Multi-parametric neuroimaging reproducibility: a 3-T resource study

Cross-Modality Reference and Feature Mutual-Projection for 3D Brain MRI Image Super-Resolution

High-resolution (HR) magnetic resonance imaging (MRI) can reveal rich anatomical structures for clinical diagnoses. However, due to hardware and signal-to-noise ratio limitations, MRI images are often collected with low resolution (LR) which is not conducive to diagnosing and analyzing clinical diseases. Recently, deep learning super-resolution (SR) methods have demonstrated great potential in enhancing the resolution of MRI images; however, most of them did not take the cross-modality and internal priors of MR seriously, which hinders the SR performance. In this paper, we propose a cross-modality reference and feature mutual-projection (CRFM) method to enhance the spatial resolution of brain MRI images. Specifically, we feed the gradients of HR MRI images from referenced imaging modality into the SR network to transform true clear textures to LR feature maps. Meanwhile, we design a plug-in feature mutual-projection (FMP) method to capture the cross-scale dependency and cross-modality similarity details of MRI images. Finally, we fuse all feature maps with parallel attentions to produce and refine the HR features adaptively. Extensive experiments on MRI images in the image domain and k-space show that our CRFM method outperforms existing state-of-the-art MRI SR methods.

Normative Neuroimaging Library: Designing a Comprehensive and Demographically Diverse Dataset of Healthy Controls to Support Traumatic Brain Injury Diagnostic and Therapeutic Development

The past decade has seen impressive advances in neuroimaging, moving from qualitative to quantitative outputs. Available techniques now allow for the inference of microscopic changes occurring in white and gray matter, along with alterations in physiology and function. These existing and emerging techniques hold the potential of providing unprecedented capabilities in achieving a diagnosis and predicting outcomes for traumatic brain injury (TBI) and a variety of other neurological diseases. To see this promise move from the research lab into clinical care, an understanding is needed of what normal data look like for all age ranges, sex, and other demographic and socioeconomic categories. Clinicians can only use the results of imaging scans to support their decision-making if they know how the results for their patient compare with a normative standard. This potential for utilizing magnetic resonance imaging (MRI) in TBI diagnosis motivated the American College of Radiology and Cohen Veterans Bioscience to create a reference database of healthy individuals with neuroimaging, demographic data, and characterization of psychological functioning and neurocognitive data that will serve as a normative resource for clinicians and researchers for development of diagnostics and therapeutics for TBI and other brain disorders. The goal of this article is to introduce the large, well-curated Normative Neuroimaging Library (NNL) to the research community. NNL consists of data collected from ∼1900 healthy participants. The highlights of NNL are (1) data are collected across a diverse population, including civilians, veterans, and active-duty service members with an age range (18-64 years) not well represented in existing datasets; (2) comprehensive structural and functional neuroimaging acquisition with state-of-the-art sequences (including structural, diffusion, and functional MRI; raw scanner data are preserved, allowing higher quality data to be derived in the future; standardized imaging acquisition protocols across sites reflect sequences and parameters often recommended for use with various neurological and psychiatric conditions, including TBI, post-traumatic stress disorder, stroke, neurodegenerative disorders, and neoplastic disease); and (3) the collection of comprehensive demographic details, medical history, and a broad structured clinical assessment, including cognition and psychological scales, relevant to multiple neurological conditions with functional sequelae. Thus, NNL provides a demographically diverse population of healthy individuals who can serve as a comparison group for brain injury study and clinical samples, providing a strong foundation for precision medicine. Use cases include the creation of imaging-derived phenotypes (IDPs), derivation of reference ranges of imaging measures, and use of IDPs as training samples for artificial intelligence-based biomarker development and for normative modeling to help identify injury-induced changes as outliers for precision diagnosis and targeted therapeutic development. On its release, NNL is poised to support the use of advanced imaging in clinician decision support tools, the validation of imaging biomarkers, and the investigation of brain-behavior anomalies, moving the field toward precision medicine.

