthy controls, and considered 28 features, capturing information on T2-lesion load, regional brain and spinal cord volumes, resting-state functional centrality ("hubness"), microstructural tissue integrity of major white matter (WM) tracts and performance on multiple cognitive tests. We used a discriminative event-based model to estimate the sequence of biomarker abnormality in MS progression in general, as well as specific models for worsening physical disability and cognitive impairment. RESULTS: We demonstrated that grey matter (GM) atrophy of the cerebellum, thalamus, and changes in corticospinal tracts are early events in MS pathology, whereas other WM tracts as well as the cognitive domains of working memory, attention, and executive function are consistently late events. The models for disability and cognition show early functional changes of the default-mode network and earlier changes in spinal cord volume compared to the general MS population. Overall, GM atrophy seems crucial due to its early involvement in the disease course, whereas WM tract integrity appears to be affected relatively late despite the early onset of WM lesions. CONCLUSION: Data-driven modelling revealed the relative occurrence of both imaging and non-imaging events as MS progresses, providing insights into disease propagation mechanisms, and allowing fine-grained staging of patients for monitoring purposes.

White matter microstructure disruption in early stage amyloid pathology.

INTRODUCTION: Amyloid beta (Aß) accumulation is the first pathological hallmark of Alzheimer's disease (AD), and it is associated with altered white matter (WM) microstructure. We aimed to investigate this relationship at a regional level in a cognitively unimpaired cohort. METHODS: We included 179 individuals from the European Medical Information Framework for AD (EMIF-AD) preclinAD study, who underwent diffusion magnetic resonance (MR) to determine tract-level fractional anisotropy (FA); mean, radial, and axial diffusivity (MD/RD/AxD); and dynamic [18F]flutemetamol) positron emission tomography (PET) imaging to assess amyloid burden. RESULTS: Regression analyses showed a non-linear relationship between regional amyloid burden and WM microstructure. Low amyloid burden was associated with increased FA and decreased MD/RD/AxD, followed by decreased FA and increased MD/RD/AxD upon higher amyloid burden. The strongest association was observed between amyloid burden in the precuneus and body of the corpus callosum (CC) FA and diffusivity (MD/RD) measures. In addition, amyloid burden in the anterior cingulate cortex strongly related to AxD and RD measures in the genu CC. DISCUSSION: Early amyloid deposition is associated with changes in WM microstructure. The non-linear relationship might reflect multiple stages of axonal damage.

The relation between APOE genotype and cerebral microbleeds in cognitively unimpaired middle- and old-aged individuals.

Positive associations between cerebral microbleeds (CMBs) and APOE-ε4 (apolipoprotein E) genotype have been reported in Alzheimer's disease, but show conflicting results. We investigated the effect of APOE genotype on CMBs in a cohort of cognitively unimpaired middle- and old-aged individuals enriched for APOE-ε4 genotype. Participants from ALFA (Alzheimer and Families) cohort were included and their magnetic resonance scans assessed (n = 564, 50% APOE-ε4 carriers). Quantitative magnetic resonance analyses included visual ratings, atrophy measures, and white matter hyperintensity (WMH) segmentations. The prevalence of CMBs was 17%, increased with age (p < 0.05), and followed an increasing trend paralleling APOE-ε4 dose. The number of CMBs was significantly higher in APOE-ε4 homozygotes compared to heterozygotes and non-carriers (p < 0.05). This association was driven by lobar CMBs (p < 0.05). CMBs co-localized with WMH (p < 0.05). No associations between CMBs and APOE-ε2, gray matter volumes, and cognitive performance were found. Our results suggest that cerebral vessels of APOE-ε4 homozygous are more fragile, especially in lobar locations. Co-occurrence of CMBs and WMH suggests that such changes localize in areas with increased vascular vulnerability.

Multitracer model for staging cortical amyloid deposition using PET imaging.

