Increased power by harmonizing structural MRI site differences with the ComBat batch adjustment method in ENIGMA

SAN: Mitigating spatial covariance heterogeneity in cortical thickness data collected from multiple scanners or sites

In neuroimaging studies, combining data collected from multiple study sites or scanners is becoming common to increase the reproducibility of scientific discoveries. At the same time, unwanted variations arise by using different scanners (inter-scanner biases), which need to be corrected before downstream analyses to facilitate replicable research and prevent spurious findings. While statistical harmonization methods such as ComBat have become popular in mitigating inter-scanner biases in neuroimaging, recent methodological advances have shown that harmonizing heterogeneous covariances results in higher data quality. In vertex-level cortical thickness data, heterogeneity in spatial autocorrelation is a critical factor that affects covariance heterogeneity. Our work proposes a new statistical harmonization method called spatial autocorrelation normalization (SAN) that preserves homogeneous covariance vertex-level cortical thickness data across different scanners. We use an explicit Gaussian process to characterize scanner-invariant and scanner-specific variations to reconstruct spatially homogeneous data across scanners. SAN is computationally feasible, and it easily allows the integration of existing harmonization methods. We demonstrate the utility of the proposed method using cortical thickness data from the Social Processes Initiative in the Neurobiology of the Schizophrenia(s) (SPINS) study. SAN is publicly available as an R package.

Sex Differences in the Brain's White Matter Microstructure during Development assessed using Advanced Diffusion MRI Models

Typical sex differences in white matter (WM) microstructure during development are incompletely understood. Here we evaluated sex differences in WM microstructure during typical brain development using a sample of neurotypical individuals across a wide developmental age (N=239, aged 5-22 years). We used the conventional diffusion-weighted MRI (dMRI) model, diffusion tensor imaging (DTI), and two advanced dMRI models, the tensor distribution function (TDF) and neurite orientation dispersion density imaging (NODDI) to assess WM microstructure. WM microstructure exhibited significant, regionally consistent sex differences across the brain during typical development. Additionally, the TDF model was most sensitive in detecting sex differences. These findings highlight the importance of considering sex in neurodevelopmental research and underscore the value of the advanced TDF model.

RASopathies influences on neuroanatomical variation in children

BACKGROUND

RASopathies are a group of disorders characterized by pathogenic mutations in the Ras-mitogen-activated protein kinase (Ras/MAPK) signaling pathway. Distinct pathogenic variants in genes encoding proteins in the Ras/MAPK pathway cause Noonan syndrome (NS) and neurofibromatosis type 1 (NF1), which are associated with increased risk for autism spectrum disorder (ASD) and attention deficit and hyperactivity disorder (ADHD).

METHODS

This study examines the effect RASopathies (NS and NF1) has on human neuroanatomy, specifically on surface area (SA), cortical thickness (CT), and subcortical volumes. We compared structural T1-weighted images, using vertex-based analysis for cortical measures and Desikan ROI parcellation for subcortical volumes on children with RASopathies (n=91, mean age = 8.81, SD = 2.12) to sex- and age-matched TD (n=74, mean age=9.07, SD = 1.77).

RESULTS

Compared to TD, RASopathies had convergent effects on SA and CT, exhibiting increased SA in the precentral gyrus, decreased SA in occipital regions, and thinner CT in the precentral gyrus. RASopathies exhibit divergent effects on subcortical volumes, with syndrome-specific influences from NS and NF1. Overall children with NS display decreased volumes in striatal and thalamic structures and children with NF1 display increased volumes in the hippocampus, amygdala, and thalamus.

CONCLUSIONS

Our study reveals the converging and diverging neuroanatomical effects of RASopathies on human neurodevelopment. The convergence of cortical effects on SA and CT indicates a shared influence of Ras/MAPK hyperactivation on the human brain. Therefore, considering these measures as objective outcome indicators for targeted treatments is imperative.

Childhood trauma moderates schizotypy-related brain morphology: analyses of 1182 healthy individuals from the ENIGMA schizotypy working group

BACKGROUND

Schizotypy represents an index of psychosis-proneness in the general population, often associated with childhood trauma exposure. Both schizotypy and childhood trauma are linked to structural brain alterations, and it is possible that trauma exposure moderates the extent of brain morphological differences associated with schizotypy.

METHODS

We addressed this question using data from a total of 1182 healthy adults (age range: 18-65 years old, 647 females/535 males), pooled from nine sites worldwide, contributing to the Enhancing NeuroImaging Genetics through Meta-Analysis (ENIGMA) Schizotypy working group. All participants completed both the Schizotypal Personality Questionnaire Brief version (SPQ-B), and the Childhood Trauma Questionnaire (CTQ), and underwent a 3D T1-weighted brain MRI scan from which regional indices of subcortical gray matter volume and cortical thickness were determined.

RESULTS

A series of multiple linear regressions revealed that differences in cortical thickness in four regions-of-interest were significantly associated with interactions between schizotypy and trauma; subsequent moderation analyses indicated that increasing levels of schizotypy were associated with thicker left caudal anterior cingulate gyrus, right middle temporal gyrus and insula, and thinner left caudal middle frontal gyrus, in people exposed to higher (but not low or average) levels of childhood trauma. This was found in the context of morphological changes directly associated with increasing levels of schizotypy or increasing levels of childhood trauma exposure.

CONCLUSIONS

These results suggest that alterations in brain regions critical for higher cognitive and integrative processes that are associated with schizotypy may be enhanced in individuals exposed to high levels of trauma.

The contribution of first-episode illness characteristics and cumulative antipsychotic usage to progressive structural brain changes over a long-term follow-up in schizophrenia

Exposure to antipsychotics as well as certain first-episode illness characteristics have been associated with greater gray matter (GM) deficits in the early phase of schizophrenia. Whether the first-episode illness characteristics affect the long-term progression of the structural brain changes remain unexplored. We therefore assessed the role of first-episode illness characteristics and life-time antipsychotic use in relation to long-term structural brain GM changes in schizophrenia. Individuals with schizophrenia (SZ, n = 29) and non-psychotic controls (n = 61) from the Northern Finland Birth Cohort 1966 underwent structural MRI at the ages of 34 (baseline) and 43 (follow-up) years. At follow-up, the average duration of illness was 19.8 years. Voxel-based morphometry was used to assess the effects of predictors on longitudinal GM changes in schizophrenia-relevant brain areas. Younger age of onset (AoO), higher cumulative antipsychotic dose and severity of symptoms were associated with greater GM deficits in the SZ group at follow-up. None of the first-episode illness characteristics were associated with longitudinal GM changes during 9-year follow-up period. We conclude that a younger AoO and high life-time antipsychotic use may contribute to progression of structural brain changes in schizophrenia. Apart from AoO, other first-episode illness characteristics may not contribute to longitudinal GM changes in midlife.

SAN: mitigating spatial covariance heterogeneity in cortical thickness data collected from multiple scanners or sites

In neuroimaging studies, combining data collected from multiple study sites or scanners is becoming common to increase the reproducibility of scientific discoveries. At the same time, unwanted variations arise by using different scanners (inter-scanner biases), which need to be corrected before downstream analyses to facilitate replicable research and prevent spurious findings. While statistical harmonization methods such as ComBat have become popular in mitigating inter-scanner biases in neuroimaging, recent methodological advances have shown that harmonizing heterogeneous covariances results in higher data quality. In vertex-level cortical thickness data, heterogeneity in spatial autocorrelation is a critical factor that affects covariance heterogeneity. Our work proposes a new statistical harmonization method called SAN (Spatial Autocorrelation Normalization) that preserves homogeneous covariance vertex-level cortical thickness data across different scanners. We use an explicit Gaussian process to characterize scanner-invariant and scanner-specific variations to reconstruct spatially homogeneous data across scanners. SAN is computationally feasible, and it easily allows the integration of existing harmonization methods. We demonstrate the utility of the proposed method using cortical thickness data from the Social Processes Initiative in the Neurobiology of the Schizophrenia(s) (SPINS) study. SAN is publicly available as an R package.

Lifespan reference curves for harmonizing multi-site regional brain white matter metrics from diffusion MRI

Age-related white matter (WM) microstructure maturation and decline occur throughout the human lifespan, complementing the process of gray matter development and degeneration. Here, we create normative lifespan reference curves for global and regional WM microstructure by harmonizing diffusion MRI (dMRI)-derived data from ten public datasets (N = 40,898 subjects; age: 3-95 years; 47.6% male). We tested three harmonization methods on regional diffusion tensor imaging (DTI) based fractional anisotropy (FA), a metric of WM microstructure, extracted using the ENIGMA-DTI pipeline. ComBat-GAM harmonization provided multi-study trajectories most consistent with known WM maturation peaks. Lifespan FA reference curves were validated with test-retest data and used to assess the effect of the ApoE4 risk factor for dementia in WM across the lifespan. We found significant associations between ApoE4 and FA in WM regions associated with neurodegenerative disease even in healthy individuals across the lifespan, with regional age-by-genotype interactions. Our lifespan reference curves and tools to harmonize new dMRI data to the curves are publicly available as eHarmonize (https://github.com/ahzhu/eharmonize).

