Impact of scanner hardware and imaging protocol on image quality and compartment volume precision in the ADNI cohort

nnSegNeXt: A 3D Convolutional Network for Brain Tissue Segmentation Based on Quality Evaluation

Accurate and automated segmentation of brain tissue images can significantly streamline clinical diagnosis and analysis. Manual delineation needs improvement due to its laborious and repetitive nature, while automated techniques encounter challenges stemming from disparities in magnetic resonance imaging (MRI) acquisition equipment and accurate labeling. Existing software packages, such as FSL and FreeSurfer, do not fully replace ground truth segmentation, highlighting the need for an efficient segmentation tool. To better capture the essence of cerebral tissue, we introduce nnSegNeXt, an innovative segmentation architecture built upon the foundations of quality assessment. This pioneering framework effectively addresses the challenges posed by missing and inaccurate annotations. To enhance the model's discriminative capacity, we integrate a 3D convolutional attention mechanism instead of conventional convolutional blocks, enabling simultaneous encoding of contextual information through the incorporation of multiscale convolutional features. Our methodology was evaluated on four multi-site T1-weighted MRI datasets from diverse sources, magnetic field strengths, scanning parameters, temporal instances, and neuropsychiatric conditions. Empirical evaluations on the HCP, SALD, and IXI datasets reveal that nnSegNeXt surpasses the esteemed nnUNet, achieving Dice coefficients of 0.992, 0.987, and 0.989, respectively, and demonstrating superior generalizability across four distinct projects with Dice coefficients ranging from 0.967 to 0.983. Additionally, extensive ablation studies have been implemented to corroborate the effectiveness of the proposed model. These findings represent a notable advancement in brain tissue analysis, suggesting that nnSegNeXt holds the promise to significantly refine clinical workflows.

Voxel-based morphometry in single subjects without a scanner-specific normal database using a convolutional neural network

OBJECTIVES

Reliable detection of disease-specific atrophy in individual T1w-MRI by voxel-based morphometry (VBM) requires scanner-specific normal databases (NDB), which often are not available. The aim of this retrospective study was to design, train, and test a deep convolutional neural network (CNN) for single-subject VBM without the need for a NDB (CNN-VBM).

MATERIALS AND METHODS

The training dataset comprised 8945 T1w scans from 65 different scanners. The gold standard VBM maps were obtained by conventional VBM with a scanner-specific NDB for each of the 65 scanners. CNN-VBM was tested in an independent dataset comprising healthy controls (n = 37) and subjects with Alzheimer's disease (AD, n = 51) or frontotemporal lobar degeneration (FTLD, n = 30). A scanner-specific NDB for the generation of the gold standard VBM maps was available also for the test set. The technical performance of CNN-VBM was characterized by the Dice coefficient of CNN-VBM maps relative to VBM maps from scanner-specific VBM. For clinical testing, VBM maps were categorized visually according to the clinical diagnoses in the test set by two independent readers, separately for both VBM methods.

RESULTS

The VBM maps from CNN-VBM were similar to the scanner-specific VBM maps (median Dice coefficient 0.85, interquartile range [0.81, 0.90]). Overall accuracy of the visual categorization of the VBM maps for the detection of AD or FTLD was 89.8% for CNN-VBM and 89.0% for scanner-specific VBM.

CONCLUSION

CNN-VBM without NDB provides a similar performance in the detection of AD- and FTLD-specific atrophy as conventional VBM.

CLINICAL RELEVANCE STATEMENT

A deep convolutional neural network for voxel-based morphometry eliminates the need of scanner-specific normal databases without relevant performance loss and, therefore, could pave the way for the widespread clinical use of voxel-based morphometry to support the diagnosis of neurodegenerative diseases.

KEY POINTS

• The need of normal databases is a barrier for widespread use of voxel-based brain morphometry. • A convolutional neural network achieved a similar performance for detection of atrophy than conventional voxel-based morphometry. • Convolutional neural networks can pave the way for widespread clinical use of voxel-based morphometry.

Combining neuropsychological assessment and structural neuroimaging to identify early Alzheimer's disease in a memory clinic cohort

INTRODUCTION

The current study examined the contributions of comprehensive neuropsychological assessment and volumetric assessment of selected mesial temporal subregions on structural magnetic resonance imaging (MRI) to identify patients with amnestic mild cognitive impairment (aMCI) and mild probable Alzheimer's disease (AD) dementia in a memory clinic cohort.

METHODS

Comprehensive neuropsychological assessment and automated entorhinal, transentorhinal, and hippocampal volume measurements were conducted in 40 healthy controls, 38 patients with subjective memory symptoms, 16 patients with aMCI, 16 patients with mild probable AD dementia. Multinomial logistic regression was used to compare the neuropsychological and MRI measures.

RESULTS

Combining the neuropsychological and MRI measures improved group membership prediction over the MRI measures alone but did not improve group membership prediction over the neuropsychological measures alone.

CONCLUSION

Comprehensive neuropsychological assessment was an important tool to evaluate cognitive impairment. The mesial temporal volumetric MRI measures contributed no diagnostic value over and above the determinations made through neuropsychological assessment.

Removing outliers from the normative database improves regional atrophy detection in single-subject voxel-based morphometry

PURPOSE

Single-subject voxel-based morphometry (VBM) compares an individual T1-weighted MRI to a sample of normal MRI in a normative database (NDB) to detect regional atrophy. Outliers in the NDB might result in reduced sensitivity of VBM. The primary aim of the current study was to propose a method for outlier removal ("NDB cleaning") and to test its impact on the performance of VBM for detection of Alzheimer's disease (AD) and frontotemporal lobar degeneration (FTLD).

METHODS

T1-weighted MRI of 81 patients with biomarker-confirmed AD (n = 51) or FTLD (n = 30) and 37 healthy subjects with simultaneous FDG-PET/MRI were included as test dataset. Two different NDBs were used: a scanner-specific NDB (37 healthy controls from the test dataset) and a non-scanner-specific NDB comprising 164 normal T1-weighted MRI from 164 different MRI scanners. Three different quality metrics based on leave-one-out testing of the scans in the NDB were implemented. A scan was removed if it was an outlier with respect to one or more quality metrics. VBM maps generated with and without NDB cleaning were assessed visually for the presence of AD or FTLD.

RESULTS

Specificity of visual interpretation of the VBM maps for detection of AD or FTLD was 100% in all settings. Sensitivity was increased by NDB cleaning with both NDBs. The effect was statistically significant for the multiple-scanner NDB (from 0.47 [95%-CI 0.36-0.58] to 0.61 [0.49-0.71]).

CONCLUSION

NDB cleaning has the potential to improve the sensitivity of VBM for the detection of AD or FTLD without increasing the risk of false positive findings.

Accuracy of MRI-Based Radiomics in Diagnosis of Placenta Accreta Spectrum: A PRISMA Systematic Review and Meta-Analysis

BACKGROUND Placenta accreta syndrome (PAS) can lead to severe obstetric bleeding, and can be life-threatening. This study aimed to assess the precision of radiomics features derived from magnetic resonance imaging (MRI) for diagnosing PAS. MATERIAL AND METHODS A comprehensive search was conducted in the databases PubMed, Embase, Web of Science, and the Cochrane library from inception to October 2023. We included diagnostic accuracy studies utilizing radiomics-MRI in PAS patients, with histopathology serving as the reference standard. The overall diagnostic odds ratio (DOR), sensitivity, specificity, and area under the curve (AUC) were computed to gauge the diagnostic accuracy of MRI-based radiomic features in PAS patients. Quality assessment was performed using the Quality Assessment of Diagnostic Accuracy Studies 2. Statistical analyses were carried out using Stata 14.2, MetaDiSc 1.4, and Review Manager 5.3 software. RESULTS Seven studies involving 672 patients were incorporated. The aggregated DOR, sensitivity, specificity, and AUC for radiomics in detecting PAS were 78% (confidence interval32, 191), 87% (76%, 93%), 92% (89%, 94%), and 0.93 (0.91-0.95), respectively. The meta-analysis revealed notable heterogeneity among the included studies, with no evidence of a threshold effect. Subgroup analysis demonstrated that, in comparison to manual segmentation and validation groups with ≤100 cases and internal validation datasets, automated segmentation, validation groups with >100 cases, and external validation datasets exhibited superior diagnostic performance . CONCLUSIONS Our findings indicate that MRI-based radiomic features perform well in assessing the diagnostic risk of PAS during prenatal diagnosis. This noninvasive and convenient tool may prove valuable in facilitating the identification of PAS.

