Harmonization of multi-site MRS data with ComBat

Rethinking MRI as a measurement device through modular and portable pipelines

The premise of MRI as a reliable measurement device is limited by proprietary barriers and inconsistent implementations, which prevent the establishment of measurement uncertainties. As a result, biomedical studies that rely on these methods are plagued by systematic variance, undermining the perceived promise of quantitative imaging biomarkers (QIBs) and hindering their clinical translation. This review explores the added value of open-source measurement pipelines in minimizing variability sources that would otherwise remain unknown. First, we introduce a tiered benchmarking framework (from black-box to glass-box) that exposes how opacity at different workflow stages propagates measurement uncertainty. Second, we provide a concise glossary to promote consistent terminology for strategies that enhance reproducibility before acquisition or enable valid post-hoc pooling of QIBs. Building on this foundation, we present two illustrative measurement workflows that decouple workflow logic from the orchestration of computational processes in an MRI measurement pipeline, rooted in the core principles of modularity and portability. Designed as accessible entry points for implementation, these examples serve as practical guides, helping users adapt the frameworks to their specific needs and facilitating collaboration. Through critical evaluation of existing approaches, we discuss how standardized workflows can help identify outstanding challenges in translating glass-box frameworks into clinical scanner environments. Ultimately, achieving this goal will require coordinated efforts from QIB developers, regulators, industry partners, and clinicians alike.

Atypical hierarchical brain connectivity in autism: Insights from stepwise causal analysis using Liang information flow

Autism spectrum disorder (ASD) is associated with atypical brain connectivity, yet its hierarchical organization remains underexplored. In this study, we applied the Liang information flow method to analyze stepwise causal functional connectivity in ASD, offering a novel approach to understanding how different brain networks interact. Using resting-state fMRI data from ASD individuals and healthy controls, we observed significant alterations in both positive and negative causal connections across the ventral attention network, limbic network, frontal-parietal network, and default mode network. These disruptions were detected at multiple hierarchical levels, indicating changes in communication patterns across brain regions. By leveraging features of hierarchical causal connectivity, we achieved high classification accuracy between ASD and healthy individuals. Additionally, changes in network node degrees were found to correlate with ASD clinical symptoms, particularly social and communication behaviors. Our findings provide new insights into disrupted hierarchical brain connectivity in ASD and demonstrate the potential of this approach for distinguishing ASD from typical development.

Investigation of ComBat Harmonization on Radiomic and Deep Features from Multi-Center Abdominal MRI Data

ComBat harmonization has been developed to remove non-biological variations for data in multi-center research applying artificial intelligence (AI). We investigated the effectiveness of ComBat harmonization on radiomic and deep features extracted from large, multi-center abdominal MRI data. A retrospective study was conducted on T2-weighted (T2W) abdominal MRI data retrieved from individual patients with suspected or known chronic liver disease at three study sites. MRI data were acquired using systems from three manufacturers and two field strengths. Radiomic features and deep features were extracted using the PyRadiomics pipeline and a Swin Transformer. ComBat was used to harmonize radiomic and deep features across different manufacturers and field strengths. Student's t-test, ANOVA test, and Cohen's F score were applied to assess the difference in individual features before and after ComBat harmonization. Between two field strengths, 76.7%, 52.9%, and 26.7% of radiomic features, and 89.0%, 56.5%, and 0.1% of deep features from three manufacturers were significantly different. Among the three manufacturers, 90.1% and 75.0% of radiomic features and 89.3% and 84.1% of deep features from two field strengths were significantly different. After ComBat harmonization, there were no significant differences in radiomic and deep features among manufacturers or field strengths based on t-tests or ANOVA tests. Reduced Cohen's F scores were consistently observed after ComBat harmonization. ComBat harmonization effectively harmonizes radiomic and deep features by removing the non-biological variations due to system manufacturers and/or field strengths in large multi-center clinical abdominal MRI datasets.