Structural indices of brain aging in methamphetamine use disorder

BACKGROUND

Methamphetamine use is surging globally. It has been linked to premature stroke, Parkinsonism, and dementia, suggesting that it may accelerate brain aging.

METHODS

We performed a retrospective study to determine if structural indices of brain aging were more prevalent prior to old age (26 - 54 years) in individuals with Methamphetamine Use Disorder (MUD), who were in early abstinence (M ± SD = 22.1 ± 25.6 days) than in healthy control (HC) participants. We compared T1-weighted MRI brain scans in age- and sex-matched groups (n = 89/group) on three structural features of brain aging: the brain volume/cerebrospinal fluid (BV/CSF) index, volume of white matter hypointensities/lesions, and choroid plexus volume.

RESULTS

The MUD group had a lower mean BV/CSF index and larger volumes of white matter hypointensities and choroid plexus (p-values < 0.01). Regression analyses showed significant age-by-group effects, indicating different age trajectories of the BV/CSF index and choroid plexus volume, consistent with abnormal global brain atrophy and choroid plexus pathology in the MUD group. Significant age and group main effects reflected a larger volume of white matter hypointensities for older participants across groups and for the MUD group irrespective of age. None of the three measures of brain aging correlated significantly with recent use or duration of recent abstinence from methamphetamine.

CONCLUSIONS

Premature brain pathology, which may reflect cerebrovascular damage and dysfunction of the choroid plexus, occurs in people with MUD. Such pathology may affect cognition and thereby efficacy of behavioral treatments for MUD.

Machine learning prediction of future amyloid beta positivity in amyloid-negative individuals

BACKGROUND

The pathophysiology of Alzheimer's disease (AD) involves β -amyloid (A β ) accumulation. Early identification of individuals with abnormal β -amyloid levels is crucial, but A β quantification with positron emission tomography (PET) and cerebrospinal fluid (CSF) is invasive and expensive.

METHODS

We propose a machine learning framework using standard non-invasive (MRI, demographics, APOE, neuropsychology) measures to predict future A β -positivity in A β -negative individuals. We separately study A β -positivity defined by PET and CSF.

RESULTS

Cross-validated AUC for 4-year A β conversion prediction was 0.78 for the CSF-based and 0.68 for the PET-based A β definitions. Although not trained for the clinical status-change prediction, the CSF-based model excelled in predicting future mild cognitive impairment (MCI)/dementia conversion in cognitively normal/MCI individuals (AUCs, respectively, 0.76 and 0.89 with a separate dataset).

CONCLUSION

Standard measures have potential in detecting future A β -positivity and assessing conversion risk, even in cognitively normal individuals. The CSF-based definition led to better predictions than the PET-based definition.

Using brain structural neuroimaging measures to predict psychosis onset for individuals at clinical high-risk

Machine learning approaches using structural magnetic resonance imaging (sMRI) can be informative for disease classification, although their ability to predict psychosis is largely unknown. We created a model with individuals at CHR who developed psychosis later (CHR-PS+) from healthy controls (HCs) that can differentiate each other. We also evaluated whether we could distinguish CHR-PS+ individuals from those who did not develop psychosis later (CHR-PS-) and those with uncertain follow-up status (CHR-UNK). T1-weighted structural brain MRI scans from 1165 individuals at CHR (CHR-PS+, n = 144; CHR-PS-, n = 793; and CHR-UNK, n = 228), and 1029 HCs, were obtained from 21 sites. We used ComBat to harmonize measures of subcortical volume, cortical thickness and surface area data and corrected for non-linear effects of age and sex using a general additive model. CHR-PS+ (n = 120) and HC (n = 799) data from 20 sites served as a training dataset, which we used to build a classifier. The remaining samples were used external validation datasets to evaluate classifier performance (test, independent confirmatory, and independent group [CHR-PS- and CHR-UNK] datasets). The accuracy of the classifier on the training and independent confirmatory datasets was 85% and 73% respectively. Regional cortical surface area measures-including those from the right superior frontal, right superior temporal, and bilateral insular cortices strongly contributed to classifying CHR-PS+ from HC. CHR-PS- and CHR-UNK individuals were more likely to be classified as HC compared to CHR-PS+ (classification rate to HC: CHR-PS+, 30%; CHR-PS-, 73%; CHR-UNK, 80%). We used multisite sMRI to train a classifier to predict psychosis onset in CHR individuals, and it showed promise predicting CHR-PS+ in an independent sample. The results suggest that when considering adolescent brain development, baseline MRI scans for CHR individuals may be helpful to identify their prognosis. Future prospective studies are required about whether the classifier could be actually helpful in the clinical settings.

Functional magnetic resonance imaging in schizophrenia: current evidence, methodological advances, limitations and future directions

Functional neuroimaging emerged with great promise and has provided fundamental insights into the neurobiology of schizophrenia. However, it has faced challenges and criticisms, most notably a lack of clinical translation. This paper provides a comprehensive review and critical summary of the literature on functional neuroimaging, in particular functional magnetic resonance imaging (fMRI), in schizophrenia. We begin by reviewing research on fMRI biomarkers in schizophrenia and the clinical high risk phase through a historical lens, moving from case-control regional brain activation to global connectivity and advanced analytical approaches, and more recent machine learning algorithms to identify predictive neuroimaging features. Findings from fMRI studies of negative symptoms as well as of neurocognitive and social cognitive deficits are then reviewed. Functional neural markers of these symptoms and deficits may represent promising treatment targets in schizophrenia. Next, we summarize fMRI research related to antipsychotic medication, psychotherapy and psychosocial interventions, and neurostimulation, including treatment response and resistance, therapeutic mechanisms, and treatment targeting. We also review the utility of fMRI and data-driven approaches to dissect the heterogeneity of schizophrenia, moving beyond case-control comparisons, as well as methodological considerations and advances, including consortia and precision fMRI. Lastly, limitations and future directions of research in the field are discussed. Our comprehensive review suggests that, in order for fMRI to be clinically useful in the care of patients with schizophrenia, research should address potentially actionable clinical decisions that are routine in schizophrenia treatment, such as which antipsychotic should be prescribed or whether a given patient is likely to have persistent functional impairment. The potential clinical utility of fMRI is influenced by and must be weighed against cost and accessibility factors. Future evaluations of the utility of fMRI in prognostic and treatment response studies may consider including a health economics analysis.

Examining the association between posttraumatic stress disorder and disruptions in cortical networks identified using data-driven methods

Posttraumatic stress disorder (PTSD) is associated with lower cortical thickness (CT) in prefrontal, cingulate, and insular cortices in diverse trauma-affected samples. However, some studies have failed to detect differences between PTSD patients and healthy controls or reported that PTSD is associated with greater CT. Using data-driven dimensionality reduction, we sought to conduct a well-powered study to identify vulnerable networks without regard to neuroanatomic boundaries. Moreover, this approach enabled us to avoid the excessive burden of multiple comparison correction that plagues vertex-wise methods. We derived structural covariance networks (SCNs) by applying non-negative matrix factorization (NMF) to CT data from 961 PTSD patients and 1124 trauma-exposed controls without PTSD. We used regression analyses to investigate associations between CT within SCNs and PTSD diagnosis (with and without accounting for the potential confounding effect of trauma type) and symptom severity in the full sample. We performed additional regression analyses in subsets of the data to examine associations between SCNs and comorbid depression, childhood trauma severity, and alcohol abuse. NMF identified 20 unbiased SCNs, which aligned closely with functionally defined brain networks. PTSD diagnosis was most strongly associated with diminished CT in SCNs that encompassed the bilateral superior frontal cortex, motor cortex, insular cortex, orbitofrontal cortex, medial occipital cortex, anterior cingulate cortex, and posterior cingulate cortex. CT in these networks was significantly negatively correlated with PTSD symptom severity. Collectively, these findings suggest that PTSD diagnosis is associated with widespread reductions in CT, particularly within prefrontal regulatory regions and broader emotion and sensory processing cortical regions.

Improving brain age prediction with anatomical feature attention-enhanced 3D-CNN

Currently, significant progress has been made in predicting brain age from structural Magnetic Resonance Imaging (sMRI) data using deep learning techniques. However, despite the valuable structural information they contain, the traditional engineering features known as anatomical features have been largely overlooked in this context. To address this issue, we propose an attention-based network design that integrates anatomical and deep convolutional features, leveraging an anatomical feature attention (AFA) module to effectively capture salient anatomical features. In addition, we introduce a fully convolutional network, which simplifies the extraction of deep convolutional features and overcomes the high computational memory requirements associated with deep learning. Our approach outperforms several widely-used models on eight publicly available datasets (n = 2501), with a mean absolute error (MAE) of 2.20 years in predicting brain age. Comparisons with deep learning models lacking the AFA module demonstrate that our fusion model effectively improves overall performance. These findings provide a promising approach for combining anatomical and deep convolutional features from sMRI data to predict brain age, with potential applications in clinical diagnosis and treatment, particularly for populations with age-related cognitive decline or neurological disorders.