Growing importance of brain morphometry analysis in the clinical routine: The hidden impact of MR sequence parameters

Volumetric assessment based on structural MRI is increasingly recognized as an auxiliary tool to visual reading, also in examinations acquired in the clinical routine. However, MRI acquisition parameters can significantly influence these measures, which must be considered when interpreting the results on an individual patient level. This Technical Note shall demonstrate the problem. Using data from a dedicated experiment, we show the influence of two crucial sequence parameters on the GM/WM contrast and their impact on the measured volumes. A simulated contrast derived from acquisition parameters TI/TR may serve as surrogate and is highly correlated (r=0.96) with the measured contrast.

A systematic review of (semi-)automatic quality control of T1-weighted MRI scans

PURPOSE

Artifacts in magnetic resonance imaging (MRI) scans degrade image quality and thus negatively affect the outcome measures of clinical and research scanning. Considering the time-consuming and subjective nature of visual quality control (QC), multiple (semi-)automatic QC algorithms have been developed. This systematic review presents an overview of the available (semi-)automatic QC algorithms and software packages designed for raw, structural T1-weighted (T1w) MRI datasets. The objective of this review was to identify the differences among these algorithms in terms of their features of interest, performance, and benchmarks.

METHODS

We queried PubMed, EMBASE (Ovid), and Web of Science databases on the fifth of January 2023, and cross-checked reference lists of retrieved papers. Bias assessment was performed using PROBAST (Prediction model Risk Of Bias ASsessment Tool).

RESULTS

A total of 18 distinct algorithms were identified, demonstrating significant variations in methods, features, datasets, and benchmarks. The algorithms were categorized into rule-based, classical machine learning-based, and deep learning-based approaches. Numerous unique features were defined, which can be roughly divided into features capturing entropy, contrast, and normative measures.

CONCLUSION

Due to dataset-specific optimization, it is challenging to draw broad conclusions about comparative performance. Additionally, large variations exist in the used datasets and benchmarks, further hindering direct algorithm comparison. The findings emphasize the need for standardization and comparative studies for advancing QC in MR imaging. Efforts should focus on identifying a dataset-independent measure as well as algorithm-independent methods for assessing the relative performance of different approaches.

Multi-Ethnic Norms for Volumes of Subcortical and Lobar Brain Structures Measured by Neuro I: Ethnicity May Improve the Diagnosis of Alzheimer's Disease1

Background

We previously demonstrated the validity of a regression model that included ethnicity as a novel predictor for predicting normative brain volumes in old age. The model was optimized using brain volumes measured with a standard tool FreeSurfer.

Objective

Here we further verified the prediction model using newly estimated brain volumes from Neuro I, a quantitative brain analysis system developed for Korean populations.

Methods

Lobar and subcortical volumes were estimated from MRI images of 1,629 normal Korean and 786 Caucasian subjects (age range 59-89) and were predicted in linear regression from ethnicity, age, sex, intracranial volume, magnetic field strength, and scanner manufacturers.

Results

In the regression model predicting the new volumes, ethnicity was again a substantial predictor in most regions. Additionally, the model-based z-scores of regions were calculated for 428 AD patients and the matched controls, and then employed for diagnostic classification. When the AD classifier adopted the z-scores adjusted for ethnicity, the diagnostic accuracy has noticeably improved (AUC = 0.85, ΔAUC = + 0.04, D = 4.10, p < 0.001).

Conclusions

Our results suggest that the prediction model remains robust across different measurement tool, and ethnicity significantly contributes to the establishment of norms for brain volumes and the development of a diagnostic system for neurodegenerative diseases.

A three-dimensional deep learning model for inter-site harmonization of structural MR images of the brain: Extensive validation with a multicenter dataset

In multicenter MRI studies, pooling the imaging data can introduce site-related variabilities and can therefore bias the subsequent analyses. To harmonize the intensity distributions of brain MR images in a multicenter dataset, unsupervised deep learning methods can be employed. Here, we developed a model based on cycle-consistent adversarial networks for the harmonization of T1-weighted brain MR images. In contrast to previous works, it was designed to process three-dimensional whole-brain images in a stable manner while optimizing computation resources. Using six different MRI datasets for healthy adults (n=1525 in total) with different acquisition parameters, we tested the model in (i) three pairwise harmonizations with site effects of various sizes, (ii) an overall harmonization of the six datasets with different age distributions, and (iii) a traveling-subject dataset. Our results for intensity distributions, brain volumes, image quality metrics and radiomic features indicated that the MRI characteristics at the various sites had been effectively homogenized. Next, brain age prediction experiments and the observed correlation between the gray-matter volume and age showed that thanks to an appropriate training strategy and despite biological differences between the dataset populations, the model reinforced biological patterns. Furthermore, radiologic analyses of the harmonized images attested to the conservation of the radiologic information in the original images. The robustness of the harmonization model (as judged with various datasets and metrics) demonstrates its potential for application in retrospective multicenter studies.

Deep learning-based high-accuracy detection for lumbar and cervical degenerative disease on T2-weighted MR images

PURPOSE

To develop and validate a deep learning (DL) model for detecting lumbar degenerative disease in both sagittal and axial views of T2-weighted MRI and evaluate its generalized performance in detecting cervical degenerative disease.

METHODS

T2-weighted MRI scans of 804 patients with symptoms of lumbar degenerative disease were retrospectively collected from three hospitals. The training dataset (n = 456) and internal validation dataset (n = 134) were randomly selected from the center I. Two external validation datasets comprising 100 and 114 patients were from center II and center III, respectively. A DL model based on 3D ResNet18 and transformer architecture was proposed to detect lumbar degenerative disease. In addition, a cervical MR image dataset comprising 200 patients from an independent hospital was used to evaluate the generalized performance of the DL model. The diagnostic performance was assessed by the free-response receiver operating characteristic (fROC) curve and precision-recall (PR) curve. Precision, recall, and F1-score were used to measure the DL model.

RESULTS

A total of 2497 three-dimension retrogression annotations were labeled for training (n = 1157) and multicenter validation (n = 1340). The DL model showed excellent detection efficiency in the internal validation dataset, with F1-score achieving 0.971 and 0.903 on the sagittal and axial MR images, respectively. Good performance was also observed in the external validation dataset I (F1-score, 0.768 on sagittal MR images and 0.837 on axial MR images) and external validation dataset II (F1-score, 0.787 on sagittal MR images and 0.770 on axial MR images). Furthermore, the robustness of the DL model was demonstrated via transfer learning and generalized performance evaluation on the external cervical dataset, with the F1-score yielding 0.931 and 0.919 on the sagittal and axial MR images, respectively.

CONCLUSION

The proposed DL model can automatically detect lumbar and cervical degenerative disease on T2-weighted MR images with good performance, robustness, and feasibility in clinical practice.

Intersite brain MRI volumetric biases persist even in a harmonized multisubject study of multiple sclerosis

BACKGROUND AND PURPOSE

Multicenter study designs involving a variety of MRI scanners have become increasingly common. However, these present the issue of biases in image-based measures due to scanner or site differences. To assess these biases, we imaged 11 volunteers with multiple sclerosis (MS) with scan and rescan data at four sites.

METHODS

Images were acquired on Siemens or Philips scanners at 3 Tesla. Automated white matter lesion detection and whole-brain, gray and white matter, and thalamic volumetry were performed, as well as expert manual delineations of T1 magnetization-prepared rapid acquisition gradient echo and T2 fluid-attenuated inversion recovery lesions. Random-effect and permutation-based nonparametric modeling was performed to assess differences in estimated volumes within and across sites.

RESULTS

Random-effect modeling demonstrated model assumption violations for most comparisons of interest. Nonparametric modeling indicated that site explained >50% of the variation for most estimated volumes. This expanded to >75% when data from both Siemens and Philips scanners were included. Permutation tests revealed significant differences between average inter- and intrasite differences in most estimated brain volumes (P < .05). The automatic activation of spine coil elements during some acquisitions resulted in a shading artifact in these images. Permutation tests revealed significant differences between thalamic volume measurements from acquisitions with and without this artifact.

CONCLUSION

Differences in brain volumetry persisted across MR scanners despite protocol harmonization. These differences were not well explained by variance component modeling; however, statistical innovations for mitigating intersite differences show promise in reducing biases in multicenter studies of MS.