A bimodal taxonomy of adult human brain sulcal morphology related to timing of fetal sulcation and trans-sulcal gene expression gradients

We developed a computational pipeline (now provided as a resource) for measuring morphological similarity between cortical surface sulci to construct a sulcal phenotype network (SPN) from each magnetic resonance imaging (MRI) scan in an adult cohort (n = 34,725; 45-82 years). Networks estimated from pairwise similarities of 40 sulci on 5 morphological metrics comprised two clusters of sulci, represented also by the bimodal distribution of sulci on a linear-to-complex dimension. Linear sulci were more heritable and typically located in unimodal cortex, and complex sulci were less heritable and typically located in heteromodal cortex. Aligning these results with an independent fetal brain MRI cohort (n = 228; 21-36 gestational weeks), we found that linear sulci formed earlier, and the earliest and latest-forming sulci had the least between-adult variation. Using high-resolution maps of cortical gene expression, we found that linear sulcation is mechanistically underpinned by trans-sulcal gene expression gradients enriched for developmental processes.

On the optimization of 3D inflow-based vascular-space-occupancy (iVASO) MRI for the quantification of arterial cerebral blood volume (CBVa)

PURPOSE

The inflow-based vascular-space-occupancy (iVASO) MRI was originally developed in a single-slice mode to measure arterial cerebral blood volume (CBVa). When vascular crushers are applied in iVASO, the signals can be sensitized predominantly to small pial arteries and arterioles. The purpose of this study is to perform a systematic optimization and evaluation of a 3D iVASO sequence on both 3 T and 7 T for the quantification of CBVa values in the human brain.

METHODS

Three sets of experiments were performed in three separate cohorts. (1) 3D iVASO MRI protocols were compared to single-slice iVASO, and the reproducibility of whole-brain 3D iVASO MRI was evaluated. (2) The effects from different vascular crushers in iVASO were assessed. (3) 3D iVASO MRI results were evaluated in arterial and venous blood vessels identified using ultrasmall-superparamagnetic-iron-oxides-enhanced MRI to validate its arterial origin.

RESULTS

3D iVASO scans showed signal-to-noise ratio (SNR) and CBVa measures consistent with single-slice iVASO with reasonable intrasubject reproducibility. Among the iVASO scans performed with different vascular crushers, the whole-brain 3D iVASO scan with a motion-sensitized-driven-equilibrium preparation with two binomial refocusing pulses and an effective TE of 50 ms showed the best suppression of macrovascular signals, with a relatively low specific absorption rate. When no vascular crusher was applied, the CBVa maps from 3D iVASO scans showed large CBVa values in arterial vessels but well-suppressed signals in venous vessels.

CONCLUSION

A whole-brain 3D iVASO MRI scan was optimized for CBVa measurement in the human brain. When only microvascular signals are desired, a motion-sensitized-driven-equilibrium-based vascular crusher with binomial refocusing pulses can be applied in 3D iVASO.

Sliding windows analysis can undo the effects of preprocessing when applied to fMRI data

Resting-state fMRI (rs-fMRI) data is used to study the intrinsic functional connectivity (FC) in the human brain. In the past decade, interest has focused on studying the temporal dynamics of FC on short timescales, ranging from seconds to minutes. These studies of time-varying FC (TVFC) have enabled the classification of whole-brain dynamic FC profiles into distinct "brain states", defined as recurring whole-brain connectivity profiles reliably observed across subjects and sessions. The analysis of rs-fMRI data is complicated by the fact that the measured BOLD signal consists of changes induced by neuronal activation, as well as non-neuronal nuisance fluctuations that should be removed prior to further analysis. Thus, the data undergoes significant preprocesing prior to analysis. In previous work [24], we illustrated the potential pitfalls involved with using modular preprocessing pipelines, showing how later preprocessing steps can reintroduce correlation with signal previously removed from the data. Here we show that the problem runs deeper, and that certain statistical analysis techniques can potentially interact with preprocessing and reintroduce correlations with previously removed signal. One such technique is the popular sliding window analysis, used to compute TVFC. In this paper, we discuss the problem both theoretically and empirically in application to test-retest rs-fMRI data. Importantly, we show that we are able to obtain essentially the same brain states and state transitions when analyzing motion induced signal as we do when analyzing the preprocessed but windowed data. Our results cast doubt on whether the estimated brain states obtained using sliding window analysis are neuronal in nature, or simply reflect non-neuronal nuisance signal variation (e.g., motion).

