Site effects how-to and when: An overview of retrospective techniques to accommodate site effects in multi-site neuroimaging analyses

Current best practices and future opportunities for reproducible findings using large-scale neuroimaging in psychiatry

Research into the brain basis of psychopathology is challenging due to the heterogeneity of psychiatric disorders, extensive comorbidities, underdiagnosis or overdiagnosis, multifaceted interactions with genetics and life experiences, and the highly multivariate nature of neural correlates. Therefore, increasingly larger datasets that measure more variables in larger cohorts are needed to gain insights. In this review, we present current "best practice" approaches for using existing databases, collecting and sharing new repositories for big data analyses, and future directions for big data in neuroimaging and psychiatry with an emphasis on contributing to collaborative efforts and the challenges of multi-study data analysis.

DeepComBat: A statistically motivated, hyperparameter-robust, deep learning approach to harmonization of neuroimaging data

Neuroimaging data acquired using multiple scanners or protocols are increasingly available. However, such data exhibit technical artifacts across batches which introduce confounding and decrease reproducibility. This is especially true when multi-batch data are analyzed using complex downstream models which are more likely to pick up on and implicitly incorporate batch-related information. Previously proposed image harmonization methods have sought to remove these batch effects; however, batch effects remain detectable in the data after applying these methods. We present DeepComBat, a deep learning harmonization method based on a conditional variational autoencoder and the ComBat method. DeepComBat combines the strengths of statistical and deep learning methods in order to account for the multivariate relationships between features while simultaneously relaxing strong assumptions made by previous deep learning harmonization methods. As a result, DeepComBat can perform multivariate harmonization while preserving data structure and avoiding the introduction of synthetic artifacts. We apply this method to cortical thickness measurements from a cognitive-aging cohort and show DeepComBat qualitatively and quantitatively outperforms existing methods in removing batch effects while preserving biological heterogeneity. Additionally, DeepComBat provides a new perspective for statistically motivated deep learning harmonization methods.

ComBatLS: A location- and scale-preserving method for multi-site image harmonization

Recent work has leveraged massive datasets and advanced harmonization methods to construct normative models of neuroanatomical features and benchmark individuals' morphology. However, current harmonization tools do not preserve the effects of biological covariates including sex and age on features' variances; this failure may induce error in normative scores, particularly when such factors are distributed unequally across sites. Here, we introduce a new extension of the popular ComBat harmonization method, ComBatLS, that preserves biological variance in features' locations and scales. We use UK Biobank data to show that ComBatLS robustly replicates individuals' normative scores better than other ComBat methods when subjects are assigned to sex-imbalanced synthetic "sites". Additionally, we demonstrate that ComBatLS significantly reduces sex biases in normative scores compared to traditional methods. Finally, we show that ComBatLS successfully harmonizes consortium data collected across over 50 studies. R implementation of ComBatLS is available at https://github.com/andy1764/ComBatFamily.

Harmonizing three-dimensional MRI using pseudo-warping field guided GAN

In pursuit of cultivating automated models for magnetic resonance imaging (MRI) to aid in diagnostics, an escalating demand for extensive, multisite, and heterogeneous brain imaging datasets has emerged. This potentially introduces biased outcomes when directly applied for subsequent analysis. Researchers have endeavored to address this issue by pursuing the harmonization of MRIs. However, most existing image-based harmonization methods for MRI are tailored for 2D slices, which may introduce inter-slice variations when they are combined into a 3D volume. In this study, we aim to resolve inconsistencies between slices by introducing a pseudo-warping field. This field is created randomly and utilized to transform a slice into an artificially warped subsequent slice. The objective of this pseudo-warping field is to ensure that generators can consistently harmonize adjacent slices to another domain, without being affected by the varying content present in different slices. Furthermore, we construct unsupervised spatial and recycle loss to enhance the spatial accuracy and slice-wise consistency across the 3D images. The results demonstrate that our model effectively mitigates inter-slice variations and successfully preserves the anatomical details of the images during the harmonization process. Compared to generative harmonization models that employ 3D operators, our model exhibits greater computational efficiency and flexibility.