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.

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.

Positron emission tomography and magnetic resonance imaging methods and datasets within the Dominantly Inherited Alzheimer Network (DIAN)

The Dominantly Inherited Alzheimer Network (DIAN) is an international collaboration studying autosomal dominant Alzheimer disease (ADAD). ADAD arises from mutations occurring in three genes. Offspring from ADAD families have a 50% chance of inheriting their familial mutation, so non-carrier siblings can be recruited for comparisons in case-control studies. The age of onset in ADAD is highly predictable within families, allowing researchers to estimate an individual's point in the disease trajectory. These characteristics allow candidate AD biomarker measurements to be reliably mapped during the preclinical phase. Although ADAD represents a small proportion of AD cases, understanding neuroimaging-based changes that occur during the preclinical period may provide insight into early disease stages of 'sporadic' AD also. Additionally, this study provides rich data for research in healthy aging through inclusion of the non-carrier controls. Here we introduce the neuroimaging dataset collected and describe how this resource can be used by a range of researchers.

Adversarially-Regularized Mixed Effects Deep Learning (ARMED) Models Improve Interpretability, Performance, and Generalization on Clustered (non-iid) Data

Natural science datasets frequently violate assumptions of independence. Samples may be clustered (e.g., by study site, subject, or experimental batch), leading to spurious associations, poor model fitting, and confounded analyses. While largely unaddressed in deep learning, this problem has been handled in the statistics community through mixed effects models, which separate cluster-invariant fixed effects from cluster-specific random effects. We propose a general-purpose framework for Adversarially-Regularized Mixed Effects Deep learning (ARMED) models through non-intrusive additions to existing neural networks: 1) an adversarial classifier constraining the original model to learn only cluster-invariant features, 2) a random effects subnetwork capturing cluster-specific features, and 3) an approach to apply random effects to clusters unseen during training. We apply ARMED to dense, convolutional, and autoencoder neural networks on 4 datasets including simulated nonlinear data, dementia prognosis and diagnosis, and live-cell image analysis. Compared to prior techniques, ARMED models better distinguish confounded from true associations in simulations and learn more biologically plausible features in clinical applications. They can also quantify inter-cluster variance and visualize cluster effects in data. Finally, ARMED matches or improves performance on data from clusters seen during training (5-28% relative improvement) and generalization to unseen clusters (2-9% relative improvement) versus conventional models.

A comprehensive evaluation of multicentric reliability of single-subject cortical morphological networks on traveling subjects

Despite the prevalence of research on single-subject cerebral morphological networks in recent years, whether they can offer a reliable way for multicentric studies remains largely unknown. Using two multicentric datasets of traveling subjects, this work systematically examined the inter-site test-retest (TRT) reliabilities of single-subject cerebral morphological networks, and further evaluated the effects of several key factors. We found that most graph-based network measures exhibited fair to excellent reliabilities regardless of different analytical pipelines. Nevertheless, the reliabilities were affected by choices of morphological index (fractal dimension > sulcal depth > gyrification index > cortical thickness), brain parcellation (high-resolution > low-resolution), thresholding method (proportional > absolute), and network type (binarized > weighted). For the factor of similarity measure, its effects depended on the thresholding method used (absolute: Kullback-Leibler divergence > Jensen-Shannon divergence; proportional: Jensen-Shannon divergence > Kullback-Leibler divergence). Furthermore, longer data acquisition intervals and different scanner software versions significantly reduced the reliabilities. Finally, we showed that inter-site reliabilities were significantly lower than intra-site reliabilities for single-subject cerebral morphological networks. Altogether, our findings propose single-subject cerebral morphological networks as a promising approach for multicentric human connectome studies, and offer recommendations on how to determine analytical pipelines and scanning protocols for obtaining reliable results.

BrainSuite BIDS App: Containerized Workflows for MRI Analysis

There has been a concerted effort by the neuroimaging community to establish standards for computational methods for data analysis that promote reproducibility and portability. In particular, the Brain Imaging Data Structure (BIDS) specifies a standard for storing imaging data, and the related BIDS App methodology provides a standard for implementing containerized processing environments that include all necessary dependencies to process BIDS datasets using image processing workflows. We present the BrainSuite BIDS App, which encapsulates the core MRI processing functionality of BrainSuite within the BIDS App framework. Specifically, the BrainSuite BIDS App implements a participant-level workflow comprising three pipelines and a corresponding set of group-level analysis workflows for processing the participant-level outputs. The BrainSuite Anatomical Pipeline (BAP) extracts cortical surface models from a T1-weighted (T1w) MRI. It then performs surface-constrained volumetric registration to align the T1w MRI to a labeled anatomical atlas, which is used to delineate anatomical regions of interest in the MRI brain volume and on the cortical surface models. The BrainSuite Diffusion Pipeline (BDP) processes diffusion-weighted imaging (DWI) data, with steps that include coregistering the DWI data to the T1w scan, correcting for geometric image distortion, and fitting diffusion models to the DWI data. The BrainSuite Functional Pipeline (BFP) performs fMRI processing using a combination of FSL, AFNI, and BrainSuite tools. BFP coregisters the fMRI data to the T1w image, then transforms the data to the anatomical atlas space and to the Human Connectome Project's grayordinate space. Each of these outputs can then be processed during group-level analysis. The outputs of BAP and BDP are analyzed using the BrainSuite Statistics in R (bssr) toolbox, which provides functionality for hypothesis testing and statistical modeling. The outputs of BFP can be analyzed using atlas-based or atlas-free statistical methods during group-level processing. These analyses include the application of BrainSync, which synchronizes the time-series data temporally and enables comparison of resting-state or task-based fMRI data across scans. We also present the BrainSuite Dashboard quality control system, which provides a browser-based interface for reviewing the outputs of individual modules of the participant-level pipelines across a study in real-time as they are generated. BrainSuite Dashboard facilitates rapid review of intermediate results, enabling users to identify processing errors and make adjustments to processing parameters if necessary. The comprehensive functionality included in the BrainSuite BIDS App provides a mechanism for rapidly deploying the BrainSuite workflows into new environments to perform large-scale studies. We demonstrate the capabilities of the BrainSuite BIDS App using structural, diffusion, and functional MRI data from the Amsterdam Open MRI Collection's Population Imaging of Psychology dataset.

Validation of cross-sectional and longitudinal ComBat harmonization methods for magnetic resonance imaging data on a travelling subject cohort

Background

The growth in multi-center neuroimaging studies generated a need for methods that mitigate the differences in hardware and acquisition protocols across sites i.e., scanner effects. ComBat harmonization methods have shown promise but have not yet been tested on all the data types commonly studied with magnetic resonance imaging (MRI). This study aimed to validate neuroCombat, longCombat and gamCombat on both structural and diffusion metrics in both cross-sectional and longitudinal data.

Methods

We used a travelling subject design whereby 73 healthy volunteers contributed 161 scans across two sites and four machines using one T1 and five diffusion MRI protocols. Scanner was defined as a composite of site, machine and protocol. A common pipeline extracted two structural metrics (volumes and cortical thickness) and two diffusion tensor imaging metrics (mean diffusivity and fractional anisotropy) for seven regions of interest including gray and (except for cortical thickness) white matter regions.

Results

Structural data exhibited no significant scanner effect and therefore did not benefit from harmonization in our particular cohort. Indeed, attempting harmonization obscured the true biological effect for some regions of interest. Diffusion data contained marked scanner effects and was successfully harmonized by all methods, resulting in smaller scanner effects and better detection of true biological effects. LongCombat less effectively reduced the scanner effect for cross-sectional white matter data but had a slightly lower probability of incorrectly finding group differences in simulations, compared to neuroCombat and gamCombat. False positive rates for all methods and all metrics did not significantly exceed 5%.

Conclusions

Statistical harmonization of structural data is not always necessary and harmonization in the absence of a scanner effect may be harmful. Harmonization of diffusion MRI data is highly recommended with neuroCombat, longCombat and gamCombat performing well in cross-sectional and longitudinal settings.