OBJECTIVE: To develop and evaluate a model for staging cortical amyloid deposition using PET with high generalizability. METHODS: Three thousand twenty-seven individuals (1,763 cognitively unimpaired [CU], 658 impaired, 467 with Alzheimer disease [AD] dementia, 111 with non-AD dementia, and 28 with missing diagnosis) from 6 cohorts (European Medical Information Framework for AD, Alzheimer's and Family, Alzheimer's Biomarkers in Daily Practice, Amsterdam Dementia Cohort, Open Access Series of Imaging Studies [OASIS]-3, Alzheimer's Disease Neuroimaging Initiative [ADNI]) who underwent amyloid PET were retrospectively included; 1,049 individuals had follow-up scans. With application of dataset-specific cutoffs to global standard uptake value ratio (SUVr) values from 27 regions, single-tracer and pooled multitracer regional rankings were constructed from the frequency of abnormality across 400 CU individuals (100 per tracer). The pooled multitracer ranking was used to create a staging model consisting of 4 clusters of regions because it displayed a high and consistent correlation with each single-tracer ranking. Relationships between amyloid stage, clinical variables, and longitudinal cognitive decline were investigated. RESULTS: SUVr abnormality was most frequently observed in cingulate, followed by orbitofrontal, precuneal, and insular cortices and then the associative, temporal, and occipital regions. Abnormal amyloid levels based on binary global SUVr classification were observed in 1.0%, 5.5%, 17.9%, 90.0%, and 100.0% of individuals in stage 0 to 4, respectively. Baseline stage predicted decline in Mini-Mental State Examination (MMSE) score (ADNI: n = 867, F = 67.37, p < 0.001; OASIS: n = 475, F = 9.12, p < 0.001) and faster progression toward an MMSE score ≤25 (ADNI: n = 787, hazard ratio [HR]stage1 2.00, HRstage2 3.53, HRstage3 4.55, HRstage4 9.91, p < 0.001; OASIS: n = 469, HRstage4 4.80, p < 0.001). CONCLUSION: The pooled multitracer staging model successfully classified the level of amyloid burden in >3,000 individuals across cohorts and radiotracers and detects preglobal amyloid burden and distinct risk profiles of cognitive decline within globally amyloid-positive individuals.

Reduced accuracy of MRI deep grey matter segmentation in multiple sclerosis: an evaluation of four automated methods against manual reference segmentations in a multi-center cohort.

BACKGROUND: Deep grey matter (DGM) atrophy in multiple sclerosis (MS) and its relation to cognitive and clinical decline requires accurate measurements. MS pathology may deteriorate the performance of automated segmentation methods. Accuracy of DGM segmentation methods is compared between MS and controls, and the relation of performance with lesions and atrophy is studied. METHODS: On images of 21 MS subjects and 11 controls, three raters manually outlined caudate nucleus, putamen and thalamus; outlines were combined by majority voting. FSL-FIRST, FreeSurfer, Geodesic Information Flow and volBrain were evaluated. Performance was evaluated volumetrically (intra-class correlation coefficient (ICC)) and spatially (Dice similarity coefficient (DSC)). Spearman's correlations of DSC with global and local lesion volume, structure of interest volume (ROIV), and normalized brain volume (NBV) were assessed. RESULTS: ICC with manual volumes was mostly good and spatial agreement was high. MS exhibited significantly lower DSC than controls for thalamus and putamen. For some combinations of structure and method, DSC correlated negatively with lesion volume or positively with NBV or ROIV. Lesion-filling did not substantially change segmentations. CONCLUSIONS: Automated methods have impaired performance in patients. Performance generally deteriorated with higher lesion volume and lower NBV and ROIV, suggesting that these may contribute to the impaired performance.

Multi-study validation of data-driven disease progression models to characterize evolution of biomarkers in Alzheimer's disease.