Individual Brain Charting dataset extension, third release for movie watching and retinotopy data

The Individual Brain Charting (IBC) is a multi-task functional Magnetic Resonance Imaging dataset acquired at high spatial-resolution and dedicated to the cognitive mapping of the human brain. It consists in the deep phenotyping of twelve individuals, covering a broad range of psychological domains suitable for functional-atlasing applications. Here, we present the inclusion of task data from both naturalistic stimuli and trial-based designs, to uncover structures of brain activation. We rely on the Fast Shared Response Model (FastSRM) to provide a data-driven solution for modelling naturalistic stimuli, typically containing many features. We show that data from left-out runs can be reconstructed using FastSRM, enabling the extraction of networks from the visual, auditory and language systems. We also present the topographic organization of the visual system through retinotopy. In total, six new tasks were added to IBC, wherein four trial-based retinotopic tasks contributed with a mapping of the visual field to the cortex. IBC is open access: source plus derivatives imaging data and meta-data are available in public repositories.

Orienting role of the putative human posterior infero-temporal area in visual attention

The dorsal attention network (DAN) is a network of brain regions essential for attentional orienting, which includes the lateral intraparietal area (LIP) and frontal eye field (FEF). Recently, the putative human dorsal posterior infero-temporal area (phPITd) has been identified as a new node of the DAN. However, its functional relationship with other areas of the DAN and its specific role in visual attention remained unclear. In this study, we analyzed a large publicly available neuroimaging dataset to investigate the intrinsic functional connectivities (FCs) of the phPITd with other brain areas. The results showed that the intrinsic FCs of the phPITd with the areas of the visual network and the DAN were significantly stronger than those with the ventral attention network (VAN) areas and areas of other networks. We further conducted individual difference analyses with a sample size of 295 participants and a series of attentional tasks to investigate which attentional components each phPITd-based DAN edge predicts. Our findings revealed that the intrinsic FC of the left phPITd with the LIPv could predict individual ability in attentional orienting, but not in alerting, executive control, and distractor suppression. Our results not only provide direct evidence of the phPITd's functional relationship with the LIPv, but also offer a comprehensive understanding of its specific role in visual attention.

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.

Distinct Volume Alterations of Thalamic Nuclei Across the Schizophrenia Spectrum

BACKGROUND AND HYPOTHESIS

Abnormal thalamic nuclei volumes and their link to cognitive impairments have been observed in schizophrenia. However, whether and how this finding extends to the schizophrenia spectrum is unknown. We hypothesized a distinct pattern of aberrant thalamic nuclei volume across the spectrum and examined its potential associations with cognitive symptoms.

STUDY DESIGN

We performed a FreeSurfer-based volumetry of T1-weighted brain MRIs from 137 healthy controls, 66 at-risk mental state (ARMS) subjects, 89 first-episode psychosis (FEP) individuals, and 126 patients with schizophrenia to estimate thalamic nuclei volumes of six nuclei groups (anterior, lateral, ventral, intralaminar, medial, and pulvinar). We used linear regression models, controlling for sex, age, and estimated total intracranial volume, both to compare thalamic nuclei volumes across groups and to investigate their associations with positive, negative, and cognitive symptoms.

STUDY RESULTS

We observed significant volume alterations in medial and lateral thalamic nuclei. Medial nuclei displayed consistently reduced volumes across the spectrum compared to controls, while lower lateral nuclei volumes were only observed in schizophrenia. Whereas positive and negative symptoms were not associated with reduced nuclei volumes across all groups, higher cognitive scores were linked to lower volumes of medial nuclei in ARMS. In FEP, cognition was not linked to nuclei volumes. In schizophrenia, lower cognitive performance was associated with lower medial volumes.

CONCLUSIONS

Results demonstrate distinct thalamic nuclei volume reductions across the schizophrenia spectrum, with lower medial nuclei volumes linked to cognitive deficits in ARMS and schizophrenia. Data suggest a distinctive trajectory of thalamic nuclei abnormalities along the course of schizophrenia.