Understanding Proton Magnetic Resonance Spectroscopy Neurochemical Changes Using Alzheimer's Disease Biofluid, PET, Postmortem Pathology Biomarkers, and APOE Genotype

In vivo proton (1H) magnetic resonance spectroscopy (MRS) is a powerful non-invasive method that can measure Alzheimer's disease (AD)-related neuropathological alterations at the molecular level. AD biomarkers include amyloid-beta (Aβ) plaques and hyperphosphorylated tau neurofibrillary tangles. These biomarkers can be detected via postmortem analysis but also in living individuals through positron emission tomography (PET) or biofluid biomarkers of Aβ and tau. This review offers an overview of biochemical abnormalities detected by 1H MRS within the biologically defined AD spectrum. It includes a summary of earlier studies that explored the association of 1H MRS metabolites with biofluid, PET, and postmortem AD biomarkers and examined how apolipoprotein e4 allele carrier status influences brain biochemistry. Studying these associations is crucial for understanding how AD pathology affects brain homeostasis throughout the AD continuum and may eventually facilitate the development of potential novel therapeutic approaches.

The Emergence of the Potential Therapeutic Targets: Ultrasound-Based Radiomics in the Prediction of Human Epidermal Growth Factor Receptor 2-Low Breast Cancer

RATIONALE AND OBJECTIVES

To evaluate whether ultrasound-based radiomics features can effectively predict HER2-low expression in patients with breast cancer (BC).

MATERIAL AND METHODS

Between January 2021 and June 2023, patients who received US scans with pathologically confirmed BC in this multicenter study were included. In total, 383 patients from institution 1 were comprised of training set, 233 patients from institution 2 were comprised of validation set and 149 patients from institution 3 were comprised of external validation set. Radiomics features were derived from conventional ultrasound (US) images. The minimum redundancy and maximum relevancy and the least absolute shrinkage and selector operation algorithm were used to generate an US-based radiomics score (RS). Multivariable logistic regression analysis was used to select variables associated with HER2 expressions. The diagnostic performance of the RS was evaluated through the area under the receiver operating characteristic curve (AUC).

RESULTS

In the training set, the RS yield an AUC of 0.81 (95%CI: 0.76-0.84) for differentiation HER2-zero from HER2-low and -positive cases, and performed well in validation set (AUC 0.84, 95%CI: 0.78-0.88) and external validation set (AUC 0.82, 95%CI: 0.73-0.90). In the subgroups analysis, the RS showed good performance in distinguishing HER2-zero from HER2 1 + , HER2 2 + and HER2-low tumors (AUC range, 0.79-0.87).

CONCLUSION

The RS based on conventional US is proven effective for predicting HER2-low expression in BC.

Development and Implementation of an Innovative Framework for Automated Radiomics Analysis in Neuroimaging

Radiomics represents an innovative approach to medical image analysis, enabling comprehensive quantitative evaluation of radiological images through advanced image processing and Machine or Deep Learning algorithms. This technique uncovers intricate data patterns beyond human visual detection. Traditionally, executing a radiomic pipeline involves multiple standardized phases across several software platforms. This could represent a limit that was overcome thanks to the development of the matRadiomics application. MatRadiomics, a freely available, IBSI-compliant tool, features its intuitive Graphical User Interface (GUI), facilitating the entire radiomics workflow from DICOM image importation to segmentation, feature selection and extraction, and Machine Learning model construction. In this project, an extension of matRadiomics was developed to support the importation of brain MRI images and segmentations in NIfTI format, thus extending its applicability to neuroimaging. This enhancement allows for the seamless execution of radiomic pipelines within matRadiomics, offering substantial advantages to the realm of neuroimaging.

Investigating the brain's neurochemical profile at midlife in relation to dementia risk factors

Changes in the brain's physiology in Alzheimer's disease are thought to occur early in the disease's trajectory. In this study our aim was to investigate the brain's neurochemical profile in a midlife cohort in relation to risk factors for future dementia using single voxel proton magnetic resonance spectroscopy. Participants in the multi-site PREVENT-Dementia study (age range 40-59 year old) underwent 3T magnetic resonance spectroscopy with the spectroscopy voxel placed in the posterior cingulate/precuneus region. Using LCModel, we quantified the absolute concentrations of myo-inositol, total N-acetylaspartate, total creatine, choline, glutathione and glutamate-glutamine for 406 participants (mean age 51.1; 65.3% female). Underlying partial volume effects were accounted for by applying a correction for the presence of cerebrospinal fluid in the magnetic resonance spectroscopy voxel. We investigated how metabolite concentrations related to apolipoprotein ɛ4 genotype, dementia family history, a risk score (Cardiovascular Risk Factors, Aging and Incidence of Dementia -CAIDE) for future dementia including non-modifiable and potentially-modifiable factors and dietary patterns (adherence to Mediterranean diet). Dementia family history was associated with decreased total N-acetylaspartate and no differences were found between apolipoprotein ɛ4 carriers and non-carriers. A higher Cardiovascular Risk Factors, Aging, and Incidence of Dementia score related to higher myo-inositol, choline, total creatine and glutamate-glutamine, an effect which was mainly driven by older age and a higher body mass index. Greater adherence to the Mediterranean diet was associated with lower choline, myo-inositol and total creatine; these effects did not survive correction for multiple comparisons. The observed associations suggest that at midlife the brain demonstrates subtle neurochemical changes in relation to both inherited and potentially modifiable risk factors for future dementia.