DeepN4: Learning N4ITK Bias Field Correction for T1-weighted Images

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 .

Multi-level feature extraction and reconstruction for 3D MRI image super-resolution

Magnetic resonance imaging (MRI) is an essential radiology technique in clinical diagnosis, but its spatial resolution may not suffice to meet the growing need for precise diagnosis due to hardware limitations and thicker slice thickness. Therefore, it is crucial to explore suitable methods to increase the resolution of MRI images. Recently, deep learning has yielded many impressive results in MRI image super-resolution (SR) reconstruction. However, current SR networks mainly use convolutions to extract relatively single image features, which may not be optimal for further enhancing the quality of image reconstruction. In this work, we propose a multi-level feature extraction and reconstruction (MFER) method to restore the degraded high-resolution details of MRI images. Specifically, to comprehensively extract different types of features, we design the triple-mixed convolution by leveraging the strengths and uniqueness of different filter operations. For the features of each level, we then apply deconvolutions to upsample them separately at the tail of the network, followed by the feature calibration of spatial and channel attention. Besides, we also use a soft cross-scale residual operation to improve the effectiveness of parameter optimization. Experiments on lesion-free and glioma datasets indicate that our method obtains superior quantitative performance and visual effects when compared with state-of-the-art MRI image SR methods.

DeepN4: Learning N4ITK Bias Field Correction for T1-weighted Images

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.

A Weakly Supervised Gradient Attribution Constraint for Interpretable Classification and Anomaly Detection

The lack of interpretability of deep learning reduces understanding of what happens when a network does not work as expected and hinders its use in critical fields like medicine, which require transparency of decisions. For example, a healthy vs pathological classification model should rely on radiological signs and not on some training dataset biases. Several post-hoc models have been proposed to explain the decision of a trained network. However, they are very seldom used to enforce interpretability during training and none in accordance with the classification. In this paper, we propose a new weakly supervised method for both interpretable healthy vs pathological classification and anomaly detection. A new loss function is added to a standard classification model to constrain each voxel of healthy images to drive the network decision towards the healthy class according to gradient-based attributions. This constraint reveals pathological structures for patient images, allowing their unsupervised segmentation. Moreover, we advocate both theoretically and experimentally, that constrained training with the simple Gradient attribution is similar to constraints with the heavier Expected Gradient, consequently reducing the computational cost. We also propose a combination of attributions during the constrained training making the model robust to the attribution choice at inference. Our proposition was evaluated on two brain pathologies: tumors and multiple sclerosis. This new constraint provides a more relevant classification, with a more pathology-driven decision. For anomaly detection, the proposed method outperforms state-of-the-art especially on difficult multiple sclerosis lesions segmentation task with a 15 points Dice improvement.

Reproducibility evaluation of the effects of MRI defacing on brain segmentation

Purpose

Recent advances in magnetic resonance (MR) scanner quality and the rapidly improving nature of facial recognition software have necessitated the introduction of MR defacing algorithms to protect patient privacy. As a result, there are a number of MR defacing algorithms available to the neuroimaging community, with several appearing in just the last 5 years. While some qualities of these defacing algorithms, such as patient identifiability, have been explored in the previous works, the potential impact of defacing on neuroimage processing has yet to be explored.

Approach

We qualitatively evaluate eight MR defacing algorithms on 179 subjects from the OASIS-3 cohort and 21 subjects from the Kirby-21 dataset. We also evaluate the effects of defacing on two neuroimaging pipelines-SLANT and FreeSurfer-by comparing the segmentation consistency between the original and defaced images.

Results

Defacing can alter brain segmentation and even lead to catastrophic failures, which are more frequent with some algorithms, such as Quickshear, MRI_Deface, and FSL_deface. Compared to FreeSurfer, SLANT is less affected by defacing. On outputs that pass the quality check, the effects of defacing are less pronounced than those of rescanning, as measured by the Dice similarity coefficient.

