An incomplete field-of-view (FOV) in diffusion magnetic resonance imaging (dMRI) can severely hinder the volumetric and bundle analyses of whole-brain white matter connectivity. Although existing works have investigated imputing the missing regions using deep generative models, it remains unclear how to specifically utilize additional information from paired multi-modality data and whether this can enhance the imputation quality and be useful for downstream tractography. To fill this gap, we propose a novel framework for imputing dMRI scans in the incomplete part of the FOV by integrating the learned diffusion features in the acquired part of the FOV to the complete brain anatomical structure. We hypothesize that by this design the proposed framework can enhance the imputation performance of the dMRI scans and therefore be useful for repairing whole-brain tractography in corrupted dMRI scans with incomplete FOV. We tested our framework on two cohorts from different sites with a total of 96 subjects and compared it with a baseline imputation method that treats the information from T1w and dMRI scans equally. The proposed framework achieved significant improvements in imputation performance, as demonstrated by angular correlation coefficient (p < 1E-5), and in downstream tractography accuracy, as demonstrated by Dice score (p < 0.01). Results suggest that the proposed framework improved imputation performance in dMRI scans by specifically utilizing additional information from paired multi-modality data, compared with the baseline method. The imputation achieved by the proposed framework enhances whole brain tractography, and therefore reduces the uncertainty when analyzing bundles associated with neurodegenerative.

Purpose:Connectome network metrics are commonly regarded as fundamental properties of the brain, and their alterations have been implicated in the development of Alzheimer's disease, multiple sclerosis, and traumatic brain injury. However, these metrics are actually estimated properties through a multistage propagation from local voxel diffusion estimations, regional tractography, and region of interest mapping. These estimation processes are significantly influenced by choices specific to imaging protocols and software, producing site-wise effects. Approach:Recent advances in disentanglement techniques offer opportunities to learn representational spaces that separate factors that cause domain shifts from intrinsic biological factors. Although these techniques have been applied in unsupervised brain anomaly detection and image-level features, their application to the unique manifold structures of connectome adjacency matrices remains unexplored. Here, we explore the conditional variational autoencoder structure for generating site-invariant representations of the connectome, allowing the harmonization of brain network measures. Results:Focusing on the context of aging, we conducted a study involving 823 patients across two sites. This approach effectively segregates site-specific influences from biological features, aligns network measures across different domains (Cohen's D < 0.2 and Mann-Whitney U - test < 0.05 ), and maintains associations with age ( 2.71 × 10 - 02 ± 2.86 × 10 - 03 error in years) and sex ( 0.92 ± 0.02 accuracy). Conclusions:Our findings demonstrate that using latent representations significantly harmonizes network measures and provides robust metrics for multi-site brain network analysis.

Curating, processing, and combining large-scale medical imaging datasets from national studies is a non-trivial task due to the intense computation and data throughput required, variability of acquired data, and associated financial overhead. Existing platforms or tools for large-scale data curation, processing, and storage have difficulty achieving a viable cost-to-scale ratio of computation speed for research purposes, either being too slow or too expensive. Additionally, management and consistency of processing large data in a team-driven manner is a non-trivial task. We design a BIDS-compliant method for an efficient and robust data processing pipeline of large-scale diffusion-weighted and T1-weighted MRI data compatible with low-cost, high-efficiency computing systems. Our method accomplishes automated querying of data available for processing and process running in a consistent and reproducible manner that has long-term stability, while using heterogenous low-cost computational resources and storage systems for efficient processing and data transfer. We demonstrate how our organizational structure permits efficiency in a semi-automated data processing pipeline and show how our method is comparable in processing time to cloud-based computation while being almost 20 times more cost-effective. Our design allows for fast data throughput speeds and low latency to reduce the time for data transfer between storage servers and computation servers, achieving an average of 0.60 Gb/s compared to 0.33 Gb/s for using cloud-based processing methods. The design of our workflow engine permits quick process running while maintaining flexibility to adapt to newly acquired data.

To date, there has been no comprehensive study characterizing the effect of diffusion-weighted magnetic resonance imaging voxel resolution on the resulting connectome for high resolution subject data. Similarity in results improved with higher resolution, even after initial down-sampling. To ensure robust tractography and connectomes, resample data to 1 mm isotropic resolution.