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.

Linewidth-related bias in modelled concentration estimates from GABA-edited 1H-MRS

J-difference-edited MRS is widely used to study GABA in the human brain. Editing for low-concentration target molecules (such as GABA) typically exhibits lower signal-to-noise ratio (SNR) than conventional non-edited MRS, varying with acquisition region, volume and duration. Moreover, spectral lineshape may be influenced by age-, pathology-, or brain-region-specific effects of metabolite T2, or by task-related blood-oxygen level dependent (BOLD) changes in functional MRS contexts. Differences in both SNR and lineshape may have systematic effects on concentration estimates derived from spectral modelling. The present study characterises the impact of lineshape and SNR on GABA+ estimates from different modelling algorithms: FSL-MRS, Gannet, LCModel, Osprey, spant and Tarquin. Publicly available multi-site GABA-edited data (222 healthy subjects from 20 sites; conventional MEGA-PRESS editing; TE = 68 ms) were pre-processed with a standardised pipeline, then filtered to apply controlled levels of Lorentzian and Gaussian linebroadening and SNR reduction. Increased Lorentzian linewidth was associated with a 2-5% decrease in GABA+ estimates per Hz, observed consistently (albeit to varying degrees) across datasets and most algorithms. Weaker, often opposing effects were observed for Gaussian linebroadening. Variations are likely caused by differing baseline parametrization and lineshape constraints between models. Effects of linewidth on other metabolites (e.g., Glx and tCr) varied, suggesting that a linewidth confound may persist after scaling to an internal reference. These findings indicate a potentially significant confound for studies where linewidth may differ systematically between groups or experimental conditions, e.g. due to T2 differences between brain regions, age, or pathology, or varying T2* due to BOLD-related changes. We conclude that linewidth effects need to be rigorously considered during experimental design and data processing, for example by incorporating linewidth into statistical analysis of modelling outcomes or development of appropriate lineshape matching algorithms.

Classifying Alzheimer's Disease Neuropathology Using Clinical and MRI Measurements

Background

Computer-aided machine learning models are being actively developed with clinically available biomarkers to diagnose Alzheimer's disease (AD) in living persons. Despite considerable work with cross-sectional in vivo data, many models lack validation against postmortem AD neuropathological data.

Objective

Train machine learning models to classify the presence or absence of autopsy-confirmed severe AD neuropathology using clinically available features.

Methods

AD neuropathological status are assessed at postmortem for participants from the National Alzheimer's Coordinating Center (NACC). Clinically available features are utilized, including demographics, Apolipoprotein E(APOE) genotype, and cortical thicknesses derived from ante-mortem MRI scans encompassing AD meta regions of interest (meta-ROI). Both logistic regression and random forest models are trained to identify linearly and nonlinearly separable features between participants with the presence (N = 91, age-at-MRI = 73.6±9.24, 38 women) or absence (N = 53, age-at-MRI = 68.93±19.69, 24 women) of severe AD neuropathology. The trained models are further validated in an external data set against in vivo amyloid biomarkers derived from PET imaging (amyloid-positive: N = 71, age-at-MRI = 74.17±6.37, 26 women; amyloid-negative: N = 73, age-at-MRI = 71.59±6.80, 41 women).

Results

Our models achieve a cross-validation accuracy of 84.03% in classifying the presence or absence of severe AD neuropathology, and an external-validation accuracy of 70.14% in classifying in vivo amyloid positivity status.

Conclusions

Our models show that clinically accessible features, including APOE genotype and cortical thinning encompassing AD meta-ROIs, are able to classify both postmortem confirmed AD neuropathological status and in vivo amyloid status with reasonable accuracies. These results suggest the potential utility of AD meta-ROIs in determining AD neuropathological status in living persons.