Improving across-dataset brain tissue segmentation for MRI imaging using transformer

Brain tissue segmentation has demonstrated great utility in quantifying MRI data by serving as a precursor to further post-processing analysis. However, manual segmentation is highly labor-intensive, and automated approaches, including convolutional neural networks (CNNs), have struggled to generalize well due to properties inherent to MRI acquisition, leaving a great need for an effective segmentation tool. This study introduces a novel CNN-Transformer hybrid architecture designed to improve brain tissue segmentation by taking advantage of the increased performance and generality conferred by Transformers for 3D medical image segmentation tasks. We first demonstrate the superior performance of our model on various T1w MRI datasets. Then, we rigorously validate our model's generality applied across four multi-site T1w MRI datasets, covering different vendors, field strengths, scan parameters, and neuropsychiatric conditions. Finally, we highlight the reliability of our model on test-retest scans taken in different time points. In all situations, our model achieved the greatest generality and reliability compared to the benchmarks. As such, our method is inherently robust and can serve as a valuable tool for brain related T1w MRI studies. The code for the TABS network is available at: https://github.com/raovish6/TABS.

Assessing Multi-Site rs-fMRI-Based Connectomic Harmonization Using Information Theory

Several harmonization techniques have recently been proposed for connectomics/networks derived from resting-state functional magnetic resonance imaging (rs-fMRI) acquired at multiple sites. These techniques have the objective of mitigating site-specific biases that complicate its subsequent analysis and, therefore, compromise the quality of the results when these images are analyzed together. Thus, harmonization is indispensable when large cohorts are required in which the data obtained must be independent of the particular condition of each resonator, its make and model, its calibration, and other features or artifacts that may affect the significance of the acquisition. To date, no assessment of the actual efficacy of these harmonization techniques has been proposed. In this work, we apply recently introduced Information Theory tools to analyze the effectiveness of these techniques, developing a methodology that allows us to compare different harmonization models. We demonstrate the usefulness of this methodology by applying it to some of the most widespread harmonization frameworks and datasets. As a result, we are able to show that some of these techniques are indeed ineffective since the acquisition site can still be determined from the fMRI data after the processing.

Multisite MRI reproducibility of lateral ventricular volume using the NAIMS cooperative pilot dataset

BACKGROUND AND PURPOSE

The North American Imaging in Multiple Sclerosis (NAIMS) multisite project identified interscanner reproducibility issues with T1-based whole brain volume (WBV). Lateral ventricular volume (LVV) acquired on T2-fluid-attenuated inverse recovery (FLAIR) scans has been proposed as a robust proxy measure. Therefore, we sought to determine the relative magnitude of scanner-induced T2-FLAIR-based LVV and T1-based WBV measurement errors in relation to clinically meaningful changes.

METHODS

This was a post hoc analysis of the NAIMS pilot dataset in which a relapsing-remitting MS patient with no intrastudy clinical or radiological activity was imaged twice on seven different Siemens scanners across the United States. LVV was determined using the automated NeuroSTREAM technique on T2-FLAIR and WBV was determined with SIENAX on high-resolution T1-MPRAGE. Average LVV and WBV were measured, and absolute intrascanner and interscanner coefficients of variation (CoVs) were calculated. The variabilities were compared to previously established annual pathological and clinically meaningful cutoffs of 0.40% for WBV and of 3.51% for LVV.

RESULTS

Mean LVV across all seven scan/rescan pairs was 45.87 ± 1.15 ml. Average LVV intrascanner CoV was 1.42% and interscanner CoV was 1.78%, both smaller than the reported annualized clinically meaningful cutoff of 3.51%. In contrast, intra- and interscanner CoVs for WBV (0.99% and 1.15%) were both higher than the established cutoff of 0.40%. Individually, 1/7 intrasite and 2/7 intersite pair-wise LVV comparisons were above the 3.51% cutoff, whereas 4/7 intrasite and 7/7 intersite WBV comparisons were above the 0.40% cutoff.

CONCLUSION

Fully automated LVV segmentation has higher absolute variability than WBV, but much lower relative variability compared to clinically relevant changes, and may therefore be a meaningful proxy outcome measure of neurodegeneration.

Brain atrophy measurement over a MRI scanner change in multiple sclerosis

BACKGROUND

A change in MRI hardware impacts brain volume measurements. The aim of this study was to use MRI data from multiple sclerosis (MS) patients and healthy control subjects (HCs) to statistically model how to adjust brain atrophy measures in MS patients after a major scanner upgrade.

METHODS

We scanned 20 MS patients and 26 HCs before and three months after a major scanner upgrade (1.5 T Siemens Healthineers Magnetom Avanto to 3 T Siemens Healthineers Skyra Fit). The patient group also underwent standardized serial MRIs before and after the scanner change. Percentage whole brain volume changes (PBVC) measured by Structural Image Evaluation using Normalization of Atrophy (SIENA) in the HCs was used to estimate a corrective term based on a linear model. The factor was internally validated in HCs, and then applied to the MS group.

RESULTS

Mean PBVC during the scanner change was higher in MS than HCs (-4.1 ± 0.8 % versus -3.4 ± 0.6 %). A fixed corrective term of 3.4 (95% confidence interval: 3.13-3.67)% was estimated based on the observed average changes in HCs. Age and gender did not have a significant influence on this corrective term. After adjustment, a linear mixed effects model showed that the brain atrophy measures in MS during the scanner upgrade were not anymore associated with the scanner type (old vs new scanner; p = 0.29).

CONCLUSION

A scanner change affects brain atrophy measures in longitudinal cohorts. The inclusion of a corrective term based on changes observed in HCs helps to adjust for the known and unknown factors associated with a scanner upgrade on a group level.

Harmonization of Multicenter Cortical Thickness Data by Linear Mixed Effect Model

Objective

Analyzing neuroimages being useful method in the field of neuroscience and neurology and solving the incompatibilities across protocols and vendors have become a major problem. We referred to this incompatibility as “center effects,” and in this study, we attempted to correct such center effects of cortical feature obtained from multicenter magnetic resonance images (MRIs).

Methods

For MRI of a total of 4,321 multicenter subjects, the harmonized w-score was calculated by correcting biological covariates such as age, sex, years of education, and intercranial volume (ICV) as fixed effects and center information as a random effect. Afterward, we performed classification tasks using principal component analysis (PCA) and linear discriminant analysis (LDA) to check whether the center effect was successfully corrected from the harmonized w-score.

Results

First, an experiment was conducted to predict the dataset origin of a random subject sampled from two different datasets, and it was confirmed that the prediction accuracy of linear mixed effect (LME) model-based w-score was significantly closer to the baseline than that of raw cortical thickness. As a second experiment, we classified the data of the normal and patient groups of each dataset, and LME model-based w-score, which is biological-feature-corrected values, showed higher classification accuracy than the raw cortical thickness data. Afterward, to verify the compatibility of the dataset used for LME model training and the dataset that is not, intraobject comparison and w-score RMSE calculation process were performed.

Conclusion

Through comparison between the LME model-based w-score and existing methods and several classification tasks, we showed that the LME model-based w-score sufficiently corrects the center effects while preserving the disease effects from the dataset. We also showed that the preserved disease effects have a match with well-known disease atrophy patterns such as Alzheimer’s disease or Parkinson’s disease. Finally, through intrasubject comparison, we found that the difference between centers decreases in the LME model-based w-score compared with the raw cortical thickness and thus showed that our model well-harmonizes the data that are not used for the model training.

Ten years of image analysis and machine learning competitions in dementia

Machine learning methods exploiting multi-parametric biomarkers, especially based on neuroimaging, have huge potential to improve early diagnosis of dementia and to predict which individuals are at-risk of developing dementia. To benchmark algorithms in the field of machine learning and neuroimaging in dementia and assess their potential for use in clinical practice and clinical trials, seven grand challenges have been organized in the last decade: MIRIAD (2012), Alzheimer's Disease Big Data DREAM (2014), CADDementia (2014), Machine Learning Challenge (2014), MCI Neuroimaging (2017), TADPOLE (2017), and the Predictive Analytics Competition (2019). Based on two challenge evaluation frameworks, we analyzed how these grand challenges are complementing each other regarding research questions, datasets, validation approaches, results and impact. The seven grand challenges addressed questions related to screening, clinical status estimation, prediction and monitoring in (pre-clinical) dementia. There was little overlap in clinical questions, tasks and performance metrics. Whereas this aids providing insight on a broad range of questions, it also limits the validation of results across challenges. The validation process itself was mostly comparable between challenges, using similar methods for ensuring objective comparison, uncertainty estimation and statistical testing. In general, winning algorithms performed rigorous data pre-processing and combined a wide range of input features. Despite high state-of-the-art performances, most of the methods evaluated by the challenges are not clinically used. To increase impact, future challenges could pay more attention to statistical analysis of which factors (i.e., features, models) relate to higher performance, to clinical questions beyond Alzheimer's disease, and to using testing data beyond the Alzheimer's Disease Neuroimaging Initiative. Grand challenges would be an ideal venue for assessing the generalizability of algorithm performance to unseen data of other cohorts. Key for increasing impact in this way are larger testing data sizes, which could be reached by sharing algorithms rather than data to exploit data that cannot be shared. Given the potential and lessons learned in the past ten years, we are excited by the prospects of grand challenges in machine learning and neuroimaging for the next ten years and beyond.