ABSTRACT Head motion during diffusion magnetic resonance imaging (MRI) scans can cause numerous artifacts and biases subsequent quantification. However, a thorough characterization of motion across multiple scans, cohorts, and consortiums has not been performed. To address this, we designed a study with three aims. First, we aimed to characterize subject motion across several large cohorts, utilizing 13 cohorts comprised of 16,995 imaging sessions (age 0.1–100 years, mean age = 63 years; 7220 females; 3175 cognitively impaired adults; 471 developmentally delayed children) to describe the magnitude and directions of subject movement. Second, we aimed to investigate whether state‐of‐the‐art diffusion preprocessing pipelines mitigate biases in quantitative measures of microstructure and connectivity by taking advantage of datasets with scan‐rescan acquisitions and ask whether there are detectable differences between the same subjects when scans and rescans have differing levels of motion. Third, we aimed to investigate whether there are structural connectivity differences between movers and non‐movers. We found that (1) subjects typically move 1–2 mm/min with most motion as translation in the anterior–posterior direction and rotation around the right–left axis; (2) Modern preprocessing pipelines can effectively mitigate motion to the point where biases are not detectable with current analysis techniques; and (3) There are no apparent differences in microstructure or macrostructural connections in participants who exhibit high motion versus those that exhibit low motion. Overall, characterizing motion magnitude and directions, as well as motion correlates, informs and improves motion mitigation strategies and image processing pipelines.

Thorough quality control (QC) can be time consuming when working with large-scale medical imaging datasets, yet necessary, as poor-quality data can lead to erroneous conclusions or poorly trained machine learning models. Most efforts to reduce data QC time rely on quantitative outlier detection, which cannot capture every instance of algorithm failure. Thus, there is a need to visually inspect every output of data processing pipelines in a scalable manner. We design a QC pipeline that allows for low time cost and effort across a team setting for a large database of diffusion-weighted and structural magnetic resonance images. Our proposed method satisfies the following design criteria: 1.) a consistent way to perform and manage quality control across a team of researchers, 2.) quick visualization of preprocessed data that minimizes the effort and time spent on the QC process without compromising the condition/caliber of the QC, and 3.) a way to aggregate QC results across pipelines and datasets that can be easily shared. In addition to meeting these design criteria, we also provide a comparison experiment of our method to an automated QC method for a T1-weighted dataset of N=1560 images and an inter-rater variability experiment for several processing pipelines. The experiments show mostly high agreement among raters and slight differences with the automated QC method. While researchers must spend time on robust visual QC of data, there are mechanisms by which the process can be streamlined and efficient.

While typical qualitative T1-weighted magnetic resonance images reflect scanner and protocol differences, quantitative T1 mapping aims to measure T1 independent of these effects. Changes in T1 in the brain reflect chemical and physical changes in brain tissue, such as the demyelination of axons in multiple sclerosis. Magnetization-prepared two rapid acquisition gradient echo (MP2RAGE) is an acquisition protocol that allows for efficient T1 mapping with a much lower scan time per slice compared to multi-TI inversion recovery (IR) protocols. We collect and register B1-corrected MP2RAGE acquisitions with an additional inversion time (MP3RAGE) alongside multi-TI selective inversion recovery acquisitions for four subjects and find a tissue-dependent bias between the derived T1 values. We train a patch-based ResNet-18 to calibrate the MP3RAGE T1 values to the multi-TI IR T1 values, incorporating the standard deviation of T1 calculated from a Monte Carlo simulation as an additional channel. Across four folds, the error between the MP2RAGE and T1 maps varies substantially (RMSE in white matter: 0.30 +/- 0.01 seconds, subcortical gray matter: 0.26 +/- 0.02 seconds, cortical gray matter: 0.36 +/- 0.02 seconds). Our network reduces the RMSE significantly (RMSE in white matter: 0.11 +/- 0.02 seconds, subcortical gray matter: 0.10 +/- 0.02 seconds, cortical gray matter: 0.17 +/- 0.03 seconds). Adding the standard deviation channel does not substantially change the RMSE. Using limited paired training data from both sequences, we can reduce the error between quantitative imaging methods and calibrate to one of the protocols with a neural network.