Multi-feature concatenation and multi-classifier stacking: An interpretable and generalizable machine learning method for MDD discrimination with rsfMRI

Major depressive disorder (MDD) is a serious and heterogeneous psychiatric disorder that needs accurate diagnosis. Resting-state functional MRI (rsfMRI), which captures multiple perspectives on brain structure, function, and connectivity, is increasingly applied in the diagnosis and pathological research of MDD. Different machine learning algorithms are then developed to exploit the rich information in rsfMRI and discriminate MDD patients from normal controls. Despite recent advances reported, the MDD discrimination accuracy has room for further improvement. The generalizability and interpretability of the discrimination method are not sufficiently addressed either. Here, we propose a machine learning method (MFMC) for MDD discrimination by concatenating multiple features and stacking multiple classifiers. MFMC is tested on the REST-meta-MDD data set that contains 2428 subjects collected from 25 different sites. MFMC yields 96.9% MDD discrimination accuracy, demonstrating a significant improvement over existing methods. In addition, the generalizability of MFMC is validated by the good performance when the training and testing subjects are from independent sites. The use of XGBoost as the meta classifier allows us to probe the decision process of MFMC. We identify 13 feature values related to 9 brain regions including the posterior cingulate gyrus, superior frontal gyrus orbital part, and angular gyrus, which contribute most to the classification and also demonstrate significant differences at the group level. The use of these 13 feature values alone can reach 87% of MFMC's full performance when taking all feature values. These features may serve as clinically useful diagnostic and prognostic biomarkers for MDD in the future.

Altered Effective Connectivity Among the Cerebellum and Cerebrum in Patients with Major Depressive Disorder Using Multisite Resting-State fMRI

Major depressive disorder (MDD) is a serious and widespread psychiatric disorder. Previous studies mainly focused on cerebrum functional connectivity, and the sample size was relatively small. However, functional connectivity is undirected. And, there is increasing evidence that the cerebellum is also involved in emotion and cognitive processing and makes outstanding contributions to the symptomology and pathology of depression. Therefore, we used a large sample size of resting-state functional magnetic resonance imaging (rs-fMRI) data to investigate the altered effective connectivity (EC) among the cerebellum and other cerebral cortex in patients with MDD. Here, from the perspective of data-driven analysis, we used two different atlases to divide the whole brain into different regions and analyzed the alterations of EC and EC networks in the MDD group compared with healthy controls group (HCs). The results showed that compared with HCs, there were significantly altered EC in the cerebellum-neocortex and cerebellum-basal ganglia circuits in MDD patients, which implied that the cerebellum may be a potential biomarker of depressive disorders. And, the alterations of EC brain networks in MDD patients may provide new insights into the pathophysiological mechanisms of depression.

Removal of site effects and enhancement of signal using dual projection independent component analysis for pooling multi-site MRI data

Combining magnetic resonance imaging (MRI) data from multi-site studies is a popular approach for constructing larger datasets to greatly enhance the reliability and reproducibility of neuroscience research. However, the scanner/site variability is a significant confound that complicates the interpretation of the results, so effective and complete removal of the scanner/site variability is necessary to realise the full advantages of pooling multi-site datasets. Independent component analysis (ICA) and general linear model (GLM) based harmonisation methods are the two primary methods used to eliminate scanner/site effects. Unfortunately, there are challenges with both ICA-based and GLM-based harmonisation methods to remove site effects completely when the signals of interest and scanner/site effects-related variables are correlated, which may occur in neuroscience studies. In this study, we propose an effective and powerful harmonisation strategy that implements dual projection (DP) theory based on ICA to remove the scanner/site effects more completely. This method can separate the signal effects correlated with site variables from the identified site effects for removal without losing signals of interest. Both simulations and vivo structural MRI datasets, including a dataset from Autism Brain Imaging Data Exchange II and a travelling subject dataset from the Strategic Research Program for Brain Sciences, were used to test the performance of a DP-based ICA harmonisation method. Results show that DP-based ICA harmonisation has superior performance for removing site effects and enhancing the sensitivity to detect signals of interest as compared with GLM-based and conventional ICA harmonisation methods.