Privacy-preserving harmonization via distributed ComBat

Challenges in clinical data sharing and the need to protect data privacy have led to the development and popularization of methods that do not require directly transferring patient data. In neuroimaging, integration of data across multiple institutions also introduces unwanted biases driven by scanner differences. These scanner effects have been shown by several research groups to severely affect downstream analyses. To facilitate the need of removing scanner effects in a distributed data setting, we introduce distributed ComBat, an adaptation of a popular harmonization method for multivariate data that borrows information across features. We present our fast and simple distributed algorithm and show that it yields equivalent results using data from the Alzheimer's Disease Neuroimaging Initiative. Our method enables harmonization while ensuring maximal privacy protection, thus facilitating a broad range of downstream analyses in functional and structural imaging studies.

Multimodal biological brain age prediction using magnetic resonance imaging and angiography with the identification of predictive regions

Biological brain age predicted using machine learning models based on high-resolution imaging data has been suggested as a potential biomarker for neurological and cerebrovascular diseases. In this work, we aimed to develop deep learning models to predict the biological brain age using structural magnetic resonance imaging and angiography datasets from a large database of 2074 adults (21-81 years). Since different imaging modalities can provide complementary information, combining them might allow to identify more complex aging patterns, with angiography data, for instance, showing vascular aging effects complementary to the atrophic brain tissue changes seen in T1-weighted MRI sequences. We used saliency maps to investigate the contribution of cortical, subcortical, and arterial structures to the prediction. Our results show that combining T1-weighted and angiography MR data led to a significantly improved brain age prediction accuracy, with a mean absolute error of 3.85 years comparing the predicted and chronological age. The most predictive brain regions included the lateral sulcus, the fourth ventricle, and the amygdala, while the brain arteries contributing the most to the prediction included the basilar artery, the middle cerebral artery M2 segments, and the left posterior cerebral artery. Our study proposes a framework for brain age prediction using multimodal imaging, which gives accurate predictions and allows identifying the most predictive regions for this task, which can serve as a surrogate for the brain regions that are most affected by aging.

Agreement Between Automated and Manual MRI Volumetry in Alzheimer's Disease: A Systematic Review and Meta-Analysis

BACKGROUND

Automated magnetic resonance imaging (MRI) volumetry is a promising tool to evaluate regional brain volumes in dementia and especially Alzheimer's disease (AD).

PURPOSE

To compare automated methods and the gold standard manual segmentation in measuring regional brain volumes on MRI across healthy controls, patients with mild cognitive impairment, and patients with dementia due to AD.

STUDY TYPE

Systematic review and meta-analysis.

DATA SOURCES

MEDLINE, Embase, and PsycINFO were searched through October 2021.

FIELD STRENGTH

1.0 T, 1.5 T, or 3.0 T.

ASSESSMENT

Two review authors independently identified studies for inclusion and extracted data. Methodological quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2).

STATISTICAL TESTS

Standardized mean differences (SMD; Hedges' g) were pooled using random-effects meta-analysis with robust variance estimation. Subgroup analyses were undertaken to explore potential sources of heterogeneity. Sensitivity analyses were conducted to examine the impact of the within-study correlation between effect estimates on the meta-analysis results.

RESULTS

Seventeen studies provided sufficient data to evaluate the hippocampus, lateral ventricles, and parahippocampal gyrus. The pooled SMD for the hippocampus, lateral ventricles, and parahippocampal gyrus were 0.22 (95% CI -0.50 to 0.93), 0.12 (95% CI -0.13 to 0.37), and -0.48 (95% CI -1.37 to 0.41), respectively. For the hippocampal data, subgroup analyses suggested that the pooled SMD was invariant across clinical diagnosis and field strength. Subgroup analyses could not be conducted on the lateral ventricles data and the parahippocampal gyrus data due to insufficient data. The results were robust to the selected within-study correlation value.

DATA CONCLUSION

While automated methods are generally comparable to manual segmentation for measuring hippocampal, lateral ventricle, and parahippocampal gyrus volumes, wide 95% CIs and large heterogeneity suggest that there is substantial uncontrolled variance. Thus, automated methods may be used to measure these regions in patients with AD but should be used with caution.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 3.

Mitigating site effects in covariance for machine learning in neuroimaging data

To acquire larger samples for answering complex questions in neuroscience, researchers have increasingly turned to multi‐site neuroimaging studies. However, these studies are hindered by differences in images acquired across multiple sites. These effects have been shown to bias comparison between sites, mask biologically meaningful associations, and even introduce spurious associations. To address this, the field has focused on harmonizing data by removing site‐related effects in the mean and variance of measurements. Contemporaneously with the increase in popularity of multi‐center imaging, the use of machine learning (ML) in neuroimaging has also become commonplace. These approaches have been shown to provide improved sensitivity, specificity, and power due to their modeling the joint relationship across measurements in the brain. In this work, we demonstrate that methods for removing site effects in mean and variance may not be sufficient for ML. This stems from the fact that such methods fail to address how correlations between measurements can vary across sites. Data from the Alzheimer's Disease Neuroimaging Initiative is used to show that considerable differences in covariance exist across sites and that popular harmonization techniques do not address this issue. We then propose a novel harmonization method called Correcting Covariance Batch Effects (CovBat) that removes site effects in mean, variance, and covariance. We apply CovBat and show that within‐site correlation matrices are successfully harmonized. Furthermore, we find that ML methods are unable to distinguish scanner manufacturer after our proposed harmonization is applied, and that the CovBat‐harmonized data retain accurate prediction of disease group.

Protocol for magnetic resonance imaging acquisition, quality assurance, and quality check for the Accelerator program for Discovery in Brain disorders using Stem cells

OBJECTIVE

The Accelerator program for Discovery in Brain disorders using Stem cells (ADBS) is a longitudinal study on five cohorts of patients with major psychiatric disorders from genetically high-risk families, their unaffected first-degree relatives, and healthy subjects. We describe the ADBS protocols for acquisition, quality assurance (QA), and quality check (QC) for multimodal magnetic resonance brain imaging studies.

METHODS

We describe the acquisition and QC protocols for structural, functional, and diffusion images. For QA, we acquire proton density and functional images on phantoms, along with repeated scans on human volunteer. We describe the analysis of phantom data and test-retest reliability of volumetric and diffusion measures.

RESULTS

Analysis of acquired phantom data shows linearity of proton density signal with increasing proton fraction, and an overall stability of various spatial and temporal QA measures. Examination of dice coefficient and statistical analyses of coefficient of variation in test-retest data on the human volunteer showed consistency of volumetric and diffusivity measures at whole-brain, regional, and voxel-level.

CONCLUSION

The described acquisition and QA-QC procedures can yield consistent and reliable quantitative measures. It is expected that this longitudinal neuroimaging dataset will, upon its release, serve the scientific community well and pave the way for interesting discoveries.

Integrating large-scale neuroimaging research datasets: Harmonisation of white matter hyperintensity measurements across Whitehall and UK Biobank datasets

•

We harmonised measures of WMHs across two studies on healthy ageing.

•

Specific pre-processing strategies can increase comparability across studies.

•

Modelling of biological differences is crucial to provide calibrated measures.

Large scale neuroimaging datasets present the possibility of providing normative distributions for a wide variety of neuroimaging markers, which would vastly improve the clinical utility of these measures. However, a major challenge is our current poor ability to integrate measures across different large-scale datasets, due to inconsistencies in imaging and non-imaging measures across the different protocols and populations. Here we explore the harmonisation of white matter hyperintensity (WMH) measures across two major studies of healthy elderly populations, the Whitehall II imaging sub-study and the UK Biobank. We identify pre-processing strategies that maximise the consistency across datasets and utilise multivariate regression to characterise study sample differences contributing to differences in WMH variations across studies. We also present a parser to harmonise WMH-relevant non-imaging variables across the two datasets. We show that we can provide highly calibrated WMH measures from these datasets with: (1) the inclusion of a number of specific standardised processing steps; and (2) appropriate modelling of sample differences through the alignment of demographic, cognitive and physiological variables. These results open up a wide range of applications for the study of WMHs and other neuroimaging markers across extensive databases of clinical data.