Understanding the sequence of biological and clinical events along the course of Alzheimer's disease provides insights into dementia pathophysiology and can help participant selection in clinical trials. Our objective is to train two data-driven computational models for sequencing these events, the Event Based Model (EBM) and discriminative-EBM (DEBM), on the basis of well-characterized research data, then validate the trained models on subjects from clinical cohorts characterized by less-structured data-acquisition protocols. Seven independent data cohorts were considered totalling 2389 cognitively normal (CN), 1424 mild cognitive impairment (MCI) and 743 Alzheimer's disease (AD) patients. The Alzheimer's Disease Neuroimaging Initiative (ADNI) data set was used as training set for the constriction of disease models while a collection of multi-centric data cohorts was used as test set for validation. Cross-sectional information related to clinical, cognitive, imaging and cerebrospinal fluid (CSF) biomarkers was used. Event sequences obtained with EBM and DEBM showed differences in the ordering of single biomarkers but according to both the first biomarkers to become abnormal were those related to CSF, followed by cognitive scores, while structural imaging showed significant volumetric decreases at later stages of the disease progression. Staging of test set subjects based on sequences obtained with both models showed good linear correlation with the Mini Mental State Examination score (R2EBM = 0.866; R2DEBM = 0.906). In discriminant analyses, significant differences (p-value ≤ 0.05) between the staging of subjects from training and test sets were observed in both models. No significant difference between the staging of subjects from the training and test was observed (p-value > 0.05) when considering a subset composed by 562 subjects for which all biomarker families (cognitive, imaging and CSF) are available. Event sequence obtained with DEBM recapitulates the heuristic models in a data-driven fashion and is clinically plausible. We demonstrated inter-cohort transferability of two disease progression models and their robustness in detecting AD phases. This is an important step towards the adoption of data-driven statistical models into clinical domain.

SVM recursive feature elimination analyses of structural brain MRI predicts near-term relapses in patients with clinically isolated syndromes suggestive of multiple sclerosis.

Machine learning classification is an attractive approach to automatically differentiate patients from healthy subjects, and to predict future disease outcomes. A clinically isolated syndrome (CIS) is often the first presentation of multiple sclerosis (MS), but it is difficult at onset to predict who will have a second relapse and hence convert to clinically definite MS. In this study, we thus aimed to distinguish CIS converters from non-converters at onset of a CIS, using recursive feature elimination and weight averaging with support vector machines. We also sought to assess the influence of cohort size and cross-validation methods on the accuracy estimate of the classification. We retrospectively collected 400 patients with CIS from six European MAGNIMS MS centres. Patients underwent brain MRI at onset of a CIS according to local standard-of-care protocols. The diagnosis of clinically definite MS at one-year follow-up was the standard against which the accuracy of the model was tested. For each patient, we derived MRI-based features, such as grey matter probability, white matter lesion load, cortical thickness, and volume of specific cortical and white matter regions. Features with little contribution to the classification model were removed iteratively through an interleaved sample bootstrapping and feature averaging approach. Classification of CIS outcome at one-year follow-up was performed with 2-fold, 5-fold, 10-fold and leave-one-out cross-validation for each centre cohort independently and in all patients together. The estimated classification accuracy across centres ranged from 64.9% to 88.1% using 2-fold cross-validation and from 73% to 92.9% using leave-one-out cross-validation. The classification accuracy estimate was higher in single-centre, smaller data sets than in combinations of data sets, being the lowest when all patients were merged together. Regional MRI features such as WM lesions, grey matter probability in the thalamus and the precuneus or cortical thickness in the cuneus and inferior temporal gyrus predicted the occurrence of a second relapse in patients at onset of a CIS using support vector machines. The increased accuracy estimate of the classification achieved with smaller and single-centre samples may indicate a model bias (overfitting) when data points were limited, but also more homogeneous. We provide an overview of classifier performance from a range of cross-validation schemes to give insight into the variability across schemes. The proposed recursive feature elimination approach with weight averaging can be used both in single- and multi-centre data sets in order to bridge the gap between group-level comparisons and making predictions for individual patients.

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

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

Localization of the Epileptogenic Zone Using Interictal MEG and Machine Learning in a Large Cohort of Drug-Resistant Epilepsy Patients.