Conclusions

The effects of defacing are noticeable and should not be disregarded. Extra attention, in particular, should be paid to the possibility of catastrophic failures. It is crucial to adopt a robust defacing algorithm and perform a thorough quality check before releasing defaced datasets. To improve the reliability of analysis in scenarios involving defaced MRIs, it is encouraged to include multiple brain segmentation pipelines.

Improving Structural MRI Preprocessing with Hybrid Transformer GANs

Magnetic resonance imaging (MRI) is a technique that is widely used in practice to evaluate any pathologies in the human body. One of the areas of interest is the human brain. Naturally, MR images are low-resolution and contain noise due to signal interference, the patient's body's radio-frequency emissions and smaller Tesla coil counts in the machinery. There is a need to solve this problem, as MR tomographs that have the capability of capturing high-resolution images are extremely expensive and the length of the procedure to capture such images increases by the order of magnitude. Vision transformers have lately shown state-of-the-art results in super-resolution tasks; therefore, we decided to evaluate whether we can employ them for structural MRI super-resolution tasks. A literature review showed that similar methods do not focus on perceptual image quality because upscaled images are often blurry and are subjectively of poor quality. Knowing this, we propose a methodology called HR-MRI-GAN, which is a hybrid transformer generative adversarial network capable of increasing resolution and removing noise from 2D T1w MRI slice images. Experiments show that our method quantitatively outperforms other SOTA methods in terms of perceptual image quality and is capable of subjectively generalizing to unseen data. During the experiments, we additionally identified that the visual saliency-induced index metric is not applicable to MRI perceptual quality assessment and that general-purpose denoising networks are effective when removing noise from MR images.

Corticocerebellar White Matter Integrity Is Related to Naming Outcome in Post-Stroke Aphasia

Studies have shown that the integrity of white matter tracts connecting different regions in the left cerebral hemisphere is important for aphasia recovery after stroke. However, the impact of the underlying structural connection between the cortex and the cerebellum in post-stroke aphasia is poorly understood. We studied the microstructural integrity of the cerebellum and the corticocerebellar connections and their role in picture naming. Fifty-six patients with left cerebral infarcts (sparing the cerebellum) underwent diffusion tensor imaging (DTI) and Boston Naming Test. We compared the fractional anisotropy (FA) and mean diffusivity (MD) values of the right and the left cerebellum (lobular gray and white matter structures) and cerebellocortical connections. Recursive feature elimination and Spearman correlation analyses were performed to evaluate the relationship between naming performance and the corticocerebellar connections. We found that the right, relative to left, cerebellar structures and their connections with the left cerebrum showed lower FA and higher MD values, both reflecting lower microstructural integrity. This trend was not observed in the healthy controls. Higher MD values of the right major cerebellar outflow tract were associated with poorer picture naming performance. Our study provides the first DTI data demonstrating the critical importance of ascending and descending corticocerebellar connections for naming outcomes after stroke.

A reproducibility evaluation of the effects of MRI defacing on brain segmentation

Purpose

Recent advances in magnetic resonance (MR) scanner quality and the rapidly improving nature of facial recognition software have necessitated the introduction of MR defacing algorithms to protect patient privacy. As a result, there are a number of MR defacing algorithms available to the neuroimaging community, with several appearing in just the last five years. While some qualities of these defacing algorithms, such as patient identifiability, have been explored in previous works, the potential impact of defacing on neuroimage processing has yet to be explored.

Approach

We qualitatively evaluate eight MR defacing algorithms on 179 subjects from the OASIS-3 cohort and the 21 subjects from the Kirby-21 dataset. We also evaluate the effects of defacing on two neuroimaging pipelines-SLANT and FreeSurfer-by comparing the segmentation consistency between the original and defaced images.