Comparison of structural MRI brain measures between 1.5 and 3 T: Data from the Lothian Birth Cohort 1936

Multi‐scanner MRI studies are reliant on understanding the apparent differences in imaging measures between different scanners. We provide a comprehensive analysis of T₁‐weighted and diffusion MRI (dMRI) structural brain measures between a 1.5 T GE Signa Horizon HDx and a 3 T Siemens Magnetom Prisma using 91 community‐dwelling older participants (aged 82 years). Although we found considerable differences in absolute measurements (global tissue volumes were measured as ~6–11% higher and fractional anisotropy [FA] was 33% higher at 3 T than at 1.5 T), between‐scanner consistency was good to excellent for global volumetric and dMRI measures (intraclass correlation coefficient [ICC] range: .612–.993) and fair to good for 68 cortical regions (FreeSurfer) and cortical surface measures (mean ICC: .504–.763). Between‐scanner consistency was fair for dMRI measures of 12 major white matter tracts (mean ICC: .475–.564), and the general factors of these tracts provided excellent consistency (ICC ≥ .769). Whole‐brain structural networks provided good to excellent consistency for global metrics (ICC ≥ .612). Although consistency was poor for individual network connections (mean ICCs: .275−.280), this was driven by a large difference in network sparsity (.599 vs. .334), and consistency was improved when comparing only the connections present in every participant (mean ICCs: .533–.647). Regression‐based k‐fold cross‐validation showed that, particularly for global volumes, between‐scanner differences could be largely eliminated (R² range .615–.991). We conclude that low granularity measures of brain structure can be reliably matched between the scanners tested, but caution is warranted when combining high granularity information from different scanners.

Grey matter alterations in restless legs syndrome: A coordinate-based meta-analysis

Brain structural abnormalities in idiopathic restless legs syndrome have long been debated. Voxel-based morphometry is an objective structural magnetic resonance imaging technique to investigate regional grey matter volume or density differences between groups. In the last decade, voxel-based morphometry studies have exhibited inconsistent and conflicting findings regarding the presence and localization of brain grey matter alterations in restless legs syndrome. We therefore conducted a coordinate-based meta-analysis to quantitatively examine whether there were consistent grey matter findings in restless legs syndrome using the latest algorithms, seed-based d mapping with permutation of subject images. We included 12 voxel-based morphometry studies (13 datasets, 375 patients and 385 healthy controls). Our coordinate-based meta-analysis did not identify evidence of consistent grey matter alterations in restless legs syndrome. Grey matter alterations via voxel-based morphometry analysis are not therefore recommended to be used as a reliable surrogate neuroimaging marker for restless legs syndrome. This lack of consistency may be attributed to differences in sample size, genetics, gender distribution and age at onset, clinical heterogeneity (clinical course, anatomical distribution of symptoms, disease severity, disease duration, abnormal sensory profiles and comorbidity), and variations in imaging acquisition, data processing and statistical strategies. Longitudinal studies with multimodal neuroimaging techniques are needed to determine whether structural changes are dynamic and secondary to functional abnormalities.

Support vector machine-based classification of schizophrenia patients and healthy controls using structural magnetic resonance imaging from two independent sites

Structural brain alterations have been repeatedly reported in schizophrenia; however, the pathophysiology of its alterations remains unclear. Multivariate pattern recognition analysis such as support vector machines can classify patients and healthy controls by detecting subtle and spatially distributed patterns of structural alterations. We aimed to use a support vector machine to distinguish patients with schizophrenia from control participants on the basis of structural magnetic resonance imaging data and delineate the patterns of structural alterations that significantly contributed to the classification performance. We used independent datasets from different sites with different magnetic resonance imaging scanners, protocols and clinical characteristics of the patient group to achieve a more accurate estimate of the classification performance of support vector machines. We developed a support vector machine classifier using the dataset from one site (101 participants) and evaluated the performance of the trained support vector machine using a dataset from the other site (97 participants) and vice versa. We assessed the performance of the trained support vector machines in each support vector machine classifier. Both support vector machine classifiers attained a classification accuracy of >70% with two independent datasets indicating a consistently high performance of support vector machines even when used to classify data from different sites, scanners and different acquisition protocols. The regions contributing to the classification accuracy included the bilateral medial frontal cortex, superior temporal cortex, insula, occipital cortex, cerebellum, and thalamus, which have been reported to be related to the pathogenesis of schizophrenia. These results indicated that the support vector machine could detect subtle structural brain alterations and might aid our understanding of the pathophysiology of these changes in schizophrenia, which could be one of the diagnostic findings of schizophrenia.

Detect and correct bias in multi-site neuroimaging datasets

The desire to train complex machine learning algorithms and to increase the statistical power in association studies drives neuroimaging research to use ever-larger datasets. The most obvious way to increase sample size is by pooling scans from independent studies. However, simple pooling is often ill-advised as selection, measurement, and confounding biases may creep in and yield spurious correlations. In this work, we combine 35,320 magnetic resonance images of the brain from 17 studies to examine bias in neuroimaging. In the first experiment, Name That Dataset, we provide empirical evidence for the presence of bias by showing that scans can be correctly assigned to their respective dataset with 71.5% accuracy. Given such evidence, we take a closer look at confounding bias, which is often viewed as the main shortcoming in observational studies. In practice, we neither know all potential confounders nor do we have data on them. Hence, we model confounders as unknown, latent variables. Kolmogorov complexity is then used to decide whether the confounded or the causal model provides the simplest factorization of the graphical model. Finally, we present methods for dataset harmonization and study their ability to remove bias in imaging features. In particular, we propose an extension of the recently introduced ComBat algorithm to control for global variation across image features, inspired by adjusting for unknown population stratification in genetics. Our results demonstrate that harmonization can reduce dataset-specific information in image features. Further, confounding bias can be reduced and even turned into a causal relationship. However, harmonization also requires caution as it can easily remove relevant subject-specific information. Code is available at https://github.com/ai-med/Dataset-Bias.

Interobserver variability in quality assessment of magnetic resonance images

Background

The perceptual quality of magnetic resonance (MR) images influences diagnosis and may compromise the treatment. The purpose of this study was to evaluate how the image quality changes influence the interobserver variability of their assessment.

Methods

For the variability evaluation, a dataset containing distorted MRI images was prepared and then assessed by 31 experienced medical professionals (radiologists). Differences between observers were analyzed using the Fleiss’ kappa. However, since the kappa evaluates the agreement among radiologists taking into account aggregated decisions, a typically employed criterion of the image quality assessment (IQA) performance was used to provide a more thorough analysis. The IQA performance of radiologists was evaluated by comparing the Spearman correlation coefficients, ρ, between individual scores with the mean opinion scores (MOS) composed of the subjective opinions of the remaining professionals.

Results

The experiments show that there is a significant agreement among radiologists (κ=0.12; 95% confidence interval [CI]: 0.118, 0.121; P<0.001) on the quality of the assessed images. The resulted κ is strongly affected by the subjectivity of the assigned scores, separately presenting close scores. Therefore, the ρ was used to identify poor performance cases and to confirm the consistency of the majority of collected scores (ρ_mean = 0.5706). The results for interns (ρ_mean = 0.6868) supports the finding that the quality assessment of MR images can be successfully taught.

Conclusions

The agreement observed among radiologists from different imaging centers confirms the subjectivity of the perception of MR images. It was shown that the image content and severity of distortions affect the IQA. Furthermore, the study highlights the importance of the psychosomatic condition of the observers and their attitude.