Fighting the scanner effect in brain MRI segmentation with a progressive level-of-detail network trained on multi-site data

Many clinical and research studies of the human brain require accurate structural MRI segmentation. While traditional atlas-based methods can be applied to volumes from any acquisition site, recent deep learning algorithms ensure high accuracy only when tested on data from the same sites exploited in training (i.e., internal data). Performance degradation experienced on external data (i.e., unseen volumes from unseen sites) is due to the inter-site variability in intensity distributions, and to unique artefacts caused by different MR scanner models and acquisition parameters. To mitigate this site-dependency, often referred to as the scanner effect, we propose LOD-Brain, a 3D convolutional neural network with progressive levels-of-detail (LOD), able to segment brain data from any site. Coarser network levels are responsible for learning a robust anatomical prior helpful in identifying brain structures and their locations, while finer levels refine the model to handle site-specific intensity distributions and anatomical variations. We ensure robustness across sites by training the model on an unprecedentedly rich dataset aggregating data from open repositories: almost 27,000 T1w volumes from around 160 acquisition sites, at 1.5 - 3T, from a population spanning from 8 to 90 years old. Extensive tests demonstrate that LOD-Brain produces state-of-the-art results, with no significant difference in performance between internal and external sites, and robust to challenging anatomical variations. Its portability paves the way for large-scale applications across different healthcare institutions, patient populations, and imaging technology manufacturers. Code, model, and demo are available on the project website.

White and gray matter alterations in bipolar I and bipolar II disorder subtypes compared with healthy controls - exploring associations with disease course and polygenic risk

Patients with bipolar disorder (BD) show alterations in both gray matter volume (GMV) and white matter (WM) integrity compared with healthy controls (HC). However, it remains unclear whether the phenotypically distinct BD subtypes (BD-I and BD-II) also exhibit brain structural differences. This study investigated GMV and WM differences between HC, BD-I, and BD-II, along with clinical and genetic associations. N = 73 BD-I, n = 63 BD-II patients and n = 136 matched HC were included. Using voxel-based morphometry and tract-based spatial statistics, main effects of group in GMV and fractional anisotropy (FA) were analyzed. Associations between clinical and genetic features and GMV or FA were calculated using regression models. For FA but not GMV, we found significant differences between groups. BD-I patients showed lower FA compared with BD-II patients (ptfce-FWE = 0.006), primarily in the anterior corpus callosum. Compared with HC, BD-I patients exhibited lower FA in widespread clusters (ptfce-FWE < 0.001), including almost all major projection, association, and commissural fiber tracts. BD-II patients also demonstrated lower FA compared with HC, although less pronounced (ptfce-FWE = 0.049). The results remained unchanged after controlling for clinical and genetic features, for which no independent associations with FA or GMV emerged. Our findings suggest that, at a neurobiological level, BD subtypes may reflect distinct degrees of disease expression, with increasing WM microstructure disruption from BD-II to BD-I. This differential magnitude of microstructural alterations was not clearly linked to clinical and genetic variables. These findings should be considered when discussing the classification of BD subtypes within the spectrum of affective disorders.

Large-Scale Neuroimaging of Mental Illness

Neuroimaging has provided important insights into the brain variations related to mental illness. Inconsistencies in prior studies, however, call for methods that lead to more replicable and generalizable brain markers that can reliably predict illness severity, treatment course, and prognosis. A paradigm shift is underway with large-scale international research teams actively pooling data and resources to drive consensus findings and test emerging methods aimed at achieving the goals of precision psychiatry. In parallel with large-scale psychiatric genomics studies, international consortia combining neuroimaging data are mapping the transdiagnostic brain signatures of mental illness on an unprecedented scale. This chapter discusses the major challenges, recent findings, and a roadmap for developing better neuroimaging-based tools and markers for mental illness.

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).

Empirical assessment of the assumptions of ComBat with diffusion tensor imaging

Purpose

Diffusion tensor imaging (DTI) is a magnetic resonance imaging technique that provides unique information about white matter microstructure in the brain but is susceptible to confounding effects introduced by scanner or acquisition differences. ComBat is a leading approach for addressing these site biases. However, despite its frequent use for harmonization, ComBat's robustness toward site dissimilarities and overall cohort size have not yet been evaluated in terms of DTI.

Approach

As a baseline, we match N = 358 participants from two sites to create a "silver standard" that simulates a cohort for multi-site harmonization. Across sites, we harmonize mean fractional anisotropy and mean diffusivity, calculated using participant DTI data, for the regions of interest defined by the JHU EVE-Type III atlas. We bootstrap 10 iterations at 19 levels of total sample size, 10 levels of sample size imbalance between sites, and 6 levels of mean age difference between sites to quantify (i) β AGE , the linear regression coefficient of the relationship between FA and age; (ii) γ ^ s f * , the ComBat-estimated site-shift; and (iii) δ ^ s f * , the ComBat-estimated site-scaling. We characterize the reliability of ComBat by evaluating the root mean squared error in these three metrics and examine if there is a correlation between the reliability of ComBat and a violation of assumptions.