Efficacy of MRI data harmonization in the age of machine learning: a multicenter study across 36 datasets

Pooling publicly-available MRI data from multiple sites allows to assemble extensive groups of subjects, increase statistical power, and promote data reuse with machine learning techniques. The harmonization of multicenter data is necessary to reduce the confounding effect associated with non-biological sources of variability in the data. However, when applied to the entire dataset before machine learning, the harmonization leads to data leakage, because information outside the training set may affect model building, and potentially falsely overestimate performance. We propose a 1) measurement of the efficacy of data harmonization; 2) harmonizer transformer, i.e., an implementation of the ComBat harmonization allowing its encapsulation among the preprocessing steps of a machine learning pipeline, avoiding data leakage by design. We tested these tools using brain T1-weighted MRI data from 1740 healthy subjects acquired at 36 sites. After harmonization, the site effect was removed or reduced, and we showed the data leakage effect in predicting individual age from MRI data, highlighting that introducing the harmonizer transformer into a machine learning pipeline allows for avoiding data leakage by design.

Beyond the Global Brain Differences: Intraindividual Variability Differences in 1q21.1 Distal and 15q11.2 BP1-BP2 Deletion Carriers

BACKGROUND

Carriers of the 1q21.1 distal and 15q11.2 BP1-BP2 copy number variants exhibit regional and global brain differences compared with noncarriers. However, interpreting regional differences is challenging if a global difference drives the regional brain differences. Intraindividual variability measures can be used to test for regional differences beyond global differences in brain structure.

METHODS

Magnetic resonance imaging data were used to obtain regional brain values for 1q21.1 distal deletion (n = 30) and duplication (n = 27) and 15q11.2 BP1-BP2 deletion (n = 170) and duplication (n = 243) carriers and matched noncarriers (n = 2350). Regional intra-deviation scores, i.e., the standardized difference between an individual's regional difference and global difference, were used to test for regional differences that diverge from the global difference.

RESULTS

For the 1q21.1 distal deletion carriers, cortical surface area for regions in the medial visual cortex, posterior cingulate, and temporal pole differed less and regions in the prefrontal and superior temporal cortex differed more than the global difference in cortical surface area. For the 15q11.2 BP1-BP2 deletion carriers, cortical thickness in regions in the medial visual cortex, auditory cortex, and temporal pole differed less and the prefrontal and somatosensory cortex differed more than the global difference in cortical thickness.

CONCLUSIONS

We find evidence for regional effects beyond differences in global brain measures in 1q21.1 distal and 15q11.2 BP1-BP2 copy number variants. The results provide new insight into brain profiling of the 1q21.1 distal and 15q11.2 BP1-BP2 copy number variants, with the potential to increase understanding of the mechanisms involved in altered neurodevelopment.

Intrusive Traumatic Re-Experiencing Domain: Functional Connectivity Feature Classification by the ENIGMA PTSD Consortium

Background

Intrusive traumatic re-experiencing domain (ITRED) was recently introduced as a novel perspective on posttraumatic psychopathology, proposing to focus research of posttraumatic stress disorder (PTSD) on the unique symptoms of intrusive and involuntary re-experiencing of the trauma, namely, intrusive memories, nightmares, and flashbacks. The aim of the present study was to explore ITRED from a neural network connectivity perspective.

Methods

Data were collected from 9 sites taking part in the ENIGMA (Enhancing Neuro Imaging Genetics through Meta Analysis) PTSD Consortium (n= 584) and included itemized PTSD symptom scores and resting-state functional connectivity (rsFC) data. We assessed the utility of rsFC in classifying PTSD, ITRED-only (no PTSD diagnosis), and trauma-exposed (TE)-only (no PTSD or ITRED) groups using a machine learning approach, examining well-known networks implicated in PTSD. A random forest classification model was built on a training set using cross-validation, and the averaged cross-validation model performance for classification was evaluated using the area under the curve. The model was tested using a fully independent portion of the data (test dataset), and the test area under the curve was evaluated.

Results

rsFC signatures differentiated TE-only participants from PTSD and ITRED-only participants at about 60% accuracy. Conversely, rsFC signatures did not differentiate PTSD from ITRED-only individuals (45% accuracy). Common features differentiating TE-only participants from PTSD and ITRED-only participants mainly involved default mode network-related pathways. Some unique features, such as connectivity within the frontoparietal network, differentiated TE-only participants from one group (PTSD or ITRED-only) but to a lesser extent from the other group.

Conclusions

Neural network connectivity supports ITRED as a novel neurobiologically based approach to classifying posttrauma psychopathology.

Recalibrating single-study effect sizes using hierarchical Bayesian models

Introduction

There are growing concerns about commonly inflated effect sizes in small neuroimaging studies, yet no study has addressed recalibrating effect size estimates for small samples. To tackle this issue, we propose a hierarchical Bayesian model to adjust the magnitude of single-study effect sizes while incorporating a tailored estimation of sampling variance.

Methods

We estimated the effect sizes of case-control differences on brain structural features between individuals who were dependent on alcohol, nicotine, cocaine, methamphetamine, or cannabis and non-dependent participants for 21 individual studies (Total cases: 903; Total controls: 996). Then, the study-specific effect sizes were modeled using a hierarchical Bayesian approach in which the parameters of the study-specific effect size distributions were sampled from a higher-order overarching distribution. The posterior distribution of the overarching and study-specific parameters was approximated using the Gibbs sampling method.

Results

The results showed shrinkage of the posterior distribution of the study-specific estimates toward the overarching estimates given the original effect sizes observed in individual studies. Differences between the original effect sizes (i.e., Cohen's d) and the point estimate of the posterior distribution ranged from 0 to 0.97. The magnitude of adjustment was negatively correlated with the sample size (r = -0.27, p < 0.001) and positively correlated with empirically estimated sampling variance (r = 0.40, p < 0.001), suggesting studies with smaller samples and larger sampling variance tended to have greater adjustments.

Discussion

Our findings demonstrate the utility of the hierarchical Bayesian model in recalibrating single-study effect sizes using information from similar studies. This suggests that Bayesian utilization of existing knowledge can be an effective alternative approach to improve the effect size estimation in individual studies, particularly for those with smaller samples.

Neuroimaging-based classification of PTSD using data-driven computational approaches: A multisite big data study from the ENIGMA-PGC PTSD consortium

BACKGROUND

Recent advances in data-driven computational approaches have been helpful in devising tools to objectively diagnose psychiatric disorders. However, current machine learning studies limited to small homogeneous samples, different methodologies, and different imaging collection protocols, limit the ability to directly compare and generalize their results. Here we aimed to classify individuals with PTSD versus controls and assess the generalizability using a large heterogeneous brain datasets from the ENIGMA-PGC PTSD Working group.

METHODS

We analyzed brain MRI data from 3,477 structural-MRI; 2,495 resting state-fMRI; and 1,952 diffusion-MRI. First, we identified the brain features that best distinguish individuals with PTSD from controls using traditional machine learning methods. Second, we assessed the utility of the denoising variational autoencoder (DVAE) and evaluated its classification performance. Third, we assessed the generalizability and reproducibility of both models using leave-one-site-out cross-validation procedure for each modality.

RESULTS

We found lower performance in classifying PTSD vs. controls with data from over 20 sites (60 % test AUC for s-MRI, 59 % for rs-fMRI and 56 % for d-MRI), as compared to other studies run on single-site data. The performance increased when classifying PTSD from HC without trauma history in each modality (75 % AUC). The classification performance remained intact when applying the DVAE framework, which reduced the number of features. Finally, we found that the DVAE framework achieved better generalization to unseen datasets compared with the traditional machine learning frameworks, albeit performance was slightly above chance.

CONCLUSION

These results have the potential to provide a baseline classification performance for PTSD when using large scale neuroimaging datasets. Our findings show that the control group used can heavily affect classification performance. The DVAE framework provided better generalizability for the multi-site data. This may be more significant in clinical practice since the neuroimaging-based diagnostic DVAE classification models are much less site-specific, rendering them more generalizable.

Diffusion-based structural connectivity patterns of multiple sclerosis phenotypes

BACKGROUND

We aimed to describe the severity of the changes in brain diffusion-based connectivity as multiple sclerosis (MS) progresses and the microstructural characteristics of these networks that are associated with distinct MS phenotypes.

METHODS

Clinical information and brain MRIs were collected from 221 healthy individuals and 823 people with MS at 8 MAGNIMS centres. The patients were divided into four clinical phenotypes: clinically isolated syndrome, relapsing-remitting, secondary progressive and primary progressive. Advanced tractography methods were used to obtain connectivity matrices. Then, differences in whole-brain and nodal graph-derived measures, and in the fractional anisotropy of connections between groups were analysed. Support vector machine algorithms were used to classify groups.