Results

Defacing can alter brain segmentation and even lead to catastrophic failures, which are more frequent with some algorithms such as Quickshear, MRI_Deface, and FSL_deface. Compared to FreeSurfer, SLANT is less affected by defacing. On outputs that pass the quality check, the effects of defacing are less pronounced than those of rescanning, as measured by the Dice similarity coefficient.

Conclusions

The effects of defacing are noticeable and should not be disregarded. Extra attention, in particular, should be paid to the possibility of catastrophic failures. It is crucial to adopt a robust defacing algorithm and perform a thorough quality check before releasing defaced datasets. To improve the reliability of analysis in scenarios involving defaced MRIs, it's encouraged to include multiple brain segmentation pipelines.

Evaluation of T2-prepared blood oxygenation level dependent functional magnetic resonance imaging with an event-related task: Hemodynamic response function and reproducibility

T2-prepared (T2prep) blood oxygenation level dependent (BOLD) functional MRI (fMRI) is an alternative fMRI approach developed to mitigate the susceptibility artifacts that are typically observed in brain regions near air-filled cavities, bleeding and calcification, and metallic objects in echo-planar-imaging (EPI) based fMRI images. Here, T2prep BOLD fMRI was evaluated in an event-related paradigm for the first time. Functional experiments were performed using gradient-echo (GRE) EPI, spin-echo (SE) EPI, and T2prep BOLD fMRI during an event-related visual task in 10 healthy human subjects. Each fMRI method was performed with a low (3.4 × 3.4 × 4 mm3) and a high (1.5 mm isotropic) spatial resolution on 3T and a high resolution (1.5 mm isotropic) on 7T. Robust activation were detected in the visual cortex with all three fMRI methods. In each group of fMRI scans (3T low resolution, 3T high resolution, and 7T high resolution), GRE EPI showed the highest signal change (ΔS/S), largest full-width-at-half-maximum (FWHM) and longest time-to-peak (TTP) extracted from the hemodynamic response functions (HRF), indicating substantial signal contribution from large draining veins which have longer response times than microvessels. In contrast, T2prep BOLD showed the lowest ΔS/S, smallest FWHM, and shortest TTP, suggesting that T2prep BOLD may have a purer T2-weighted BOLD contrast that is more sensitive to microvessels compared to GRE/SE EPI BOLD. This trend was more obvious in fMRI scans performed with a lower spatial resolution on a lower field (3T with a 3.4 × 3.4 × 4 mm3 voxel). Scan-rescan reproducibility in the same subjects was comparable among the three fMRI methods. The results from the current study are expected to be useful to establish T2prep BOLD as a useful alternative fMRI approach for event-related fMRI in brain regions with large susceptibility artifacts.

Multisite test-retest reliability and compatibility of brain metrics derived from FreeSurfer versions 7.1, 6.0, and 5.3

Automatic neuroimaging processing tools provide convenient and systematic methods for extracting features from brain magnetic resonance imaging scans. One tool, FreeSurfer, provides an easy-to-use pipeline to extract cortical and subcortical morphometric measures. There have been over 25 stable releases of FreeSurfer, with different versions used across published works. The reliability and compatibility of regional morphometric metrics derived from the most recent version releases have yet to be empirically assessed. Here, we used test-retest data from three public data sets to determine within-version reliability and between-version compatibility across 42 regional outputs from FreeSurfer versions 7.1, 6.0, and 5.3. Cortical thickness from v7.1 was less compatible with that of older versions, particularly along the cingulate gyrus, where the lowest version compatibility was observed (intraclass correlation coefficient 0.37-0.61). Surface area of the temporal pole, frontal pole, and medial orbitofrontal cortex, also showed low to moderate version compatibility. We confirm low compatibility between v6.0 and v5.3 of pallidum and putamen volumes, while those from v7.1 were compatible with v6.0. Replication in an independent sample showed largely similar results for measures of surface area and subcortical volumes, but had lower overall regional thickness reliability and compatibility. Batch effect correction may adjust for some inter-version effects when most sites are run with one version, but results vary when more sites are run with different versions. Age associations in a quality controlled independent sample (N = 106) revealed version differences in results of downstream statistical analysis. We provide a reference to highlight the regional metrics that may yield recent version-related inconsistencies in published findings. An interactive viewer is provided at http://data.brainescience.org/Freesurfer_Reliability/.