Results

ComBat remains well behaved for β AGE when N > 162 and when the mean age difference is less than 4 years. The assumptions of the ComBat model regarding the normality of residual distributions are not violated as the model becomes unstable.

Conclusion

Prior to harmonization of DTI data with ComBat, the input cohort should be examined for size and covariate distributions of each site. Direct assessment of residual distributions is less informative on stability than bootstrap analysis. We caution use ComBat of in situations that do not conform to the above thresholds.

Harmonized diffusion MRI data and white matter measures from the Adolescent Brain Cognitive Development Study

The Adolescent Brain Cognitive Development (ABCD) Study® has collected data from over 10,000 children across 21 sites, providing insights into adolescent brain development. However, site-specific scanner variability has made it challenging to use diffusion MRI (dMRI) data from this study. To address this, a dataset of harmonized and processed ABCD dMRI data (from release 3) has been created, comprising quality-controlled imaging data from 9,345 subjects, focusing exclusively on the baseline session, i.e., the first time point of the study. This resource required substantial computational time (approx. 50,000 CPU hours) for harmonization, whole-brain tractography, and white matter parcellation. The dataset includes harmonized dMRI data, 800 white matter clusters, 73 anatomically labeled white matter tracts in full and low resolution, and 804 different dMRI-derived measures per subject (72.3 TB total size). Accessible via the NIMH Data Archive, it offers a large-scale dMRI dataset for studying structural connectivity in child and adolescent neurodevelopment. Additionally, several post-harmonization experiments were conducted to demonstrate the success of the harmonization process on the ABCD dataset.

Heritability estimation of subcortical volumes in a multi-ethnic multi-site cohort study

Heritability of regional subcortical brain volumes (rSBVs) describes the role of genetics in middle and inner brain development. rSBVs are highly heritable in adults but are not characterized well in adolescents. The Adolescent Brain Cognitive Development study (ABCD), taken over 22 US sites, provides data to characterize the heritability of subcortical structures in adolescence. In ABCD, site-specific effects co-occur with genetic effects which can bias heritability estimates. Existing methods adjusting for site effects require additional steps to adjust for site effects and can lead to inconsistent estimation. We propose a random-effect model-based method of moments approach that is a single step estimator and is a theoretically consistent estimator even when sites are imbalanced and performs well under simulations. We compare methods on rSBVs from ABCD. The proposed approach yielded heritability estimates similar to previous results derived from single-site studies. The cerebellum cortex and hippocampus were the most heritable regions (> 50%).

Brain structural covariance network features are robust markers of early heavy alcohol use

BACKGROUND AND AIMS

Recently, we demonstrated that a distinct pattern of structural covariance networks (SCN) from magnetic resonance imaging (MRI)-derived measurements of brain cortical thickness characterized young adults with alcohol use disorder (AUD) and predicted current and future problematic drinking in adolescents relative to controls. Here, we establish the robustness and value of SCN for identifying heavy alcohol users in three additional independent studies.

DESIGN AND SETTING

Cross-sectional and longitudinal studies using data from the Pediatric Imaging, Neurocognition and Genetics (PING) study (n = 400, age range = 14-22 years), the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA) (n = 272, age range = 17-22 years) and the Human Connectome Project (HCP) (n = 375, age range = 22-37 years).

CASES

Cases were defined based on heavy alcohol use patterns or former alcohol use disorder (AUD) diagnoses: 50, 68 and 61 cases were identified. Controls had none or low alcohol use or absence of AUD: 350, 204 and 314 controls were selected.

MEASUREMENTS

Graph theory metrics of segregation and integration were used to summarize SCN.

FINDINGS

Mirroring our prior findings, and across the three data sets, cases had a lower clustering coefficient [area under the curve (AUC) = -0.029, P = 0.002], lower modularity (AUC = -0.14, P = 0.004), lower average shortest path length (AUC = -0.078, P = 0.017) and higher global efficiency (AUC = 0.007, P = 0.010). Local efficiency differences were marginal (AUC = -0.017, P = 0.052). That is, cases exhibited lower network segregation and higher integration, suggesting that adjacent nodes (i.e. brain regions) were less similar in thickness whereas spatially distant nodes were more similar.