RESULTS

Clinically isolated syndrome and relapsing-remitting patients shared similar network changes relative to controls. However, most global and local network properties differed in secondary progressive patients compared with the other groups, with lower fractional anisotropy in most connections. Primary progressive participants had fewer differences in global and local graph measures compared with clinically isolated syndrome and relapsing-remitting patients, and reductions in fractional anisotropy were only evident for a few connections. The accuracy of support vector machine to discriminate patients from healthy controls based on connection was 81%, and ranged between 64% and 74% in distinguishing among the clinical phenotypes.

CONCLUSIONS

In conclusion, brain connectivity is disrupted in MS and has differential patterns according to the phenotype. Secondary progressive is associated with more widespread changes in connectivity. Additionally, classification tasks can distinguish between MS types, with subcortical connections being the most important factor.

STAREE-Mind Imaging Study: a randomised placebo-controlled trial of atorvastatin for prevention of cerebrovascular decline and neurodegeneration in older individuals

Introduction

Cerebrovascular disease and neurodegeneration are causes of cognitive decline and dementia, for which primary prevention options are currently lacking. Statins are well-tolerated and widely available medications that potentially have neuroprotective effects. The STAREE-Mind Imaging Study is a randomised, double-blind, placebo-controlled clinical trial that will investigate the impact of atorvastatin on markers of neurovascular health and brain atrophy in a healthy, older population using MRI. This is a nested substudy of the 'Statins for Reducing Events in the Elderly' (STAREE) primary prevention trial.

Methods

Participants aged 70 years or older (n=340) will be randomised to atorvastatin or placebo. Comprehensive brain MRI assessment will be undertaken at baseline and up to 4 years follow-up, including structural, diffusion, perfusion and susceptibility imaging. The primary outcome measures will be change in brain free water fraction (a composite marker of vascular leakage, neuroinflammation and neurodegeneration) and white matter hyperintensity volume (small vessel disease). Secondary outcomes will include change in perivascular space volume (glymphatic drainage), cortical thickness, hippocampal volume, microbleeds and lacunae, prefrontal cerebral perfusion and white matter microstructure.

Ethics and dissemination

Academic publications from this work will address the current uncertainty regarding the impact of statins on brain structure and vascular integrity. This study will inform the utility of repurposing these well-tolerated, inexpensive and widely available drugs for primary prevention of neurological outcomes in older individuals. Ethics approval was given by Monash University Human Research Ethics Committee, Protocol 12206.

Trial registration number

ClinicalTrials.gov Identifier: NCT05586750.

Multimodal Neuroimaging Summary Scores as Neurobiological Markers of Psychosis

BACKGROUND AND HYPOTHESIS

Structural brain alterations are well-established features of schizophrenia but they do not effectively predict disease/disease risk. Similar to polygenic risk scores in genetics, we integrated multifactorial aspects of brain structure into a summary "Neuroscore" and examined its potential as a marker of disease.

STUDY DESIGN

We extracted measures from T1-weighted scans and diffusion tensor imaging (DTI) models from three studies with schizophrenia and healthy individuals. We calculated individual-level summary scores (Neuroscores) for T1-weighted and DTI measures and a combined score (Multimodal Neuroscore-MM). We assessed each score's ability to differentiate schizophrenia cases from controls and its relationship to clinical symptomatology, intelligence quotient (IQ), and medication dosage. We assessed Neuroscore specificity by performing all analyses in a more inclusive psychosis sample and by using scores generated from MDD effect sizes.

STUDY RESULTS

All Neuroscores significantly differentiated schizophrenia cases from controls (T1 d = 0.56, DTI d = 0.29, MM d = 0.64) to a greater degree than individual brain regions. Higher Neuroscores (ie, increased liability) were associated with lower IQ (T1 β = -0.26, DTI β = -0.15, MM β = -0.30). Higher T1-weighted Neuroscores were associated with higher positive and negative symptom severity (Positive β = 0.21, Negative β = 0.16); Higher Multimodal Neuroscores were associated with higher positive symptom severity (β = 0.30). SZ Neuroscores outperformed MDD Neuroscores in predicting IQ (T1: z = 3.5, q = 0.0007; MM: z = 1.8, q = 0.05).

CONCLUSIONS

Neuroscores are a step toward leveraging widespread structural brain alterations in psychosis to identify robust neurobiological markers of disease. Future studies will assess ways to improve neuroscore calculation, including developing the optimal methods to calculate neuroscores and considering disorder overlap.

Effects of MRI scanner manufacturers in classification tasks with deep learning models

Deep learning has become a leading subset of machine learning and has been successfully employed in diverse areas, ranging from natural language processing to medical image analysis. In medical imaging, researchers have progressively turned towards multi-center neuroimaging studies to address complex questions in neuroscience, leveraging larger sample sizes and aiming to enhance the accuracy of deep learning models. However, variations in image pixel/voxel characteristics can arise between centers due to factors including differences in magnetic resonance imaging scanners. Such variations create challenges, particularly inconsistent performance in machine learning-based approaches, often referred to as domain shift, where the trained models fail to achieve satisfactory or improved results when confronted with dissimilar test data. This study analyzes the performance of multiple disease classification tasks using multi-center MRI data obtained from three widely used scanner manufacturers (GE, Philips, and Siemens) across several deep learning-based networks. Furthermore, we investigate the efficacy of mitigating scanner vendor effects using ComBat-based harmonization techniques when applied to multi-center datasets of 3D structural MR images. Our experimental results reveal a substantial decline in classification performance when models trained on one type of scanner manufacturer are tested with data from different manufacturers. Moreover, despite applying ComBat-based harmonization, the harmonized images do not demonstrate any noticeable performance enhancement for disease classification tasks.

Cortical morphology in patients with the deficit and non-deficit syndrome of schizophrenia: a worldwide meta- and mega-analyses

Converging evidence suggests that schizophrenia (SZ) with primary, enduring negative symptoms (i.e., Deficit SZ (DSZ)) represents a distinct entity within the SZ spectrum while the neurobiological underpinnings remain undetermined. In the largest dataset of DSZ and Non-Deficit (NDSZ), we conducted a meta-analysis of data from 1560 individuals (168 DSZ, 373 NDSZ, 1019 Healthy Controls (HC)) and a mega-analysis of a subsampled data from 944 individuals (115 DSZ, 254 NDSZ, 575 HC) collected across 9 worldwide research centers of the ENIGMA SZ Working Group (8 in the mega-analysis), to clarify whether they differ in terms of cortical morphology. In the meta-analysis, sites computed effect sizes for differences in cortical thickness and surface area between SZ and control groups using a harmonized pipeline. In the mega-analysis, cortical values of individuals with schizophrenia and control participants were analyzed across sites using mixed-model ANCOVAs. The meta-analysis of cortical thickness showed a converging pattern of widespread thinner cortex in fronto-parietal regions of the left hemisphere in both DSZ and NDSZ, when compared to HC. However, DSZ have more pronounced thickness abnormalities than NDSZ, mostly involving the right fronto-parietal cortices. As for surface area, NDSZ showed differences in fronto-parietal-temporo-occipital cortices as compared to HC, and in temporo-occipital cortices as compared to DSZ. Although DSZ and NDSZ show widespread overlapping regions of thinner cortex as compared to HC, cortical thinning seems to better typify DSZ, being more extensive and bilateral, while surface area alterations are more evident in NDSZ. Our findings demonstrate for the first time that DSZ and NDSZ are characterized by different neuroimaging phenotypes, supporting a nosological distinction between DSZ and NDSZ and point toward the separate disease hypothesis.

A systematic comparison of VBM pipelines and their application to age prediction

Voxel-based morphometry (VBM) analysis is commonly used for localized quantification of gray matter volume (GMV). Several alternatives exist to implement a VBM pipeline. However, how these alternatives compare and their utility in applications, such as the estimation of aging effects, remain largely unclear. This leaves researchers wondering which VBM pipeline they should use for their project. In this study, we took a user-centric perspective and systematically compared five VBM pipelines, together with registration to either a general or a study-specific template, utilizing three large datasets (n>500 each). Considering the known effect of aging on GMV, we first compared the pipelines in their ability of individual-level age prediction and found markedly varied results. To examine whether these results arise from systematic differences between the pipelines, we classified them based on their GMVs, resulting in near-perfect accuracy. To gain deeper insights, we examined the impact of different VBM steps using the region-wise similarity between pipelines. The results revealed marked differences, largely driven by segmentation and registration steps. We observed large variability in subject-identification accuracies, highlighting the interpipeline differences in individual-level quantification of GMV. As a biologically meaningful criterion we correlated regional GMV with age. The results were in line with the age-prediction analysis, and two pipelines, CAT and the combination of fMRIPrep for tissue characterization with FSL for registration, reflected age information better.