T1-weighted MRI-driven Brain Age Estimation in Alzheimer's Disease and Parkinson's Disease

Neuroimaging-driven brain age estimation has introduced a robust (reliable and heritable) biomarker for detecting and monitoring neurodegenerative diseases. Here, we computed and compared brain age in Alzheimer's disease (AD) and Parkinson's disease (PD) patients using an advanced machine learning procedure involving T1-weighted MRI scans and gray matter (GM) and white matter (WM) models. Brain age estimation frameworks were built using 839 healthy individuals and then the brain estimated age difference (Brain-EAD: chronological age subtracted from brain estimated age) was assessed in a large sample of PD patients (n = 160) and AD patients (n = 129), respectively. The mean Brain-EADs for GM were +9.29 ± 6.43 years for AD patients versus +1.50 ± 6.03 years for PD patients. For WM, the mean Brain-EADs were +8.85 ± 6.62 years for AD patients versus +2.47 ± 5.85 years for PD patients. In addition, PD patients showed a significantly higher WM Brain-EAD than GM Brain-EAD. In a direct comparison between PD and AD patients, we observed significantly higher Brain-EAD values in AD patients for both GM and WM. A comparison of the Brain-EAD between PD and AD patients revealed that AD patients may have a significantly "older-appearing" brain than PD patients.

Basal forebrain volume reliably predicts the cortical spread of Alzheimer's degeneration

Alzheimer's disease neurodegeneration is thought to spread across anatomically and functionally connected brain regions. However, the precise sequence of spread remains ambiguous. The prevailing model used to guide in vivo human neuroimaging and non-human animal research assumes that Alzheimer's degeneration starts in the entorhinal cortices, before spreading to the temporoparietal cortex. Challenging this model, we previously provided evidence that in vivo markers of neurodegeneration within the nucleus basalis of Meynert (NbM), a subregion of the basal forebrain heavily populated by cortically projecting cholinergic neurons, precedes and predicts entorhinal degeneration. There have been few systematic attempts at directly comparing staging models using in vivo longitudinal biomarker data, and none to our knowledge testing if comparative evidence generalizes across independent samples. Here we addressed the sequence of pathological staging in Alzheimer's disease using two independent samples of the Alzheimer's Disease Neuroimaging Initiative (n1 = 284; n2 = 553) with harmonized CSF assays of amyloid-β and hyperphosphorylated tau (pTau), and longitudinal structural MRI data over 2 years. We derived measures of grey matter degeneration in a priori NbM and the entorhinal cortical regions of interest. To examine the spreading of degeneration, we used a predictive modelling strategy that tests whether baseline grey matter volume in a seed region accounts for longitudinal change in a target region. We demonstrated that predictive spread favoured the NbM→entorhinal over the entorhinal→NbM model. This evidence generalized across the independent samples. We also showed that CSF concentrations of pTau/amyloid-β moderated the observed predictive relationship, consistent with evidence in rodent models of an underlying trans-synaptic mechanism of pathophysiological spread. The moderating effect of CSF was robust to additional factors, including clinical diagnosis. We then applied our predictive modelling strategy to an exploratory whole-brain voxel-wise analysis to examine the spatial specificity of the NbM→entorhinal model. We found that smaller baseline NbM volumes predicted greater degeneration in localized regions of the entorhinal and perirhinal cortices. By contrast, smaller baseline entorhinal volumes predicted degeneration in the medial temporal cortex, recapitulating a prior influential staging model. Our findings suggest that degeneration of the basal forebrain cholinergic projection system is a robust and reliable upstream event of entorhinal and neocortical degeneration, calling into question a prevailing view of Alzheimer's disease pathogenesis.

Towards Precision Medicine in Psychosis: Benefits and Challenges of Multimodal Multicenter Studies-PSYSCAN: Translating Neuroimaging Findings From Research into Clinical Practice

In the last 2 decades, several neuroimaging studies investigated brain abnormalities associated with the early stages of psychosis in the hope that these could aid the prediction of onset and clinical outcome. Despite advancements in the field, neuroimaging has yet to deliver. This is in part explained by the use of univariate analytical techniques, small samples and lack of statistical power, lack of external validation of potential biomarkers, and lack of integration of nonimaging measures (eg, genetic, clinical, cognitive data). PSYSCAN is an international, longitudinal, multicenter study on the early stages of psychosis which uses machine learning techniques to analyze imaging, clinical, cognitive, and biological data with the aim of facilitating the prediction of psychosis onset and outcome. In this article, we provide an overview of the PSYSCAN protocol and we discuss benefits and methodological challenges of large multicenter studies that employ neuroimaging measures.

Measured volumes using segmented tissue probability data obtained using statistical parametric mapping 12 were not influenced by the contrasts of analyzed images

The aim of this study was to evaluate changes in measured volumes using Statistical Parametric Mapping (SPM) 12 caused by contrast changes in magnetic resonance (MR) image. Twenty-one healthy subjects participated in the study. From all subjects, 3D T1-weighted images (T1WIs) were obtained using a 3T scanner. In the first step of creating reference volume data, we used SPM12 to binarize all the segmented data. In the second step, we assigned simulated 3D-T1WI signal intensities to each tissue image and used the following values. The last step was integration of each tissue image to generate 3D-T1WI simulated reference volume data for each participant. To create the reference 3D-T1WIs with various contrasts from the reference volume data, we varied the signal intensity of gray matter from 900 to 600, 700, 1100, 1300, and 1400. The reference 3D-T2WI was acquired using the method used for 3D-T1WIs. Then, six 3D-T1WIs were processed using intrasubject bias-correction processing with SPM12, resulting in six new 3D-T1WIs of nonuniform signal intensities. Thirteen volume data sets were segmented into native-space tissue probability data using SPM12. Examination of the 3D data without nonuniform signal intensity showed that significant differences in measured volumes were not observed on repeated analysis of variance, but examination of the 3D data with nonuniform signal intensity did show significant differences in measured volumes in gray matter and CSF but not in white matter. Measured volumes using segmented tissue probability data obtained using SPM12 were not influenced by the contrasts of analyzed images.

Inter-site harmonization based on dual generative adversarial networks for diffusion tensor imaging: application to neonatal white matter development

Background

Site-specific variations are challenges for pooling analyses in multi-center studies. This work aims to propose an inter-site harmonization method based on dual generative adversarial networks (GANs) for diffusion tensor imaging (DTI) derived metrics on neonatal brains.

Results

DTI-derived metrics (fractional anisotropy, FA; mean diffusivity, MD) are obtained on age-matched neonates without magnetic resonance imaging (MRI) abnormalities: 42 neonates from site 1 and 42 neonates from site 2. Significant inter-site differences of FA can be observed. The proposed harmonization approach and three conventional methods (the global-wise scaling, the voxel-wise scaling, and the ComBat) are performed on DTI-derived metrics from two sites. During the tract-based spatial statistics, inter-site differences can be removed by the proposed dual GANs method, the voxel-wise scaling, and the ComBat. Among these methods, the proposed method holds the lowest median values in absolute errors and root mean square errors. During the pooling analysis of two sites, Pearson correlation coefficients between FA and the postmenstrual age after harmonization are larger than those before harmonization. The effect sizes (Cohen’s d between males and females) are also maintained by the harmonization procedure.

Conclusions

The proposed dual GANs-based harmonization method is effective to harmonize neonatal DTI-derived metrics from different sites. Results in this study further suggest that the GANs-based harmonization is a feasible pre-processing method for pooling analyses in multi-center studies.

Differentiation of schizophrenia using structural MRI with consideration of scanner differences: A real-world multisite study

AIM

Neuroimaging studies have revealed that patients with schizophrenia exhibit reduced gray matter volume in various regions. With these findings, various studies have indicated that structural MRI can be useful for the diagnosis of schizophrenia. However, multisite studies are limited. Here, we evaluated a simple model that could be used to differentiate schizophrenia from control subjects considering MRI scanner differences employing voxel-based morphometry.

METHODS

Subjects were 541 patients with schizophrenia and 1252 healthy volunteers. Among them, 95 patients and 95 controls (Dataset A) were used for the generation of regions of interest (ROI), and the rest (Dataset B) were used to evaluate our method. The two datasets were comprised of different subjects. Three-dimensional T1-weighted MRI scans were taken for all subjects and gray-matter images were extracted. To differentiate schizophrenia, we generated ROI for schizophrenia from Dataset A. Then, we determined volume within the ROI for each subject from Dataset B. Using the extracted volume data, we calculated a differentiation feature considering age, sex, and intracranial volume for each MRI scanner. Receiver-operator curve analyses were performed to evaluate the differentiation feature.

RESULTS

The area under the curve ranged from 0.74 to 0.84, with accuracy from 69% to 76%. Receiver-operator curve analysis with all samples revealed an area under the curve of 0.76 and an accuracy of 73%.