ConvNets for automatic detection of polyglutamine SCAs from brain MRIs: state of the art applications

Polyglutamine spinocerebellar ataxias (polyQ SCAs) are a group of neurodegenerative diseases, clinically and genetically heterogeneous, characterized by loss of balance and motor coordination due to dysfunction of the cerebellum and its connections. The diagnosis of each type of polyQ SCA, alongside with genetic tests, includes medical images analysis, and its automation may help specialists to distinguish between each type. Convolutional neural networks (ConvNets or CNNs) have been recently used for medical image processing, with outstanding results. In this work, we present the main clinical and imaging features of polyglutamine SCAs, and the basics of CNNs. Finally, we review studies that have used this approach to automatically process brain medical images and may be applied to SCAs detection. We conclude by discussing the possible limitations and opportunities of using ConvNets for SCAs diagnose in the future.

CerebNet: A fast and reliable deep-learning pipeline for detailed cerebellum sub-segmentation

Quantifying the volume of the cerebellum and its lobes is of profound interest in various neurodegenerative and acquired diseases. Especially for the most common spinocerebellar ataxias (SCA), for which the first antisense oligonculeotide-base gene silencing trial has recently started, there is an urgent need for quantitative, sensitive imaging markers at pre-symptomatic stages for stratification and treatment assessment. This work introduces CerebNet, a fully automated, extensively validated, deep learning method for the lobular segmentation of the cerebellum, including the separation of gray and white matter. For training, validation, and testing, T1-weighted images from 30 participants were manually annotated into cerebellar lobules and vermal sub-segments, as well as cerebellar white matter. CerebNet combines FastSurferCNN, a UNet-based 2.5D segmentation network, with extensive data augmentation, e.g. realistic non-linear deformations to increase the anatomical variety, eliminating additional preprocessing steps, such as spatial normalization or bias field correction. CerebNet demonstrates a high accuracy (on average 0.87 Dice and 1.742mm Robust Hausdorff Distance across all structures) outperforming state-of-the-art approaches. Furthermore, it shows high test-retest reliability (average ICC >0.97 on OASIS and Kirby) as well as high sensitivity to disease effects, including the pre-ataxic stage of spinocerebellar ataxia type 3 (SCA3). CerebNet is compatible with FreeSurfer and FastSurfer and can analyze a 3D volume within seconds on a consumer GPU in an end-to-end fashion, thus providing an efficient and validated solution for assessing cerebellum sub-structure volumes. We make CerebNet available as source-code (https://github.com/Deep-MI/FastSurfer).

A benchmark for hypothalamus segmentation on T1-weighted MR images

The hypothalamus is a small brain structure that plays essential roles in sleep regulation, body temperature control, and metabolic homeostasis. Hypothalamic structural abnormalities have been reported in neuropsychiatric disorders, such as schizophrenia, amyotrophic lateral sclerosis, and Alzheimer's disease. Although mag- netic resonance (MR) imaging is the standard examination method for evaluating this region, hypothalamic morphological landmarks are unclear, leading to subjec- tivity and high variability during manual segmentation. Due to these limitations, it is common to find contradicting results in the literature regarding hypothalamic volumetry. To the best of our knowledge, only two automated methods are available in the literature for hypothalamus segmentation, the first of which is our previous method based on U-Net. However, both methods present performance losses when predicting images from different datasets than those used in training. Therefore, this project presents a benchmark consisting of a diverse T1-weighted MR image dataset comprising 1381 subjects from IXI, CC359, OASIS, and MiLI (the latter created specifically for this benchmark). All data were provided using automatically generated hypothalamic masks and a subset containing manually annotated masks. As a baseline, a method for fully automated segmentation of the hypothalamus on T1-weighted MR images with a greater generalization ability is presented. The pro- posed method is a teacher-student-based model with two blocks: segmentation and correction, where the second corrects the imperfections of the first block. After using three datasets for training (MiLI, IXI, and CC359), the prediction performance of the model was measured on two test sets: the first was composed of data from IXI, CC359, and MiLI, achieving a Dice coefficient of 0.83; the second was from OASIS, a dataset not used for training, achieving a Dice coefficient of 0.74. The dataset, the baseline model, and all necessary codes to reproduce the experiments are available at https://github.com/MICLab-Unicamp/HypAST and https://sites.google.com/ view/calgary-campinas-dataset/hypothalamus-benchmarking. In addition, a leaderboard will be maintained with predictions for the test set submitted by anyone working on the same task.