Improved estimation of the risk of manic relapse by combining clinical and brain scan data

INTRODUCTION

Estimating the risk of manic relapse could help the psychiatrist individually adjust the treatment to the risk. Some authors have attempted to estimate this risk from baseline clinical data. Still, no studies have assessed whether the estimation could improve by adding structural magnetic resonance imaging (MRI) data. We aimed to evaluate it.

MATERIAL AND METHODS

We followed a cohort of 78 patients with a manic episode without mixed symptoms (bipolar type I or schizoaffective disorder) at 2-4-6-9-12-15-18 months and up to 10 years. Within a cross-validation scheme, we created and evaluated a Cox lasso model to estimate the risk of manic relapse using both clinical and MRI data.

RESULTS

The model successfully estimated the risk of manic relapse (Cox regression of the time to relapse as a function of the estimated risk: hazard ratio (HR)=2.35, p=0.027; area under the curve (AUC)=0.65, expected calibration error (ECE)<0.2). The most relevant variables included in the model were the diagnosis of schizoaffective disorder, poor impulse control, unusual thought content, and cerebellum volume decrease. The estimations were poorer when we used clinical or MRI data separately.

CONCLUSION

Combining clinical and MRI data may improve the risk of manic relapse estimation after a manic episode. We provide a website that estimates the risk according to the model to facilitate replication by independent groups before translation to clinical settings.

MISPEL: A supervised deep learning harmonization method for multi-scanner neuroimaging data

Large-scale data obtained from aggregation of already collected multi-site neuroimaging datasets has brought benefits such as higher statistical power, reliability, and robustness to the studies. Despite these promises from growth in sample size, substantial technical variability stemming from differences in scanner specifications exists in the aggregated data and could inadvertently bias any downstream analyses on it. Such a challenge calls for data normalization and/or harmonization frameworks, in addition to comprehensive criteria to estimate the scanner-related variability and evaluate the harmonization frameworks. In this study, we propose MISPEL (Multi-scanner Image harmonization via Structure Preserving Embedding Learning), a supervised multi-scanner harmonization method that is naturally extendable to more than two scanners. We also designed a set of criteria to investigate the scanner-related technical variability and evaluate the harmonization techniques. As an essential requirement of our criteria, we introduced a multi-scanner matched dataset of 3T T1 images across four scanners, which, to the best of our knowledge is one of the few datasets of this kind. We also investigated our evaluations using two popular segmentation frameworks: FSL and segmentation in statistical parametric mapping (SPM). Lastly, we compared MISPEL to popular methods of normalization and harmonization, namely White Stripe, RAVEL, and CALAMITI. MISPEL outperformed these methods and is promising for many other neuroimaging modalities.

[Identification of Characteristic lncRNA Molecular Markers in Osteoarthritis by Integrating GEO Database and Machine Learning Strategies and Experimental Validation]

Objective

To screen for long non-coding RNA (lncRNA) molecular markers characteristic of osteoarthritis (OA) by utilizing the Gene Expression Omnibus (GEO) database combined with machine learning.

Methods

The samples of 185 OA patients and 76 healthy individuals as normal controls were included in the study. GEO datasets were screened for differentially expressed lncRNAs. Three algorithms, the least absolute shrinkage and selection operator (LASSO), support vector machine recursive feature elimination (SVM-RFE), and random forest (RF), were used to screen for candidate lncRNA models and receiver operating characteristic (ROC) curves were plotted to evaluate the models. We collected the peripheral blood samples of 30 clinical OA patients and 15 health controls and measured the immunoinflammatory indicators. RT-PCR was performed for quantitative analysis of the expression of lncRNA molecular markers in peripheral blood mononuclear cells (PBMC). Pearson analysis was performed to examine the correlation between lncRNA and indicators for inflammation of the immune system.

Results

A total of 14 key markers were identified with LASSO, 6 genes were identified with SVM-RFE, and 24 genes were identified with RF. Venn diagram was used to screen for overlapping genes identified with the three algorithms, showing HOTAIR, H19, MIR155 HG, and NKILA to be the overlapping genes. The ROC curves showed that these four lncRNAs all had an area under the curve ( AUC) greater than 0.7. The RT-PCR findings revealed relatively elevated expression of HOTAIR, H19, and MIR155HG and decreased expression of NKILA in the PBMC of OA patients compared with those of the normal group ( P<0.01). The results were consistent with the bioinformatics predictions. Pearson analysis showed that the candidate lncRNAs were correlated with clinical indicators for inflammation.

Conclusion

HOTAIR, H19, MIR155 HG, and NKILA can be used as molecular markers for the clinical diagnosis of OA and are correlate with clinical indicators of inflammation of the immune system.

A Brain Morphometry Study with Across-Site Harmonization Using a ComBat-Generalized Additive Model in Children and Adolescents

Regional anatomical structures of the brain are intimately connected to functions corresponding to specific regions and the temporospatial pattern of genetic expression and their functions from the fetal period to old age. Therefore, quantitative brain morphometry has often been employed in neuroscience investigations, while controlling for the scanner effect of the scanner is a critical issue for ensuring accuracy in brain morphometric studies of rare orphan diseases due to the lack of normal reference values available for multicenter studies. This study aimed to provide across-site normal reference values of global and regional brain volumes for each sex and age group in children and adolescents. We collected magnetic resonance imaging (MRI) examinations of 846 neurotypical participants aged 6.0-17.9 years (339 male and 507 female participants) from 5 institutions comprising healthy volunteers or neurotypical patients without neurological disorders, neuropsychological disorders, or epilepsy. Regional-based analysis using the CIVET 2.1.0. pipeline provided regional brain volumes, and the measurements were across-site combined using ComBat-GAM harmonization. The normal reference values of global and regional brain volumes and lateral indices in our study could be helpful for evaluating the characteristics of the brain morphology of each individual in a clinical setting and investigating the brain morphology of ultra-rare diseases.

Mitochondrial function-associated genes underlie cortical atrophy in prodromal synucleinopathies

Isolated rapid eye movement sleep behaviour disorder (iRBD) is a sleep disorder characterized by the loss of rapid eye movement sleep muscle atonia and the appearance of abnormal movements and vocalizations during rapid eye movement sleep. It is a strong marker of incipient synucleinopathy such as dementia with Lewy bodies and Parkinson's disease. Patients with iRBD already show brain changes that are reminiscent of manifest synucleinopathies including brain atrophy. However, the mechanisms underlying the development of this atrophy remain poorly understood. In this study, we performed cutting-edge imaging transcriptomics and comprehensive spatial mapping analyses in a multicentric cohort of 171 polysomnography-confirmed iRBD patients [67.7 ± 6.6 (49-87) years; 83% men] and 238 healthy controls [66.6 ± 7.9 (41-88) years; 77% men] with T1-weighted MRI to investigate the gene expression and connectivity patterns associated with changes in cortical thickness and surface area in iRBD. Partial least squares regression was performed to identify the gene expression patterns underlying cortical changes in iRBD. Gene set enrichment analysis and virtual histology were then done to assess the biological processes, cellular components, human disease gene terms, and cell types enriched in these gene expression patterns. We then used structural and functional neighbourhood analyses to assess whether the atrophy patterns in iRBD were constrained by the brain's structural and functional connectome. Moreover, we used comprehensive spatial mapping analyses to assess the specific neurotransmitter systems, functional networks, cytoarchitectonic classes, and cognitive brain systems associated with cortical changes in iRBD. All comparisons were tested against null models that preserved spatial autocorrelation between brain regions and compared to Alzheimer's disease to assess the specificity of findings to synucleinopathies. We found that genes involved in mitochondrial function and macroautophagy were the strongest contributors to the cortical thinning occurring in iRBD. Moreover, we demonstrated that cortical thinning was constrained by the brain's structural and functional connectome and that it mapped onto specific networks involved in motor and planning functions. In contrast with cortical thickness, changes in cortical surface area were related to distinct genes, namely genes involved in the inflammatory response, and to different spatial mapping patterns. The gene expression and connectivity patterns associated with iRBD were all distinct from those observed in Alzheimer's disease. In summary, this study demonstrates that the development of brain atrophy in synucleinopathies is constrained by specific genes and networks.

Brain morphometry in older adults with and without dementia using extremely rapid structural scans

T1-weighted structural MRI is widely used to measure brain morphometry (e.g., cortical thickness and subcortical volumes). Accelerated scans as fast as one minute or less are now available but it is unclear if they are adequate for quantitative morphometry. Here we compared the measurement properties of a widely adopted 1.0 mm resolution scan from the Alzheimer's Disease Neuroimaging Initiative (ADNI = 5'12'') with two variants of highly accelerated 1.0 mm scans (compressed-sensing, CSx6 = 1'12''; and wave-controlled aliasing in parallel imaging, WAVEx9 = 1'09'') in a test-retest study of 37 older adults aged 54 to 86 (including 19 individuals diagnosed with a neurodegenerative dementia). Rapid scans produced highly reliable morphometric measures that largely matched the quality of morphometrics derived from the ADNI scan. Regions of lower reliability and relative divergence between ADNI and rapid scan alternatives tended to occur in midline regions and regions with susceptibility-induced artifacts. Critically, the rapid scans yielded morphometric measures similar to the ADNI scan in regions of high atrophy. The results converge to suggest that, for many current uses, extremely rapid scans can replace longer scans. As a final test, we explored the possibility of a 0'49'' 1.2 mm CSx6 structural scan, which also showed promise. Rapid structural scans may benefit MRI studies by shortening the scan session and reducing cost, minimizing opportunity for movement, creating room for additional scan sequences, and allowing for the repetition of structural scans to increase precision of the estimates.