Defacing is often applied to head magnetic resonance image (MRI) datasets prior to public release to address privacy concerns. The alteration of facial and nearby voxels has provoked discussions about the true capability of these techniques to ensure privacy as well as their impact on downstream tasks. With advancements in deep generative models, the extent to which defacing can protect privacy is uncertain. Additionally, while the altered voxels are known to contain valuable anatomical information, their potential to support research beyond the anatomical regions directly affected by defacing remains uncertain. To evaluate these considerations, we develop a refacing pipeline that recovers faces in defaced head MRIs using cascaded diffusion probabilistic models (DPMs). The DPMs are trained on images from 180 subjects and tested on images from 484 unseen subjects, 469 of whom are from a different dataset. To assess whether the altered voxels in defacing contain universally useful information, we also predict computed tomography (CT)-derived skeletal muscle radiodensity from facial voxels in both defaced and original MRIs. The results show that DPMs can generate high-fidelity faces that resemble the original faces from defaced images, with surface distances to the original faces significantly smaller than those of a population average face (p < 0.05). This performance also generalizes well to previously unseen datasets. For skeletal muscle radiodensity predictions, using defaced images results in significantly weaker Spearman's rank correlation coefficients compared to using original images (p < 10-4). For shin muscle, the correlation is statistically significant (p < 0.05) when using original images but not statistically significant (p > 0.05) when any defacing method is applied, suggesting that defacing might not only fail to protect privacy but also eliminate valuable information.

Estimated brain age from magnetic resonance image (MRI) and its deviation from chronological age can provide early insights into potential neurodegenerative diseases, supporting early detection and implementation of prevention strategies to slow disease progression and onset. Diffusion MRI (dMRI), a widely used modality for brain age estimation, presents an opportunity to build an earlier biomarker for neurodegenerative disease prediction because it captures subtle microstructural changes that precede more perceptible macrostructural changes. However, the coexistence of macro- and micro-structural information in dMRI raises the question of whether current dMRI-based brain age estimation models are leveraging the intended microstructural information or if they inadvertently rely on the macrostructural information. To develop a microstructure-specific brain age, we propose a method for brain age identification from dMRI that mitigates the model's use of macrostructural information by non-rigidly registering all images to a standard template. Imaging data from 13,398 participants across 12 datasets were used for the training and evaluation. We compare our brain age models, trained with and without macrostructural information mitigated, with an architecturally similar T1-weighted (T1w) MRI-based brain age model and two recent, popular, openly available T1w MRI-based brain age models that primarily use macrostructural information. We observe difference between our dMRI-based brain age and T1w MRI-based brain age across stages of neurodegeneration, with dMRI-based brain age being older than T1w MRI-based brain age in participants transitioning from cognitively normal (CN) to mild cognitive impairment (MCI) (p-value = 0.023), but younger in participants already diagnosed with Alzheimer's disease (AD) (p-value < 0.001). Classifiers using T1w MRI-based brain ages generally outperform those using dMRI-based brain age in classifying CN versus AD participants. Conversely, dMRI-based brain age may offer advantages over T1w MRI-based brain age in predicting the transition from CN to MCI.

Paired inspiratory-expiratory computed tomography (CT) scans enable quantification of gas trapping alterations due to small airway disease and emphysema through the motion of the lung tissue for people with chronic obstructive pulmonary disease (COPD). Deformable image registration of these paired CT scans is often used to assess the regional volumetric changes in the lung. However, variations in reconstruction protocols, particularly the reconstruction kernels between paired inspiratory-expiratory scans are often overlooked, and these variations introduce errors during quantitative image analysis. In this work, we propose a two-stage pipeline to harmonize reconstruction kernels between paired inspiratory-expiratory scans and perform deformable image registration for data acquired from the COPDGene study. We use a cycle generative adversarial network (GAN) for image synthesis to harmonize inspiratory scans reconstructed with a hard kernel (BONE) to match expiratory scans reconstructed with a soft kernel (STANDARD). We then perform deformable image registration to register the expiratory scans to the inspiratory scans. We validated harmonization by measuring emphysema using a publicly available segmentation algorithm, both before and after harmonization. Our results show that harmonization significantly reduces inconsistencies in emphysema measurement, decreasing the median emphysema scores from 10.479% to 3.039% with a reference median score of 1.305% from the STANDARD kernel as a harmonization target. We validate the registration accuracy by observing the Dice overlap between emphysema regions on the inspiratory, expiratory and deformed images. The Dice coefficient between the fixed inspiratory emphysema masks and deformably registered emphysema masks increases across different stages of registration with statistical significance ( p<0.001). Additionally, we show that deformable registration is robust to kernel variation.