Image data harmonization tools for the analysis of post-traumatic epilepsy development in preclinical multisite MRI studies

Preclinical MRI studies have been utilized for the discovery of biomarkers that predict post-traumatic epilepsy (PTE). However, these single site studies often lack statistical power due to limited and homogeneous datasets. Therefore, multisite studies, such as the Epilepsy Bioinformatics Study for Antiepileptogenic Therapy (EpiBioS4Rx), are developed to create large, heterogeneous datasets that can lead to more statistically significant results. EpiBioS4Rx collects preclinical data internationally across sites, including the United States, Finland, and Australia. However, in doing so, there are robust normalization and harmonization processes that are required to obtain statistically significant and generalizable results. This work describes the tools and procedures used to harmonize multisite, multimodal preclinical imaging data acquired by EpiBioS4Rx. There were four main harmonization processes that were utilized, including file format harmonization, naming convention harmonization, image coordinate system harmonization, and diffusion tensor imaging (DTI) metrics harmonization. By using Python tools and bash scripts, the file formats, file names, and image coordinate systems are harmonized across all the sites. To harmonize DTI metrics, values are estimated for each voxel in an image to generate a histogram representing the whole image. Then, the Quantitative Imaging Toolkit (QIT) modules are utilized to scale the mode to a value of one and depict the subsequent harmonized histogram. The standardization of file formats, naming conventions, coordinate systems, and DTI metrics are qualitatively assessed. The histograms of the DTI metrics were generated for all the individual rodents per site. For inter-site analysis, an average of the individual scans was calculated to create a histogram that represents each site. In order to ensure the analysis can be run at the level of individual animals, the sham and TBI cohort were analyzed separately, which depicted the same harmonization factor. The results demonstrate that these processes qualitatively standardize the file formats, naming conventions, coordinate systems, and DTI metrics of the data. This assists in the ability to share data across the study, as well as disseminate tools that can help other researchers to strengthen the statistical power of their studies and analyze data more cohesively.

The robustness of persistent homology of brain networks to data acquisition-related non-neural variability in resting state fMRI

There is increasing interest in investigating brain function based on functional connectivity networks (FCN) obtained from resting-state functional magnetic resonance imaging (fMRI). FCNs, typically obtained using measures of time series association such as Pearson's correlation, are sensitive to data acquisition parameters such as sampling period. This introduces non-neural variability in data pooled from different acquisition protocols and MRI scanners, negating the advantages of larger sample sizes in pooled data. To address this, we hypothesize that the topology or shape of brain networks must be preserved irrespective of how densely it is sampled, and metrics which capture this topology may be statistically similar across sampling periods, thereby alleviating this source of non-neural variability. Accordingly, we present an end-to-end pipeline that uses persistent homology (PH), a branch of topological data analysis, to demonstrate similarity across FCNs acquired at different temporal sampling periods. PH, as a technique, extracts topological features by capturing the network organization across all continuous threshold values, as opposed to graph theoretic methods, which fix a discrete network topology by thresholding the connectivity matrix. The extracted topological features are encoded in the form of persistent diagrams that can be compared against one another using the earth-moving metric, also popularly known as the Wasserstein distance. We extract topological features from three data cohorts, each acquired at different temporal sampling periods and demonstrate that these features are statistically the same, hence, empirically showing that PH may be robust to changes in data acquisition parameters such as sampling period.

Addressing multi-site functional MRI heterogeneity through dual-expert collaborative learning for brain disease identification

Several studies employ multi-site rs-fMRI data for major depressive disorder (MDD) identification, with a specific site as the to-be-analyzed target domain and other site(s) as the source domain. But they usually suffer from significant inter-site heterogeneity caused by the use of different scanners and/or scanning protocols and fail to build generalizable models that can well adapt to multiple target domains. In this article, we propose a dual-expert fMRI harmonization (DFH) framework for automated MDD diagnosis. Our DFH is designed to simultaneously exploit data from a single labeled source domain/site and two unlabeled target domains for mitigating data distribution differences across domains. Specifically, the DFH consists of a domain-generic student model and two domain-specific teacher/expert models that are jointly trained to perform knowledge distillation through a deep collaborative learning module. A student model with strong generalizability is finally derived, which can be well adapted to unseen target domains and analysis of other brain diseases. To the best of our knowledge, this is among the first attempts to investigate multi-target fMRI harmonization for MDD diagnosis. Comprehensive experiments on 836 subjects with rs-fMRI data from 3 different sites show the superiority of our method. The discriminative brain functional connectivities identified by our method could be regarded as potential biomarkers for fMRI-related MDD diagnosis.