Improving Alzheimer's Disease Diagnosis With Multi-Modal PET Embedding Features by a 3D Multi-Task MLP-Mixer Neural Network

Positron emission tomography (PET) with fluorodeoxyglucose (FDG) or florbetapir (AV45) has been proved effective in the diagnosis of Alzheimer's disease. However, the expensive and radioactive nature of PET has limited its application. Here, employing multi-layer perceptron mixer architecture, we present a deep learning model, namely 3-dimensional multi-task multi-layer perceptron mixer, for simultaneously predicting the standardized uptake value ratios (SUVRs) for FDG-PET and AV45-PET from the cheap and widely used structural magnetic resonance imaging data, and the model can be further used for Alzheimer's disease diagnosis based on embedding features derived from SUVR prediction. Experiment results demonstrate the high prediction accuracy of the proposed method for FDG/AV45-PET SUVRs, where we achieved Pearson's correlation coefficients of 0.66 and 0.61 respectively between the estimated and actual SUVR and the estimated SUVRs also show high sensitivity and distinct longitudinal patterns for different disease status. By taking into account PET embedding features, the proposed method outperforms other competing methods on five independent datasets in the diagnosis of Alzheimer's disease and discriminating between stable and progressive mild cognitive impairments, achieving the area under receiver operating characteristic curves of 0.968 and 0.776 respectively on ADNI dataset, and generalizes better to other external datasets. Moreover, the top-weighted patches extracted from the trained model involve important brain regions related to Alzheimer's disease, suggesting good biological interpretability of our proposed method."

Neurobiological Clusters Are Associated With Trajectories of Overall Psychopathology in Youth

BACKGROUND

Integrating multiple neuroimaging modalities to identify clusters of individuals and then associating these clusters with psychopathology is a promising approach for understanding neurobiological mechanisms that underlie psychopathology and the extent to which these features are associated with clinical symptoms.

METHODS

We leveraged neuroimaging data from T1-weighted, diffusion-weighted, and resting-state functional magnetic resonance images from the Adolescent Brain Cognitive Development (ABCD) Study (N = 8035) and used similarity network fusion and spectral clustering to identify subgroups of participants. We examined neuroimaging measures as a function of clustering profiles using 1, 2, or 3 imaging modalities (i.e., data combinations), calculated the stability of the clustering assignment in each respective data combination, and compared the consistency of clusters across different data combinations. We then compared the extent to which clusters were associated with overall psychopathology at the baseline assessment and at 2 yearly follow-up visits.

RESULTS

Each data combination resulted in optimal clusters ranging from 2 to 4 subgroups for each data combination. Clusters were stable across subsampling of the ABCD Study cohort. Widespread structural measures (surface area, fractional anisotropy, and mean diffusivity) were important features contributing to clustering across different data combinations. Five of the seven data combinations were associated with overall psychopathology, both at baseline and over time (d = 0.08-0.41). Generally, lower global cortical volume and surface area, widespread reduced fractional anisotropy, and increased radial diffusivity were associated with increased overall psychopathology.

CONCLUSIONS

Profiles constructed from neuroimaging data combinations are associated with concurrent and future psychopathology trajectories.

ComBat Harmonization: Empirical Bayes versus fully Bayes approaches

Studying small effects or subtle neuroanatomical variation requires large-scale sample size data. As a result, combining neuroimaging data from multiple datasets is necessary. Variation in acquisition protocols, magnetic field strength, scanner build, and many other non-biologically related factors can introduce undesirable bias into studies. Hence, harmonization is required to remove the bias-inducing factors from the data. ComBat is one of the most common methods applied to features from structural images. ComBat models the data using a hierarchical Bayesian model and uses the empirical Bayes approach to infer the distribution of the unknown factors. The empirical Bayes harmonization method is computationally efficient and provides valid point estimates. However, it tends to underestimate uncertainty. This paper investigates a new approach, fully Bayesian ComBat, where Monte Carlo sampling is used for statistical inference. When comparing fully Bayesian and empirical Bayesian ComBat, we found Empirical Bayesian ComBat more effectively removed scanner strength information and was much more computationally efficient. Conversely, fully Bayesian ComBat better preserved biological disease and age-related information while performing more accurate harmonization on traveling subjects. The fully Bayesian approach generates a rich posterior distribution, which is useful for generating simulated imaging features for improving classifier performance in a limited data setting. We show the generative capacity of our model for augmenting and improving the detection of patients with Alzheimer's disease. Posterior distributions for harmonized imaging measures can also be used for brain-wide uncertainty comparison and more principled downstream statistical analysis.Code for our new fully Bayesian ComBat extension is available at https://github.com/batmanlab/BayesComBat.

Challenges of implementing computer-aided diagnostic models for neuroimages in a clinical setting

Advances in artificial intelligence have cultivated a strong interest in developing and validating the clinical utilities of computer-aided diagnostic models. Machine learning for diagnostic neuroimaging has often been applied to detect psychological and neurological disorders, typically on small-scale datasets or data collected in a research setting. With the collection and collation of an ever-growing number of public datasets that researchers can freely access, much work has been done in adapting machine learning models to classify these neuroimages by diseases such as Alzheimer's, ADHD, autism, bipolar disorder, and so on. These studies often come with the promise of being implemented clinically, but despite intense interest in this topic in the laboratory, limited progress has been made in clinical implementation. In this review, we analyze challenges specific to the clinical implementation of diagnostic AI models for neuroimaging data, looking at the differences between laboratory and clinical settings, the inherent limitations of diagnostic AI, and the different incentives and skill sets between research institutions, technology companies, and hospitals. These complexities need to be recognized in the translation of diagnostic AI for neuroimaging from the laboratory to the clinic.

Image harmonization: A review of statistical and deep learning methods for removing batch effects and evaluation metrics for effective harmonization

Magnetic resonance imaging and computed tomography from multiple batches (e.g. sites, scanners, datasets, etc.) are increasingly used alongside complex downstream analyses to obtain new insights into the human brain. However, significant confounding due to batch-related technical variation, called batch effects, is present in this data; direct application of downstream analyses to the data may lead to biased results. Image harmonization methods seek to remove these batch effects and enable increased generalizability and reproducibility of downstream results. In this review, we describe and categorize current approaches in statistical and deep learning harmonization methods. We also describe current evaluation metrics used to assess harmonization methods and provide a standardized framework to evaluate newly-proposed methods for effective harmonization and preservation of biological information. Finally, we provide recommendations to end-users to advocate for more effective use of current methods and to methodologists to direct future efforts and accelerate development of the field.

Harmonizing PTSD severity scales across instruments and sites

OBJECTIVE

The variety of instruments used to assess posttraumatic stress disorder (PTSD) allows for flexibility, but also creates challenges for data synthesis. The objective of this work was to use a multisite mega analysis to derive quantitative recommendations for equating scores across measures of PTSD severity.

METHOD

Empirical Bayes harmonization and linear models were used to describe and mitigate site and covariate effects. Quadratic models for converting scores across PTSD assessments were constructed using bootstrapping and tested on hold out data.

RESULTS

We aggregated 17 data sources and compiled an n = 5,634 sample of individuals who were assessed for PTSD symptoms. We confirmed our hypothesis that harmonization and covariate adjustments would significantly improve inference of scores across instruments. Harmonization significantly reduced cross-dataset variance (28%, p < .001), and models for converting scores across instruments were well fit (median R² = 0.985) with an average root mean squared error of 1.46 on sum scores.

CONCLUSIONS

These methods allow PTSD symptom severity to be placed on multiple scales and offers interesting empirical perspectives on the role of harmonization in the behavioral sciences. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Thalamic nuclei volumes in schizophrenia and bipolar spectrum disorders - Associations with diagnosis and clinical characteristics

BACKGROUND

The thalamus is central to brain functions ranging from primary sensory processing to higher-order cognition. Structural deficits in thalamic association nuclei such as the pulvinar and mediodorsal nuclei have previously been reported in schizophrenia. However, the specificity with regards to clinical presentation, and whether or not bipolar disorder (BD) is associated with similar alterations is unclear.