PURPOSE:Diffusion weighted MRI (dMRI) and its models of neural structure provide insight into human brain organization and variations in white matter. A recent study by McMaster, et al. showed that complex graph measures of the connectome, the graphical representation of a tractogram, vary with spatial sampling changes, but biases introduced by anisotropic voxels in the process have not been well characterized. This study uses microstructural measures (fractional anisotropy and mean diffusivity) and white matter bundle properties (bundle volume, length, and surface area) to further understand the effect of anisotropic voxels on microstructure and tractography. METHODS:The statistical significance of the selected measures derived from dMRI data were assessed by comparing three white matter bundles at different spatial resolutions with 44 subjects from the Human Connectome Project - Young Adult dataset scan/rescan data using the Wilcoxon Signed-Rank test. The original isotropic resolution (1.25 mm isotropic) was explored with 6 anisotropic resolutions with 0.25 mm incremental steps in the z dimension. Then, all generated resolutions were upsampled to 1.25 mm isotropic and 1 mm isotropic. RESULTS:There were statistically significant differences between at least one microstructural and one bundle measure at every resolution (p≤0.05, corrected for multiple comparisons). Cohen's d coefficient evaluated the effect size of anisotropic voxels on microstructure and tractography. CONCLUSION:Fractional anisotropy and mean diffusivity cannot be recovered with basic up-sampling from low quality data with gold-standard data with the methods selected for this study. However, the bundle measures across our selected regions of interest become more repeatable when voxels are resampled to 1 mm isotropic.

Diffusion MRI (dMRI) streamline tractography, the gold standard for in vivo estimation of brain white matter (WM) pathways, has long been considered indicative of macroscopic relationships with WM microstructure. However, recent advances in tractography demonstrated that convolutional recurrent neural networks (CoRNN) trained with a teacher-student framework have the ability to learn and propagate streamlines directly from T1 and anatomical contexts. Training for this network has previously relied on high-resolution dMRI. In this paper, we generalize the training mechanism to traditional clinical resolution data, which allows generalizability across sensitive and susceptible study populations. We train CoRNN on a small subset of the Baltimore Longitudinal Study of Aging (BLSA), which better resembles clinical protocols. Then, we define a metric, termed the epsilon ball seeding method, to compare T1 tractography and traditional diffusion tractography at the streamline level. Under this metric, T1 tractography generated by CoRNN reproduces diffusion tractography with approximately two millimeters of error.

2D single-slice abdominal computed tomography (CT) enables the assessment of body habitus and organ health with low radiation exposure. However, single-slice data necessitates the use of 2D networks for segmentation, but these networks often struggle to capture contextual information effectively. Consequently, even when trained on identical datasets, 3D networks typically achieve superior segmentation results. In this work, we propose a novel 3D-to-2D distillation framework, leveraging pre-trained 3D models to enhance 2D single-slice segmentation. Specifically, we extract the prediction distribution centroid from the 3D representations, to guide the 2D student by learning intra- and inter-class correlation. Unlike traditional knowledge distillation methods that require the same data input, our approach employs unpaired 3D CT scans with any contrast to guide the 2D student model. Experiments conducted on 707 subjects from the single-slice Baltimore Longitudinal Study of Aging (BLSA) dataset demonstrate that state-of-the-art 2D multi-organ segmentation methods can benefit from the 3D teacher model, achieving enhanced performance in single-slice multi-organ segmentation. Notably, our approach demonstrates considerable efficacy in low-data regimes, outperforming the model trained with all available training subjects even when utilizing only 200 training subjects. Thus, this work underscores the potential to alleviate manual annotation burdens.