CONCLUSION

Structural covariance network (SCN) differences in the brain appear to constitute an early marker of heavy alcohol use in three new data sets and, more generally, demonstrate the utility of SCN-derived metrics to detect brain-related psychopathology.

Cardiometabolic health across menopausal years is linked to white matter hyperintensities up to a decade later

Introduction

The menopause transition is associated with several cardiometabolic risk factors. Poor cardiometabolic health is further linked to microvascular brain lesions, which can be detected as white matter hyperintensities (WMHs) using T2-FLAIR magnetic resonance imaging (MRI) scans. Females show higher risk for WMHs post-menopause, but it remains unclear whether changes in cardiometabolic risk factors underlie menopause-related increase in brain pathology.

Methods

In this study, we assessed whether cross-sectional measures of cardiometabolic health, including body mass index (BMI) and waist-to-hip ratio (WHR), blood lipids, blood pressure, and long-term blood glucose (HbA1c), as well as longitudinal changes in BMI and WHR, differed according to menopausal status at baseline in 9,882 UK Biobank females (age range 40-70 years, n premenopausal = 3,529, n postmenopausal = 6,353). Furthermore, we examined whether these cardiometabolic factors were associated with WMH outcomes at the follow-up assessment, on average 8.78 years after baseline.

Results

Postmenopausal females showed higher levels of baseline blood lipids (HDL β  = 0.14, p < 0.001, LDL β  = 0.20, p < 0.001, triglycerides β  = 0.12, p < 0.001) and HbA1c (β  = 0.24, p < 0.001) compared to premenopausal women, beyond the effects of age. Over time, BMI increased more in the premenopausal compared to the postmenopausal group (β  = -0.08, p < 0.001), while WHR increased to a similar extent in both groups (β  = -0.03, p = 0.102). The change in WHR was however driven by increased waist circumference only in the premenopausal group. While the group level changes in BMI and WHR were in general small, these findings point to distinct anthropometric changes in pre- and postmenopausal females over time. Higher baseline measures of BMI, WHR, triglycerides, blood pressure, and HbA1c, as well as longitudinal increases in BMI and WHR, were associated with larger WMH volumes (β range = 0.03-0.13, p ≤ 0.002). HDL showed a significant inverse relationship with WMH volume (β  = -0.27, p < 0.001).

Discussion

Our findings emphasise the importance of monitoring cardiometabolic risk factors in females from midlife through the menopause transition and into the postmenopausal phase, to ensure improved cerebrovascular outcomes in later years.

Predictive modelling of brain disorders with magnetic resonance imaging: A systematic review of modelling practices, transparency, and interpretability in the use of convolutional neural networks

Brain disorders comprise several psychiatric and neurological disorders which can be characterized by impaired cognition, mood alteration, psychosis, depressive episodes, and neurodegeneration. Clinical diagnoses primarily rely on a combination of life history information and questionnaires, with a distinct lack of discriminative biomarkers in use for psychiatric disorders. Symptoms across brain conditions are associated with functional alterations of cognitive and emotional processes, which can correlate with anatomical variation; structural magnetic resonance imaging (MRI) data of the brain are therefore an important focus of research, particularly for predictive modelling. With the advent of large MRI data consortia (such as the Alzheimer's Disease Neuroimaging Initiative) facilitating a greater number of MRI-based classification studies, convolutional neural networks (CNNs)-deep learning models well suited to image processing tasks-have become increasingly popular for research into brain conditions. This has resulted in a myriad of studies reporting impressive predictive performances, demonstrating the potential clinical value of deep learning systems. However, methodologies can vary widely across studies, making them difficult to compare and/or reproduce, potentially limiting their clinical application. Here, we conduct a qualitative systematic literature review of 55 studies carrying out CNN-based predictive modelling of brain disorders using MRI data and evaluate them based on three principles-modelling practices, transparency, and interpretability. We propose several recommendations to enhance the potential for the integration of CNNs into clinical care.