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.

Harmonization of multi-site functional MRI data with dual-projection based ICA model

Modern neuroimaging studies frequently merge magnetic resonance imaging (MRI) data from multiple sites. A larger and more diverse group of participants can increase the statistical power, enhance the reliability and reproducibility of neuroimaging research, and obtain findings more representative of the general population. However, measurement biases caused by site differences in scanners represent a barrier when pooling data collected from different sites. The existence of site effects can mask biological effects and lead to spurious findings. We recently proposed a powerful denoising strategy that implements dual-projection (DP) theory based on ICA to remove site-related effects from pooled data, demonstrating the method for simulated and in vivo structural MRI data. This study investigates the use of our DP-based ICA denoising method for harmonizing functional MRI (fMRI) data collected from the Autism Brain Imaging Data Exchange II. After frequency-domain and regional homogeneity analyses, two modalities, including amplitude of low frequency fluctuation (ALFF) and regional homogeneity (ReHo), were used to validate our method. The results indicate that DP-based ICA denoising method removes unwanted site effects for both two fMRI modalities, with increases in the significance of the associations between non-imaging variables (age, sex, etc.) and fMRI measures. In conclusion, our DP method can be applied to fMRI data in multi-site studies, enabling more accurate and reliable neuroimaging research findings.

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.

Predictive value of 18F-FDG PET/CT-based radiomics model for neoadjuvant chemotherapy efficacy in breast cancer: a multi-scanner/center study with external validation

PURPOSE

To develop and validate the predictive value of an 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) model for breast cancer neoadjuvant chemotherapy (NAC) efficacy based on the tumor-to-liver ratio (TLR) radiomic features and multiple data pre-processing methods.

METHODS

One hundred and ninety-three breast cancer patients from multiple centers were retrospectively included in this study. According to the endpoint of NAC, we divided the patients into pathological complete remission (pCR) and non-pCR groups. All patients underwent 18F-FDG PET/CT imaging before NAC treatment, and CT and PET images volume of interest (VOI) segmentation by manual segmentation and semi-automated absolute threshold segmentation, respectively. Then, feature extraction of VOI was performed with the pyradiomics package. A total of 630 models were created based on the source of radiomic features, the elimination of the batch effect approach, and the discretization method. The differences in data pre-processing approaches were compared and analyzed to identify the best-performing model, which was further tested by the permutation test.

RESULTS

A variety of data pre-processing methods contributed in varying degrees to the improvement of model effects. Among them, TLR radiomic features and Combat and Limma methods that eliminate batch effects could enhance the model prediction overall, and data discretization could be used as a potential method that can further optimize the model. A total of seven excellent models were selected and then based on the AUC of each model in the four test sets and their standard deviations, we selected the optimal model. The optimal model predicted AUC between 0.7 and 0.77 for the four test groups, with p-values less than 0.05 for the permutation test.

CONCLUSION

It is necessary to enhance the predictive effect of the model by eliminating confounding factors through data pre-processing. The model developed in this way is effective in predicting the efficacy of NAC for breast cancer.

Brain metabolites measured with magnetic resonance spectroscopy in pediatric concussion and orthopedic injury: An Advancing Concussion Assessment in Pediatrics (A-CAP) study

Millions of children sustain a concussion annually. Concussion disrupts cellular signaling and neural pathways within the brain but the resulting metabolic disruptions are not well characterized. Magnetic resonance spectroscopy (MRS) can examine key brain metabolites (e.g., N-acetyl Aspartate (tNAA), glutamate (Glx), creatine (tCr), choline (tCho), and myo-Inositol (mI)) to better understand these disruptions. In this study, we used MRS to examine differences in brain metabolites between children and adolescents with concussion versus orthopedic injury. Children and adolescents with concussion (n = 361) or orthopedic injury (OI) (n = 184) aged 8 to 17 years were recruited from five emergency departments across Canada. MRS data were collected from the left dorsolateral prefrontal cortex (L-DLPFC) using point resolved spectroscopy (PRESS) at 3 T at a mean of 12 days post-injury (median 10 days post-injury, range 2-33 days). Univariate analyses for each metabolite found no statistically significant metabolite differences between groups. Within each analysis, several covariates were statistically significant. Follow-up analyses designed to account for possible confounding factors including age, site, scanner, vendor, time since injury, and tissue type (and interactions as appropriate) did not find any metabolite group differences. In the largest sample of pediatric concussion studied with MRS to date, we found no metabolite differences between concussion and OI groups in the L-DLPFC. We suggest that at 2 weeks post-injury in a general pediatric concussion population, brain metabolites in the L-DLPFC are not specifically affected by brain injury.