Imaging findings inconsistent with those expected at specific chronological age ranges may serve as early indicators of neurological disorders and increased mortality risk. Estimation of chronological age, and deviations from expected results, from structural magnetic resonance imaging (MRI) data has become an important proxy task for developing biomarkers that are sensitive to such deviations. Complementary to structural analysis, diffusion tensor imaging (DTI) has proven effective in identifying age-related microstructural changes within the brain white matter, thereby presenting itself as a promising additional modality for brain age prediction. Although early studies have sought to harness DTI's advantages for age estimation, there is no evidence that the success of this prediction is owed to the unique microstructural and diffusivity features that DTI provides, rather than the macrostructural features that are also available in DTI data. Therefore, we seek to develop white-matter-specific age estimation to capture deviations from normal white matter aging. Specifically, we deliberately disregard the macrostructural information when predicting age from DTI scalar images, using two distinct methods. The first method relies on extracting only microstructural features from regions of interest (ROIs). The second applies 3D residual neural networks (ResNets) to learn features directly from the images, which are non-linearly registered and warped to a template to minimize macrostructural variations. When tested on unseen data, the first method yields mean absolute error (MAE) of 6.11 ± 0.19 years for cognitively normal participants and MAE of 6.62 ± 0.30 years for cognitively impaired participants, while the second method achieves MAE of 4.69 ± 0.23 years for cognitively normal participants and MAE of 4.96 ± 0.28 years for cognitively impaired participants. We find that the ResNet model captures subtler, non-macrostructural features for brain age prediction.

Purpose:As large analyses merge data across sites, a deeper understanding of variance in statistical assessment across the sources of data becomes critical for valid analyses. Diffusion tensor imaging (DTI) exhibits spatially varying and correlated noise, so care must be taken with distributional assumptions. Here, we characterize the role of physiology, subject compliance, and the interaction of the subject with the scanner in the understanding of DTI variability, as modeled in the spatial variance of derived metrics in homogeneous regions. Approach:We analyze DTI data from 1035 subjects in the Baltimore Longitudinal Study of Aging, with ages ranging from 22.4 to 103 years old. For each subject, up to 12 longitudinal sessions were conducted. We assess the variance of DTI scalars within regions of interest (ROIs) defined by four segmentation methods and investigate the relationships between the variance and covariates, including baseline age, time from the baseline (referred to as "interval"), motion, sex, and whether it is the first scan or the second scan in the session. Results:Covariate effects are heterogeneous and bilaterally symmetric across ROIs. Inter-session interval is positively related ( p ≪ 0.001 ) to FA variance in the cuneus and occipital gyrus, but negatively ( p ≪ 0.001 ) in the caudate nucleus. Males show significantly ( p ≪ 0.001 ) higher FA variance in the right putamen, thalamus, body of the corpus callosum, and cingulate gyrus. In 62 out of 176 ROIs defined by the Eve type-1 atlas, an increase in motion is associated ( p < 0.05 ) with a decrease in FA variance. Head motion increases during the rescan of DTI ( Δ μ = 0.045 mm per volume). Conclusions:The effects of each covariate on DTI variance and their relationships across ROIs are complex. Ultimately, we encourage researchers to include estimates of variance when sharing data and consider models of heteroscedasticity in analysis. This work provides a foundation for study planning to account for regional variations in metric variance.

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) βAGE, the linear regression coefficient of the relationship between FA and age; (ii) γ^sf*, the ComBat-estimated site-shift; and (iii) δ^sf*, 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 βAGE 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.

Over the last few decades, diffusion MRI (dMRI) streamline tractography has emerged as the dominant method for in vivo estimation of white matter (WM) pathways in the brain. One key limitation to this technique is that modern tractography implementations require high angular resolution diffusion imaging (HARDI). However, HARDI can be difficult to collect clinically, limiting the reach of tractography analyses to research cohorts and thus limiting many WM investigations to certain populations and pathologies. As such, a clinically viable tractography solution applicable to wider patient populations scanned as a part of routine care would be of key significance in broadening WM analyses to underfunded or rarer diseases and to the clinical setting. Such a solution would require the ability to perform arbitrary tractography analyses, use only clinical imaging for input, and be open source and widely accessible and implementable. Thus, here we evaluate our recently developed, containerized and open-source, T1-weighted (T1w) MRI-based deep learning model for streamline propagation. We empirically assess its performance against traditional dMRI-based and established atlas-based approaches in a healthy young population, an aging one, and in those with epilepsy, depression, and brain cancer. In the healthy young population, we find slightly increased error compared to traditional tractography with the deep learning model that falls within the bounds attributable to dMRI variability and is considerably less than the atlas-based approach. Further, seeking to replicate previously published dMRI tractography effects in the remaining cohorts as an initial assessment of clinical viability, we find this model successfully does so in some key cases—particularly in applications that rely on long-range streamlines including those not captured by the atlas-based approach—but importantly not all. These results suggest a deep learning-based approach to tractography with T1w MRI demonstrates promise within the limitations of our definition of clinical viability and especially over atlas-based approaches but requires refinement and more robust consideration of out-of-distribution effects prior to widespread clinical use. We also find these results raise additional questions regarding the differences in image content between dMRI and T1w MRI and their relationship to tractography. Further investigation of these questions will improve the field’s understanding of which features of the brain influence measured tractography effects.