Elucidating salient site-specific functional connectivity features and site-invariant biomarkers in schizophrenia via deep neural networks

Schizophrenia is a highly heterogeneous disorder and salient functional connectivity (FC) features have been observed to vary across study sites, warranting the need for methods that can differentiate between site-invariant FC biomarkers and site-specific salient FC features. We propose a technique named Semi-supervised learning with data HaRmonisation via Encoder-Decoder-classifier (SHRED) to examine these features from resting state functional magnetic resonance imaging scans gathered from four sites. Our approach involves an encoder-decoder-classifier architecture that simultaneously performs data harmonisation and semi-supervised learning (SSL) to deal with site differences and labelling inconsistencies across sites respectively. The minimisation of reconstruction loss from SSL was shown to improve model performance even within small datasets whilst data harmonisation often led to lower model generalisability, which was unaffected using the SHRED technique. We show that our proposed model produces site-invariant biomarkers, most notably the connection between transverse temporal gyrus and paracentral lobule. Site-specific salient FC features were also elucidated, especially implicating the paracentral lobule for our local dataset. Our examination of these salient FC features demonstrates how site-specific features and site-invariant biomarkers can be differentiated, which can deepen our understanding of the neurobiology of schizophrenia.

ComBat Harmonization of Myocardial Radiomic Features Sensitive to Cardiac MRI Acquisition Parameters

Purpose

To investigate the effect of ComBat harmonization methods on the robustness of cardiac MRI-derived radiomic features to variations in imaging parameters.

Materials and Methods

This Health Insurance Portability and Accountability Act-compliant retrospective study used a publicly available data set of 11 healthy controls (mean age, 33 years ± 16 [SD]; six men) and five patients (mean age, 52 years ± 16; four men). A single midventricular short-axis section was acquired with 3-T MRI using cine balanced steady-state free precision, T1-weighted, T2-weighted, T1 mapping, and T2 mapping imaging sequences. Each sequence was acquired using baseline parameters and after variations in flip angle, spatial resolution, section thickness, and parallel imaging. Image registration was performed for all sequences at a per-individual level. Manual myocardial contouring was performed, and 1652 radiomic features per sequence were extracted using baseline and variations in imaging parameters. Radiomic feature stability to change in imaging parameters was assessed using Cohen d sensitivity. The stability of radiomic features was assessed both without and after ComBat harmonization of radiomic features. Three ComBat methods were studied: parametric, nonparametric, and Gaussian mixture model (GMM).

Results

For all sequences combined, 51.4% of features were robust to changes in imaging parameters when no ComBat method was applied. ComBat harmonization substantially increased the number of stable features to 95.1% (95% CI: 94.9, 95.3) when parametric ComBat was used and 90.9% (95% CI: 90.6, 91.2) when nonparametric ComBat was used. GMM combat resulted in only 52.6% stable features.

Conclusion

ComBat harmonization improved the stability of radiomic features to changes in imaging parameters across all cardiac MRI sequences.Keywords: Cardiac MRI, Radiomics, ComBat, Harmonization Supplemental material is available for this article. © RSNA, 2023.

Reproducibility in Neuroimaging Analysis: Challenges and Solutions

Recent years have marked a renaissance in efforts to increase research reproducibility in psychology, neuroscience, and related fields. Reproducibility is the cornerstone of a solid foundation of fundamental research-one that will support new theories built on valid findings and technological innovation that works. The increased focus on reproducibility has made the barriers to it increasingly apparent, along with the development of new tools and practices to overcome these barriers. Here, we review challenges, solutions, and emerging best practices with a particular emphasis on neuroimaging studies. We distinguish 3 main types of reproducibility, discussing each in turn. Analytical reproducibility is the ability to reproduce findings using the same data and methods. Replicability is the ability to find an effect in new datasets, using the same or similar methods. Finally, robustness to analytical variability refers to the ability to identify a finding consistently across variation in methods. The incorporation of these tools and practices will result in more reproducible, replicable, and robust psychological and brain research and a stronger scientific foundation across fields of inquiry.

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.

Probabilistic estimation for across-batch compatibility enhancement for amyloid PET

INTRODUCTION

It is necessary to accurately account for systematic differences due to variability in scanners, radiotracers, and acquisition protocols in multisite studies combining amyloid imaging data.

METHODS

We propose Probabilistic Estimation for Across-batch Compatibility Enhancement (PEACE), a fully Bayesian multimodal extension of the widely used ComBat harmonization model, and we apply it to harmonize regional amyloid positron emission tomography data from two scanners.

RESULTS

Simulations show that PEACE recovers true harmonized values better than ComBat, even for unimodal data. PEACE harmonization of multiscanner regional amyloid imaging data yields results that agree better with longitudinal data compared to ComBat, without removing the known biological effects of age or apolipoprotein E genotype.