METHODS

We investigated thalamic nuclei volumes in 334 patients with schizophrenia spectrum disorders (SSD) (median age 29 years, 59 % male), 322 patients with BD (30 years, 40 % male), and 826 healthy controls (HC) (34 years, 54 % male). Volumes of 25 thalamic nuclei were extracted from T1-weighted magnetic resonance imaging using an automated Bayesian segmentation method and compared between groups. Furthermore, we explored associations with clinical characteristics across diagnostic groups, including psychotic and mood symptoms and medication use, as well as diagnostic subtype in BD.

RESULTS

Significantly smaller volumes were found in the mediodorsal, pulvinar, and lateral and medial geniculate thalamic nuclei in SSD. Similarly, smaller volumes were found in BD in the same four regions, but mediodorsal nucleus volume alterations were limited to its lateral part and pulvinar alterations to its anterior region. Smaller volumes in BD compared to HC were seen only in BD type I, not BD type II. Across diagnoses, having more negative symptoms was associated with smaller pulvinar volumes.

CONCLUSIONS

Structural alterations were found in both SSD and BD, mainly in the thalamic association nuclei. Structural deficits in the pulvinar may be of relevance for negative symptoms.

Long-term consequences of neonatal encephalopathy in the hypothermia era: protocol for a follow-up cohort study at 9 years of age

INTRODUCTION

Therapeutic hypothermia (TH) became the standard of care treatment for neonates with moderate and severe neonatal encephalopathy (NE) in most industrialized countries about 10 years ago. Although TH is effective in reducing mortality and the incidence of severe developmental disabilities, the recent literature converges in reporting frequent cognitive and behavioural difficulties at school entry in children with NE-TH. Although these challenges are deemed minor compared with cerebral palsy and intellectual disability, their impacts on a child's self-determination and family's well-being are quite significant. Therefore, the nature and extent of these difficulties need to be comprehensively described so that appropriate care can be offered.

METHODS AND ANALYSIS

The current study will be the largest follow-up study of neonates with NE treated with TH to characterize their developmental outcomes and associated brain structural profiles at 9 years of age. Specifically, we will compare executive function, attention, social cognition, behaviour, anxiety, self-esteem, peer problems, brain volume, cortical features, white matter microstructure and myelination between children with NE-TH and matched peers without NE. Associations of perinatal risk factors and structural brain integrity with cognitive, behavioural and psycho-emotional deficits will be evaluated to inform about the potential aggravating and protective factors associated with function.

ETHICS AND DISSEMINATION

This study is supported by the Canadian Institute of Health Research (202203PJT-480065-CHI-CFAC-168509), and received approval from the Pediatric Ethical Review Board of the McGill University Health Center (MP-37-2023-9320). The study findings will be disseminated in scientific journals and conferences and presented to parental associations and healthcare providers to inform best practices.

TRIAL REGISTRATION NUMBER

NCT05756296.

Bridging Big Data: Procedures for Combining Non-equivalent Cognitive Measures from the ENIGMA Consortium

Investigators in neuroscience have turned to Big Data to address replication and reliability issues by increasing sample sizes, statistical power, and representativeness of data. These efforts unveil new questions about integrating data arising from distinct sources and instruments. We focus on the most frequently assessed cognitive domain - memory testing - and demonstrate a process for reliable data harmonization across three common measures. We aggregated global raw data from 53 studies totaling N = 10,505 individuals. A mega-analysis was conducted using empirical bayes harmonization to remove site effects, followed by linear models adjusting for common covariates. A continuous item response theory (IRT) model estimated each individual's latent verbal learning ability while accounting for item difficulties. Harmonization significantly reduced inter-site variance while preserving covariate effects, and our conversion tool is freely available online. This demonstrates that large-scale data sharing and harmonization initiatives can address reproducibility and integration challenges across the behavioral sciences.

Large-scale analysis of structural brain asymmetries in schizophrenia via the ENIGMA consortium

Left-right asymmetry is an important organizing feature of the healthy brain that may be altered in schizophrenia, but most studies have used relatively small samples and heterogeneous approaches, resulting in equivocal findings. We carried out the largest case-control study of structural brain asymmetries in schizophrenia, with MRI data from 5,080 affected individuals and 6,015 controls across 46 datasets, using a single image analysis protocol. Asymmetry indexes were calculated for global and regional cortical thickness, surface area, and subcortical volume measures. Differences of asymmetry were calculated between affected individuals and controls per dataset, and effect sizes were meta-analyzed across datasets. Small average case-control differences were observed for thickness asymmetries of the rostral anterior cingulate and the middle temporal gyrus, both driven by thinner left-hemispheric cortices in schizophrenia. Analyses of these asymmetries with respect to the use of antipsychotic medication and other clinical variables did not show any significant associations. Assessment of age- and sex-specific effects revealed a stronger average leftward asymmetry of pallidum volume between older cases and controls. Case-control differences in a multivariate context were assessed in a subset of the data (N = 2,029), which revealed that 7% of the variance across all structural asymmetries was explained by case-control status. Subtle case-control differences of brain macrostructural asymmetry may reflect differences at the molecular, cytoarchitectonic, or circuit levels that have functional relevance for the disorder. Reduced left middle temporal cortical thickness is consistent with altered left-hemisphere language network organization in schizophrenia.

Applications of generative adversarial networks in neuroimaging and clinical neuroscience

Generative adversarial networks (GANs) are one powerful type of deep learning models that have been successfully utilized in numerous fields. They belong to the broader family of generative methods, which learn to generate realistic data with a probabilistic model by learning distributions from real samples. In the clinical context, GANs have shown enhanced capabilities in capturing spatially complex, nonlinear, and potentially subtle disease effects compared to traditional generative methods. This review critically appraises the existing literature on the applications of GANs in imaging studies of various neurological conditions, including Alzheimer's disease, brain tumors, brain aging, and multiple sclerosis. We provide an intuitive explanation of various GAN methods for each application and further discuss the main challenges, open questions, and promising future directions of leveraging GANs in neuroimaging. We aim to bridge the gap between advanced deep learning methods and neurology research by highlighting how GANs can be leveraged to support clinical decision making and contribute to a better understanding of the structural and functional patterns of brain diseases.

White matter microstructural metrics are sensitively associated with clinical staging in Alzheimer's disease

Introduction

White matter microstructure may be abnormal along the Alzheimer's disease (AD) continuum.

Methods

Diffusion magnetic resonance imaging (dMRI) data from the Alzheimer's Disease Neuroimaging Initiative (ADNI, n = 627), Baltimore Longitudinal Study of Aging (BLSA, n = 684), and Vanderbilt Memory & Aging Project (VMAP, n = 296) cohorts were free-water (FW) corrected and conventional, and FW-corrected microstructural metrics were quantified within 48 white matter tracts. Microstructural values were subsequently harmonized using the Longitudinal ComBat technique and inputted as independent variables to predict diagnosis (cognitively unimpaired [CU], mild cognitive impairment [MCI], AD). Models were adjusted for age, sex, race/ethnicity, education, apolipoprotein E (APOE) ε4 carrier status, and APOE ε2 carrier status.

Results

Conventional dMRI metrics were associated globally with diagnostic status; following FW correction, the FW metric itself exhibited global associations with diagnostic status, but intracellular metric associations were diminished.

Discussion

White matter microstructure is altered along the AD continuum. FW correction may provide further understanding of the white matter neurodegenerative process in AD.

Highlights

Longitudinal ComBat successfully harmonized large-scale diffusion magnetic resonance imaging (dMRI) metrics.Conventional dMRI metrics were globally sensitive to diagnostic status.Free-water (FW) correction mitigated intracellular associations with diagnostic status.The FW metric itself was globally sensitive to diagnostic status. Multivariate conventional and FW-corrected models may provide complementary information.

Parsing neurobiological heterogeneity of the clinical high-risk state for psychosis: A pseudo-continuous arterial spin labelling study

Introduction

The impact of the clinical high-risk for psychosis (CHR-P) construct is dependent on accurately predicting outcomes. Individuals with brief limited intermittent psychotic symptoms (BLIPS) have higher risk of developing a first episode of psychosis (FEP) compared to individuals with attenuated psychotic symptoms (APS). Supplementing subgroup stratification with information from candidate biomarkers based on neurobiological parameters, such as resting-state, regional cerebral blood flow (rCBF), may help refine risk estimates. Based on previous evidence, we hypothesized that individuals with BLIPS would exhibit increased rCBF compared to APS in key regions linked to dopaminergic pathways.

Methods

Data from four studies were combined using ComBat (to account for between-study differences) to analyse rCBF in 150 age- and sex-matched subjects (n = 30 healthy controls [HCs], n = 80 APS, n = 20 BLIPS and n = 20 FEP). Global gray matter (GM) rCBF was examined in addition to region-of-interest (ROI) analyses in bilateral/left/right frontal cortex, hippocampus and striatum. Group differences were assessed using general linear models: (i) alone; (ii) with global GM rCBF as a covariate; (iii) with global GM rCBF and smoking status as covariates. Significance was set at p < 0.05.