Automated multimodal segmentation of acute ischemic stroke lesions on clinical MR images

Magnetic resonance (MR) imaging (MRI) is commonly used to diagnose, assess and monitor stroke. Accurate and timely segmentation of stroke lesions provides the anatomico-structural information that can aid physicians in predicting prognosis, as well as in decision making and triaging for various rehabilitation strategies. To segment stroke lesions, MR protocols, including diffusion-weighted imaging (DWI) and T2-weighted fluid attenuated inversion recovery (FLAIR) are often utilized. These imaging sequences are usually acquired with different spatial resolutions due to time constraints. Within the same image, voxels may be anisotropic, with reduced resolution along slice direction for diffusion scans in particular. In this study, we evaluate the ability of 2D and 3D U-Net Convolutional Neural Network (CNN) architectures to segment ischemic stroke lesions using single contrast (DWI) and dual contrast images (T2w FLAIR and DWI). The predicted segmentations correlate with post-stroke motor outcome measured by the National Institutes of Health Stroke Scale (NIHSS) and Fugl-Meyer Upper Extremity (FM-UE) index based on the lesion loads overlapping the corticospinal tracts (CST), which is a neural substrate for motor movement and function. Although the four methods performed similarly, the 2D multimodal U-Net achieved the best results with a mean Dice of 0.737 (95% CI: 0.705, 0.769) and a relatively high correlation between the weighted lesion load and the NIHSS scores (both at baseline and at 90 days). A monotonically constrained quintic polynomial regression yielded R2 = 0.784 and 0.875 for weighted lesion load versus baseline and 90-Days NIHSS respectively, and better corrected Akaike information criterion (AICc) scores than those of the linear regression. In addition, using the quintic polynomial regression model to regress the weighted lesion load to the 90-Days FM-UE score results in an R2 of 0.570 with a better AICc score than that of the linear regression. Our results suggest that the multi-contrast information enhanced the accuracy of the segmentation and the prediction accuracy for upper extremity motor outcomes. Expanding the training dataset to include different types of stroke lesions and more data points will help add a temporal longitudinal aspect and increase the accuracy. Furthermore, adding patient-specific data may improve the inference about the relationship between imaging metrics and functional outcomes.

Synthetic Atrophy for Longitudinal Cortical Surface Analyses

In the fields of longitudinal cortical segmentation and surface-based cortical thickness (CT) measurement, difficulty in assessing accuracy remains a substantial limitation due to the inability of experimental validation against ground truth. Although methods have been developed to create synthetic datasets for these purposes, none provide a robust mechanism for measuring exact thickness changes with surface-based approaches. This work presents a registration-based technique for inducing synthetic cortical atrophy to create a longitudinal ground truth dataset specifically designed to address this gap in surface-based accuracy validation techniques. Across the entire brain, our method can induce up to between 0.8 and 2.5 mm of localized cortical atrophy in a given gyrus depending on the region's original thickness. By calculating the image deformation to induce this atrophy at 400% of the original resolution in each direction, we can induce a sub-voxel resolution amount of atrophy while minimizing partial volume effects. We also show that cortical segmentations of synthetically atrophied images exhibit similar segmentation error to those obtained from images of naturally atrophied brains. Importantly, our method relies exclusively on publicly available software and datasets.