BrainGB: A Benchmark for Brain Network Analysis With Graph Neural Networks

Mapping the connectome of the human brain using structural or functional connectivity has become one of the most pervasive paradigms for neuroimaging analysis. Recently, Graph Neural Networks (GNNs) motivated from geometric deep learning have attracted broad interest due to their established power for modeling complex networked data. Despite their superior performance in many fields, there has not yet been a systematic study of how to design effective GNNs for brain network analysis. To bridge this gap, we present BrainGB, a benchmark for brain network analysis with GNNs. BrainGB standardizes the process by (1) summarizing brain network construction pipelines for both functional and structural neuroimaging modalities and (2) modularizing the implementation of GNN designs. We conduct extensive experiments on datasets across cohorts and modalities and recommend a set of general recipes for effective GNN designs on brain networks. To support open and reproducible research on GNN-based brain network analysis, we host the BrainGB website at https://braingb.us with models, tutorials, examples, as well as an out-of-box Python package. We hope that this work will provide useful empirical evidence and offer insights for future research in this novel and promising direction.

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

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

Comparison of different approaches to manage multi-site magnetic resonance spectroscopy clinical data analysis

Introduction

The effects caused by differences in data acquisition can be substantial and may impact data interpretation in multi-site/scanner studies using magnetic resonance spectroscopy (MRS). Given the increasing use of multi-site studies, a better understanding of how to account for different scanners is needed. Using data from a concussion population, we compare ComBat harmonization with different statistical methods in controlling for site, vendor, and scanner as covariates to determine how to best control for multi-site data.

Methods

The data for the current study included 545 MRS datasets to measure tNAA, tCr, tCho, Glx, and mI to study the pediatric concussion acquired across five sites, six scanners, and two different MRI vendors. For each metabolite, the site and vendor were accounted for in seven different models of general linear models (GLM) or mixed-effects models while testing for group differences between the concussion and orthopedic injury. Models 1 and 2 controlled for vendor and site. Models 3 and 4 controlled for scanner. Models 5 and 6 controlled for site applied to data harmonized by vendor using ComBat. Model 7 controlled for scanner applied to data harmonized by scanner using ComBat. All the models controlled for age and sex as covariates.

Results

Models 1 and 2, controlling for site and vendor, showed no significant group effect in any metabolites, but the vendor and site were significant factors in the GLM. Model 3, which included a scanner, showed a significant group effect for tNAA and tCho, and the scanner was a significant factor. Model 4, controlling for the scanner, did not show a group effect in the mixed model. The data harmonized by the vendor using ComBat (Models 5 and 6) had no significant group effect in both the GLM and mixed models. Lastly, the data harmonized by the scanner using ComBat (Model 7) showed no significant group effect. The individual site data suggest there were no group differences.

Conclusion

Using data from a large clinical concussion population, different analysis techniques to control for site, vendor, and scanner in MRS data yielded different results. The findings support the use of ComBat harmonization for clinical MRS data, as it removes the site and vendor effects.

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.

Ultra-high field neuroimaging in psychosis: A narrative review

Schizophrenia and related psychoses are complex neuropsychiatric diseases representing dysconnectivity across multiple scales, through the micro (cellular), meso (brain network), manifest (behavioral), and social (interpersonal) levels. In vivo human neuroimaging, particularly at ultra-high field (UHF), offers unprecedented opportunity to examine multiscale dysconnectivity in psychosis. In this review, we provide an overview of the literature to date on UHF in psychosis, focusing on microscale findings from magnetic resonance spectroscopy (MRS), mesoscale studies on structural and functional magnetic resonance imaging (fMRI), and multiscale studies assessing multiple neuroimaging modalities and relating UHF findings to behavior. We highlight key insights and considerations from multiscale and longitudinal studies and provide recommendations for future research on UHF neuroimaging in psychosis.