CONCLUSION

We moderately successfully differentiated schizophrenia from control using structural MRI from differing scanners from multiple sites. This could be useful for applying neuroimaging techniques to clinical settings for the accurate diagnosis of schizophrenia.

Long-Term Test-Retest Reliability of Auditory Gamma Oscillations Between Different Clinical EEG Systems

OBJECTIVE

There is increasing interest in the utility of gamma-band activity for assessing various brain functions, including perception, language, memory, and cognition. The auditory steady-state response (ASSR) involves neural activity in the brain elicited by trains of a click sound, and its maximum response is obtained at 40 Hz (40-Hz ASSR). Abnormalities of the 40-Hz ASSR are also widely reported in patients with schizophrenia. Thus, the test-retest reliability of the ASSR is important for its clinical and translational application. However, there are only limited studies reporting the short-term reliability between acquisitions at two time points made using the same electroencephalogram (EEG) system. Furthermore, the long-term reliability between multiple EEG systems and the reliability of spontaneous gamma activity are unknown but are crucial for multicenter collaborative research.

METHODS

We examined the long-term test-retest reliability of 40-Hz ASSR oscillatory activities indexed by the phase locking factor (PLF), evoked power, and (non-phase-locked) induced power between two clinical 19-electrode EEG systems [recorded twice for EEG-1 (time1 and time2) and EEG-2 (time3 and time4)] at four time points from 14 healthy controls over a duration of 5 months. Test-retest reliability was examined using intraclass correlation coefficients (ICCs).

RESULTS

Both PLF and evoked power showed good to excellent ICCs (>0.60), mainly in the Fz-electrode, both within each EEG system-EEG-1 [(time1 vs. time2) PLF: ICC = 0.66, evoked power: ICC = 0.88] and EEG-2 [(time3 vs. time4) PLF: ICC = 0.82, evoked power: ICC = 0.77]-and between the two EEG systems [(EEG-1 vs. EEG-2) PLF: ICC = 0.73, evoked power: ICC = 0.84]. In contrast, induced power showed the highest (excellent) ICC between the two EEG systems (ICC = 0.95) mainly in the Cz-electrode. For PLF, the Fz-electrode showed better test-retest reliability across all EEG recordings than the Cz-electrode (Fz: ICC = 0.67, Cz: ICC = 0.63), whereas we found similar excellent reproducibility across all EEG recordings from both electrodes for evoked power (Fz: ICC = 0.79, Cz: ICC = 0.77) and induced power (Fz: ICC = 0.79, Cz: ICC = 0.80).

CONCLUSION

The 40-Hz ASSR oscillatory activities, including induced power, showed excellent test-retest reliability, even when using different EEG systems over a duration of 5 months. These findings confirm the utility of the 40-Hz ASSR as a reliable clinical and translatable biomarker for multicenter collaborative research.

Improving individual predictions: Machine learning approaches for detecting and attacking heterogeneity in schizophrenia (and other psychiatric diseases)

Psychiatric diseases are very heterogeneous both in clinical manifestation and etiology. With the recent rise of using machine learning techniques to attempt to diagnose and prognose these disorders, the issue of heterogeneity becomes increasingly important. With the growing interest in personalized medicine, it becomes even more important to not only classify someone as a patient with a certain disorder, its treatment needs a more precise definition of the underlying neurobiology, since different biological origins of the same disease may require (very) different treatments. We review the possible contributions that machine learning techniques could make to explore the heterogeneous nature of psychiatric disorders with a focus on schizophrenia. First we will review how heterogeneity shows up and how machine learning, or multivariate pattern recognition methods in general, can be used to discover it. Secondly, we will discuss the possible uses of these techniques to attack heterogeneity, leading to improved predictions and understanding of the neurobiological background of the disorder.

Clinical Strategies and Technical Challenges in Psychoradiology

Psychoradiology is an emerging discipline at the intersection between radiology and psychiatry. It holds promise for playing a role in clinical diagnosis, evaluation of treatment response and prognosis, and illness risk prediction for patients with psychiatric disorders. Addressing complex issues, such as the biological heterogeneity of psychiatric syndromes and unclear neurobiological mechanisms underpinning radiological abnormalities, is a challenge that needs to be resolved. With the advance of multimodal imaging and more efforts in standardization of image acquisition and analysis, psychoradiology is becoming a promising tool for the future of clinical care for patients with psychiatric disorders.

Measurement Variability Following MRI System Upgrade

Major hardware/software changes to MRI platforms, either planned or unplanned, will almost invariably occur in longitudinal studies. Our objective was to assess the resulting variability on relevant imaging measurements in such context, specifically for three Siemens Healthcare Magnetom Trio upgrades to the Prisma^fit platform. We report data acquired on three healthy volunteers scanned before and after three different platform upgrades. We assessed differences in image signal [contrast-to-noise ratio (CNR)] on T1-weighted images (T1w) and fluid-attenuated inversion recovery images (FLAIR); brain morphometry on T1w image; and small vessel disease (white matter hyperintensities; WMH) on FLAIR image. Prisma^fit upgrade resulted in higher (30%) and more variable neocortical CNR and larger brain volume and thickness mainly in frontal areas. A significant relationship was observed between neocortical CNR and neocortical volume. For FLAIR images, no significant CNR difference was observed, but WMH volumes were significantly smaller (-68%) after Prisma^fit upgrade, when compared to results on the Magnetom Trio. Together, these results indicate that Prisma^fit upgrade significantly influenced image signal, brain morphometry measures and small vessel diseases measures and that these effects need to be taken into account when analyzing results from any longitudinal study undergoing similar changes.

Lifespan normative data on rates of brain volume changes

We provide here normative values of yearly percentage brain volume change (PBVC/y) as obtained with Structural Imaging Evaluation, using Normalization, of Atrophy, a widely used open-source software, developing a PBVC/y calculator for assessing the deviation from the expected PBVC/y in patients with neurological disorders. We assessed multicenter (34 centers, 11 acquisition protocols) magnetic resonance imaging data of 720 healthy participants covering the whole adult lifespan (16-90 years). Data of 421 participants with a follow-up > 6 months were used to obtain the normative values for PBVC/y and data of 392 participants with a follow-up <1 month were selected to assess the intrasubject variability of the brain volume measurement. A mixed model evaluated PBVC/y dependence on age, sex, and magnetic resonance imaging parameters (scan vendor and magnetic field strength). PBVC/y was associated with age (p < 0.001), with 60- to 70-year-old participants showing twice more volume decrease than participants aged 30-40 years. PBVC/y was also associated with magnetic field strength, with higher decreases when measured by 1.5T than 3T scanners (p < 0.001). The variability of PBVC/y normative percentiles was narrower as the interscan interval was longer (e.g., 80th normative percentile was 50% smaller for participants with 2-year than with 1-year follow-up). The use of these normative data, eased by the freely available calculator, might help in better discriminating pathological from physiological conditions in the clinical setting.

The power of sample size through a multi-scanner approach in MR neuroimaging regression analysis: evidence from Alzheimer's disease with and without depression

The inconsistency of volumetric results often seen in MR neuroimaging studies can be partially attributed to small sample sizes and variable data analysis approaches. Increased sample size through multi-scanner studies can tackle the former, but combining data across different scanner platforms and field-strengths may introduce a variability factor capable of masking subtle statistical differences. To investigate the sample size effect on regression analysis between depressive symptoms and grey matter volume (GMV) loss in Alzheimer's disease (AD), a retrospective multi-scanner investigation was conducted. A cohort of 172 AD patients, with or without comorbid depressive symptoms, was studied. Patients were scanned with different imaging protocols in four different MRI scanners operating at either 1.5 T or 3.0 T. Acquired data were uniformly analyzed using the computational anatomy toolbox (CAT12) of the statistical parametric mapping (SPM12) software. Single- and multi-scanner regression analyses were applied to identify the anatomical pattern of correlation between GM loss and depression severity. A common anatomical pattern of correlation between GMV loss and increased depression severity, mostly involving sensorimotor areas, was identified in all patient subgroups imaged in different scanners. Analysis of the pooled multi-scanner data confirmed the above finding employing a more conservative statistical criterion. In the retrospective multi-scanner setting, a significant correlation was also exhibited for temporal and frontal areas. Increasing the sample size by retrospectively pooling multi-scanner data, irrespective of the acquisition platform and parameters employed, can facilitate the identification of anatomical areas with a strong correlation between GMV changes and depression symptoms in AD patients.