Objective: Epilepsy surgery results in seizure freedom in the majority of drug-resistant patients. To improve surgery outcome we studied whether MEG metrics combined with machine learning can improve localization of the epileptogenic zone, thereby enhancing the chance of seizure freedom. Methods: Presurgical interictal MEG recordings of 94 patients (64 seizure-free >1y post-surgery) were analyzed to extract four metrics in source space: delta power, low-to-high-frequency power ratio, functional connectivity (phase lag index), and minimum spanning tree betweenness centrality. At the group level, we estimated the overlap of the resection area with the five highest values for each metric and determined whether this overlap differed between surgery outcomes. At the individual level, those metrics were used in machine learning classifiers (linear support vector machine (SVM) and random forest) to distinguish between resection and non-resection areas and between surgery outcome groups. Results: The highest values, for all metrics, overlapped with the resection area in more than half of the patients, but the overlap did not differ between surgery outcome groups. The classifiers distinguished the resection areas from non-resection areas with 59.94% accuracy (95% confidence interval: 59.67-60.22%) for SVM and 60.34% (59.98-60.71%) for random forest, but could not differentiate seizure-free from not seizure-free patients [43.77% accuracy (42.08-45.45%) for SVM and 49.03% (47.25-50.82%) for random forest]. Significance: All four metrics localized the resection area but did not distinguish between surgery outcome groups, demonstrating that metrics derived from interictal MEG correspond to expert consensus based on several presurgical evaluation modalities, but do not yet localize the epileptogenic zone. Metrics should be improved such that they correspond to the resection area in seizure-free patients but not in patients with persistent seizures. It is important to test such localization strategies at an individual level, for example by using machine learning or individualized models, since surgery is individually tailored.

MRI predictors of amyloid pathology: results from the EMIF-AD Multimodal Biomarker Discovery study.

BACKGROUND: With the shift of research focus towards the pre-dementia stage of Alzheimer's disease (AD), there is an urgent need for reliable, non-invasive biomarkers to predict amyloid pathology. The aim of this study was to assess whether easily obtainable measures from structural MRI, combined with demographic data, cognitive data and apolipoprotein E (APOE) ε4 genotype, can be used to predict amyloid pathology using machine-learning classification. METHODS: We examined 810 subjects with structural MRI data and amyloid markers from the European Medical Information Framework for Alzheimer's Disease Multimodal Biomarker Discovery study, including subjects with normal cognition (CN, n = 337, age 66.5 ± 7.2, 50% female, 27% amyloid positive), mild cognitive impairment (MCI, n = 375, age 69.1 ± 7.5, 53% female, 63% amyloid positive) and AD dementia (n = 98, age 67.0 ± 7.7, 48% female, 97% amyloid positive). Structural MRI scans were visually assessed and Freesurfer was used to obtain subcortical volumes, cortical thickness and surface area measures. We first assessed univariate associations between MRI measures and amyloid pathology using mixed models. Next, we developed and tested an automated classifier using demographic, cognitive, MRI and APOE ε4 information to predict amyloid pathology. A support vector machine (SVM) with nested 10-fold cross-validation was applied to identify a set of markers best discriminating between amyloid positive and amyloid negative subjects. RESULTS: In univariate associations, amyloid pathology was associated with lower subcortical volumes and thinner cortex in AD-signature regions in CN and MCI. The multi-variable SVM classifier provided an area under the curve (AUC) of 0.81 ± 0.07 in MCI and an AUC of 0.74 ± 0.08 in CN. In CN, selected features for the classifier included APOE ε4, age, memory scores and several MRI measures such as hippocampus, amygdala and accumbens volumes and cortical thickness in temporal and parahippocampal regions. In MCI, the classifier including demographic and APOE ε4 information did not improve after additionally adding imaging measures. CONCLUSIONS: Amyloid pathology is associated with changes in structural MRI measures in CN and MCI. An automated classifier based on clinical, imaging and APOE ε4 data can identify the presence of amyloid pathology with a moderate level of accuracy. These results could be used in clinical trials to pre-screen subjects for anti-amyloid therapies.