Results

Whole-brain voxel-wise analyses and Bayesian ROI analyses were also conducted. No significant group differences were found in global [F(3,143) = 1,41, p = 0.24], bilateral frontal cortex [F(3,143) = 1.01, p = 0.39], hippocampus [F(3,143) = 0.63, p = 0.60] or striatum [F(3,143) = 0.52, p = 0.57] rCBF. Similar null findings were observed in lateralized ROIs (p > 0.05). All results were robust to addition of covariates (p > 0.05). No significant clusters were identified in whole-brain voxel-wise analyses (p > 0.05FWE). Weak-to-moderate evidence was found for an absence of rCBF differences between APS and BLIPS in Bayesian ROI analyses.

Conclusion

On this evidence, APS and BLIPS are unlikely to be neurobiologically distinct. Due to this and the weak-to-moderate evidence for the null hypothesis, future research should investigate larger samples of APS and BLIPS through collaboration across large-scale international consortia.

Diffusion tensor imaging metrics associated with future disability in multiple sclerosis

The relationship between brain diffusion microstructural changes and disability in multiple sclerosis (MS) remains poorly understood. We aimed to explore the predictive value of microstructural properties in white (WM) and grey matter (GM), and identify areas associated with mid-term disability in MS patients. We studied 185 patients (71% female; 86% RRMS) with the Expanded Disability Status Scale (EDSS), timed 25-foot walk (T25FW), nine-hole peg test (9HPT), and Symbol Digit Modalities Test (SDMT) at two time-points. We used Lasso regression to analyse the predictive value of baseline WM fractional anisotropy and GM mean diffusivity, and to identify areas related to each outcome at 4.1 years follow-up. Motor performance was associated with WM (T25FW: RMSE = 0.524, R² = 0.304; 9HPT dominant hand: RMSE = 0.662, R² = 0.062; 9HPT non-dominant hand: RMSE = 0.649, R² = 0.139), and SDMT with GM diffusion metrics (RMSE = 0.772, R² = 0.186). Cingulum, longitudinal fasciculus, optic radiation, forceps minor and frontal aslant were the WM tracts most closely linked to motor dysfunction, and temporal and frontal cortex were relevant for cognition. Regional specificity related to clinical outcomes provide valuable information that can be used to develop more accurate predictive models that could improve therapeutic strategies.

In vivo white matter microstructure in adolescents with early-onset psychosis: a multi-site mega-analysis

Emerging evidence suggests brain white matter alterations in adolescents with early-onset psychosis (EOP; age of onset <18 years). However, as neuroimaging methods vary and sample sizes are modest, results remain inconclusive. Using harmonized data processing protocols and a mega-analytic approach, we compared white matter microstructure in EOP and healthy controls using diffusion tensor imaging (DTI). Our sample included 321 adolescents with EOP (median age = 16.6 years, interquartile range (IQR) = 2.14, 46.4% females) and 265 adolescent healthy controls (median age = 16.2 years, IQR = 2.43, 57.7% females) pooled from nine sites. All sites extracted mean fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD), and axial diffusivity (AD) for 25 white matter regions of interest per participant. ComBat harmonization was performed for all DTI measures to adjust for scanner differences. Multiple linear regression models were fitted to investigate case-control differences and associations with clinical variables in regional DTI measures. We found widespread lower FA in EOP compared to healthy controls, with the largest effect sizes in the superior longitudinal fasciculus (Cohen's d = 0.37), posterior corona radiata (d = 0.32), and superior fronto-occipital fasciculus (d = 0.31). We also found widespread higher RD and more localized higher MD and AD. We detected significant effects of diagnostic subgroup, sex, and duration of illness, but not medication status. Using the largest EOP DTI sample to date, our findings suggest a profile of widespread white matter microstructure alterations in adolescents with EOP, most prominently in male individuals with early-onset schizophrenia and individuals with a shorter duration of illness.

Global multi-center and multi-modal magnetic resonance imaging study of obsessive-compulsive disorder: Harmonization and monitoring of protocols in healthy volunteers and phantoms

OBJECTIVES

We describe the harmonized MRI acquisition and quality assessment of an ongoing global OCD study, with the aim to translate representative, well-powered neuroimaging findings in neuropsychiatric research to worldwide populations.

METHODS

We report on T1-weighted structural MRI, resting-state functional MRI, and multi-shell diffusion-weighted imaging of 140 healthy participants (28 per site), two traveling controls, and regular phantom scans.

RESULTS

Human image quality measures (IQMs) and outcome measures showed smaller within-site variation than between-site variation. Outcome measures were less variable than IQMs, especially for the traveling controls. Phantom IQMs were stable regarding geometry, SNR, and mean diffusivity, while fMRI fluctuation was more variable between sites.

CONCLUSIONS

Variation in IQMs persists, even for an a priori harmonized data acquisition protocol, but after pre-processing they have less of an impact on the outcome measures. Continuous monitoring IQMs per site is valuable to detect potential artifacts and outliers. The inclusion of both cases and healthy participants at each site remains mandatory.

The Role of Neuroimaging in Evolving TBI Research and Clinical Practice

Neuroimaging technologies such as computed tomography (CT) and magnetic resonance imaging (MRI) have been widely adopted in the clinical diagnosis and management of traumatic brain injury (TBI), particularly at the more acute and severe levels of injury. Additionally, a number of advanced applications of MRI have been employed in TBI-related clinical research with great promise, and researchers have used these techniques to better understand underlying mechanisms, progression of secondary injury and tissue perturbation over time, and relation of focal and diffuse injury to later outcome. However, the acquisition and analysis time, the cost of these and other imaging modalities, and the need for specialized expertise have represented historical barriers in extending these tools in clinical practice. While group studies are important in detecting patterns, heterogeneity among patient presentation and limited sample sizes from which to compare individual level data to well-developed normative data have also played a role in the limited translatability of imaging to wider clinical application. Fortunately, the field of TBI has benefitted from increased public and scientific awareness of the prevalence and impact of TBI, particularly in head injury related to recent military conflicts and sport-related concussion. This awareness parallels an increase in federal funding in the United States and other countries allocated to investigation in these areas. In this article we summarize funding and publication trends since the mainstream adoption of imaging in TBI to elucidate evolving trends and priorities in the application of different techniques and patient populations. We also review recent and ongoing efforts to advance the field through promoting reproducibility, data sharing, big data analytic methods, and team science. Finally, we discuss international collaborative efforts to combine and harmonize neuroimaging, cognitive, and clinical data, both prospectively and retrospectively. Each of these represent unique, but related, efforts that facilitate closing gaps between the use of advanced imaging solely as a research tool and the use of it in clinical diagnosis, prognosis, and treatment planning and monitoring.

History of suicide attempt associated with amygdala and hippocampus changes among individuals with schizophrenia

Abnormalities in subcortical brain structures may reflect higher suicide risk in mood disorders, but less is known about its associations for schizophrenia. This cross-sectional imaging study aimed to explore whether the history of suicide attempts was associated with subcortical changes among individuals with schizophrenia. We recruited 44 individuals with schizophrenia and a history of suicide attempts (SZ-SA) and 44 individuals with schizophrenia but without a history of suicide attempts (SZ-NSA) and 44 healthy controls. Linear regression showed that SZ-SA had smaller volumes of the hippocampus (Cohen's d = -0.72), the amygdala (Cohen's d = -0.69), and some nuclei of the amygdala (Cohen's d, -0.57 to -0.72) than SZ-NSA after adjusting for age, sex, illness phase, and intracranial volume. There was no difference in the volume of the subfields of the hippocampus. It suggests the history of suicide attempts is associated with subcortical volume alterations in schizophrenia.

Harmonization of multi-scanner in vivo magnetic resonance spectroscopy: ENIGMA consortium task group considerations

Magnetic resonance spectroscopy is a powerful, non-invasive, quantitative imaging technique that allows for the measurement of brain metabolites that has demonstrated utility in diagnosing and characterizing a broad range of neurological diseases. Its impact, however, has been limited due to small sample sizes and methodological variability in addition to intrinsic limitations of the method itself such as its sensitivity to motion. The lack of standardization from a data acquisition and data processing perspective makes it difficult to pool multiple studies and/or conduct multisite studies that are necessary for supporting clinically relevant findings. Based on the experience of the ENIGMA MRS work group and a review of the literature, this manuscript provides an overview of the current state of MRS data harmonization. Key factors that need to be taken into consideration when conducting both retrospective and prospective studies are described. These include (1) MRS acquisition issues such as pulse sequence, RF and B0 calibrations, echo time, and SNR; (2) data processing issues such as pre-processing steps, modeling, and quantitation; and (3) biological factors such as voxel location, age, sex, and pathology. Various approaches to MRS data harmonization are then described including meta-analysis, mega-analysis, linear modeling, ComBat and artificial intelligence approaches. The goal is to provide both novice and experienced readers with the necessary knowledge for conducting MRS data harmonization studies.