DISCUSSION

PEACE outperforms ComBat in both unimodal and bimodal contexts, is applicable to multisite amyloid imaging data, and holds promise for the harmonization of other neuroimaging data over ComBat.

HIGHLIGHTS

We introduce PEACE, a fully Bayesian multimodal extension of ComBat harmonization.Simulations show that PEACE recovers true harmonized values better than ComBat.PEACE accurately harmonizes multiscanner regional amyloid imaging data.

Harmonized diffusion MRI data and white matter measures from the Adolescent Brain Cognitive Development Study

The Adolescent Brain Cognitive Development (ABCD) study has collected data from over 10,000 children across 21 sites, providing valuable insights into adolescent brain development. However, site-specific scanner variability has made it challenging to use diffusion MRI (dMRI) data from this study. To address this, a database of harmonized and processed ABCD dMRI data has been created, comprising quality-controlled imaging data from 9345 subjects. This resource required significant computational effort, taking ~50,000 CPU hours to harmonize the data, perform white matter parcellation, and run whole brain tractography. The database includes harmonized dMRI data, 800 white matter clusters, 73 anatomically labeled white matter tracts both in full-resolution (for analysis) and low-resolution (for visualization), and 804 different dMRI-derived measures per subject. It is available via the NIMH Data Archive and offers tremendous potential for scientific discoveries in structural connectivity studies of neurodevelopment in children and adolescents. Additionally, several post-harmonization experiments were conducted to demonstrate the success of the harmonization process on the ABCD dataset.

DeepComBat: A Statistically Motivated, Hyperparameter-Robust, Deep Learning Approach to Harmonization of Neuroimaging Data

Neuroimaging data from multiple batches (i.e. acquisition sites, scanner manufacturer, datasets, etc.) are increasingly necessary to gain new insights into the human brain. However, multi-batch data, as well as extracted radiomic features, exhibit pronounced technical artifacts across batches. These batch effects introduce confounding into the data and can obscure biological effects of interest, decreasing the generalizability and reproducibility of findings. This is especially true when multi-batch data is used alongside complex downstream analysis models, such as machine learning methods. Image harmonization methods seeking to remove these batch effects are important for mitigating these issues; however, significant multivariate batch effects remain in the data following harmonization by current state-of-the-art statistical and deep learning methods. We present DeepCombat, a deep learning harmonization method based on a conditional variational autoencoder architecture and the ComBat harmonization model. DeepCombat learns and removes subject-level batch effects by accounting for the multivariate relationships between features. Additionally, DeepComBat relaxes a number of strong assumptions commonly made by previous deep learning harmonization methods and is empirically robust across a wide range of hyperparameter choices. We apply this method to neuroimaging data from a large cognitive-aging cohort and find that DeepCombat outperforms existing methods, as assessed by a battery of machine learning methods, in removing scanner effects from cortical thickness measurements while preserving biological heterogeneity. Additionally, DeepComBat provides a new perspective for statistically-motivated deep learning harmonization methods.

Multiscale functional connectivity patterns of the aging brain learned from harmonized rsfMRI data of the multi-cohort iSTAGING study

To learn multiscale functional connectivity patterns of the aging brain, we built a brain age prediction model of functional connectivity measures at seven scales on a large fMRI dataset, consisting of resting-state fMRI scans of 4186 individuals with a wide age range (22 to 97 years, with an average of 63) from five cohorts. We computed multiscale functional connectivity measures of individual subjects using a personalized functional network computational method, harmonized the functional connectivity measures of subjects from multiple datasets in order to build a functional brain age model, and finally evaluated how functional brain age gap correlated with cognitive measures of individual subjects. Our study has revealed that functional connectivity measures at multiple scales were more informative than those at any single scale for the brain age prediction, the data harmonization significantly improved the brain age prediction performance, and the data harmonization in the functional connectivity measures' tangent space worked better than in their original space. Moreover, brain age gap scores of individual subjects derived from the brain age prediction model were significantly correlated with clinical and cognitive measures. Overall, these results demonstrated that multiscale functional connectivity patterns learned from a large-scale multi-site rsfMRI dataset were informative for characterizing the aging brain and the derived brain age gap was associated with cognitive and clinical measures.