T1-weighted (T1w) MRI has low frequency intensity artifacts due to magnetic field inhomogeneities. Removal of these biases in T1w MRI images is a critical preprocessing step to ensure spatially consistent image interpretation. N4ITK bias field correction, the current state-of-the-art, is implemented in such a way that makes it difficult to port between different pipelines and workflows, thus making it hard to reimplement and reproduce results across local, cloud, and edge platforms. Moreover, N4ITK is opaque to optimization before and after its application, meaning that methodological development must work around the inhomogeneity correction step. Given the importance of bias fields correction in structural preprocessing and flexible implementation, we pursue a deep learning approximation / reinterpretation of the N4ITK bias fields correction to create a method which is portable, flexible, and fully differentiable. In this paper, we trained a deep learning network “DeepN4” on eight independent cohorts from 72 different scanners and age ranges with N4ITK-corrected T1w MRI and bias field for supervision in log space. We found that we can closely approximate N4ITK bias fields correction with naïve networks. We evaluate the peak signal to noise ratio (PSNR) in test dataset against the N4ITK corrected images. The median PSNR of corrected images between N4ITK and DeepN4 was 47.96 dB. In addition, we assess the DeepN4 model on eight additional external datasets and show the generalizability of the approach. This study establishes that incompatible N4ITK preprocessing steps can be closely approximated by naïve deep neural networks, facilitating more flexibility. All code and models are released at https://github.com/MASILab/DeepN4.

Connectivity matrices derived from diffusion MRI (dMRI) provide an interpretable and generalizable way of understanding the human brain connectome. However, dMRI suffers from inter-site and between-scanner variation, which impedes analysis across datasets to improve robustness and reproducibility of results. To evaluate different harmonization approaches on connectivity matrices, we compared graph measures derived from these matrices before and after applying three harmonization techniques: mean shift, ComBat, and CycleGAN. The sample comprises 168 age-matched, sex-matched normal subjects from two studies: the Vanderbilt Memory and Aging Project (VMAP) and the Biomarkers of Cognitive Decline Among Normal Individuals (BIOCARD). First, we plotted the graph measures and used coefficient of variation (CoV) and the Mann-Whitney U test to evaluate different methods' effectiveness in removing site effects on the matrices and the derived graph measures. ComBat effectively eliminated site effects for global efficiency and modularity and outperformed the other two methods. However, all methods exhibited poor performance when harmonizing average betweenness centrality. Second, we tested whether our harmonization methods preserved correlations between age and graph measures. All methods except for CycleGAN in one direction improved correlations between age and global efficiency and between age and modularity from insignificant to significant with p-values less than 0.05.

Data harmonization is necessary for removing confounding effects in multi-site diffusion image analysis. One such harmonization method, LinearRISH, scales rotationally invariant spherical harmonic (RISH) features from one site (“target”) to the second (“reference”) to reduce confounding scanner effects. However, reference and target site designations are not arbitrary and resultant diffusion metrics (fractional anisotropy, mean diffusivity) are biased by this choice. In this work we propose MidRISH: rather than scaling reference RISH features to target RISH features, we project both sites to a mid-space. We validate MidRISH with the following experiments: harmonizing scanner differences from 37 matched patients free of cognitive impairment, and harmonizing acquisition and study differences on 117 matched patients free of cognitive impairment. We find that MidRISH reduces bias of reference selection while preserving harmonization efficacy of LinearRISH. Users should be cautious when performing LinearRISH harmonization. To select a reference site is to choose diffusion metric effect-size. Our proposed method eliminates the bias-inducing site selection step.