Generalized Connectivity Matrix Response Regression with Applications in Brain Connectivity Studies

Multiple-subject network data are fast emerging in recent years, where a separate connectivity matrix is measured over a common set of nodes for each individual subject, along with subject covariates information. In this article, we propose a new generalized matrix response regression model, where the observed network is treated as a matrix-valued response and the subject covariates as predictors. The new model characterizes the population-level connectivity pattern through a low-rank intercept matrix, and the effect of subject covariates through a sparse slope tensor. We develop an efficient alternating gradient descent algorithm for parameter estimation, and establish the non-asymptotic error bound for the actual estimator from the algorithm, which quantifies the interplay between the computational and statistical errors. We further show the strong consistency for graph community recovery, as well as the edge selection consistency. We demonstrate the efficacy of our method through simulations and two brain connectivity studies.

Validation of an automatic tool for the rapid measurement of brain atrophy and white matter hyperintensity: QyScore®

OBJECTIVES

QyScore® is an imaging analysis tool certified in Europe (CE marked) and the US (FDA cleared) for the automatic volumetry of grey and white matter (GM and WM respectively), hippocampus (HP), amygdala (AM), and white matter hyperintensity (WMH). Here we compare QyScore® performances with the consensus of expert neuroradiologists.

METHODS

Dice similarity coefficient (DSC) and the relative volume difference (RVD) for GM, WM volumes were calculated on 50 3DT1 images. DSC and the F1 metrics were calculated for WMH on 130 3DT1 and FLAIR images. For each index, we identified thresholds of reliability based on current literature review results. We hypothesized that DSC/F1 scores obtained using QyScore® markers would be higher than the threshold. In contrast, RVD scores would be lower. Regression analysis and Bland-Altman plots were obtained to evaluate QyScore® performance in comparison to the consensus of three expert neuroradiologists.

RESULTS

The lower bound of the DSC/F1 confidence intervals was higher than the threshold for the GM, WM, HP, AM, and WMH, and the higher bounds of the RVD confidence interval were below the threshold for the WM, GM, HP, and AM. QyScore®, compared with the consensus of three expert neuroradiologists, provides reliable performance for the automatic segmentation of the GM and WM volumes, and HP and AM volumes, as well as WMH volumes.

CONCLUSIONS

QyScore® represents a reliable medical device in comparison with the consensus of expert neuroradiologists. Therefore, QyScore® could be implemented in clinical trials and clinical routine to support the diagnosis and longitudinal monitoring of neurological diseases.

KEY POINTS

• QyScore® provides reliable automatic segmentation of brain structures in comparison with the consensus of three expert neuroradiologists. • QyScore® automatic segmentation could be performed on MRI images using different vendors and protocols of acquisition. In addition, the fast segmentation process saves time over manual and semi-automatic methods. • QyScore® could be implemented in clinical trials and clinical routine to support the diagnosis and longitudinal monitoring of neurological diseases.

FastSurferVINN: Building resolution-independence into deep learning segmentation methods-A solution for HighRes brain MRI

Leading neuroimaging studies have pushed 3T MRI acquisition resolutions below 1.0 mm for improved structure definition and morphometry. Yet, only few, time-intensive automated image analysis pipelines have been validated for high-resolution (HiRes) settings. Efficient deep learning approaches, on the other hand, rarely support more than one fixed resolution (usually 1.0 mm). Furthermore, the lack of a standard submillimeter resolution as well as limited availability of diverse HiRes data with sufficient coverage of scanner, age, diseases, or genetic variance poses additional, unsolved challenges for training HiRes networks. Incorporating resolution-independence into deep learning-based segmentation, i.e., the ability to segment images at their native resolution across a range of different voxel sizes, promises to overcome these challenges, yet no such approach currently exists. We now fill this gap by introducing a Voxel-size Independent Neural Network (VINN) for resolution-independent segmentation tasks and present FastSurferVINN, which (i) establishes and implements resolution-independence for deep learning as the first method simultaneously supporting 0.7-1.0 mm whole brain segmentation, (ii) significantly outperforms state-of-the-art methods across resolutions, and (iii) mitigates the data imbalance problem present in HiRes datasets. Overall, internal resolution-independence mutually benefits both HiRes and 1.0 mm MRI segmentation. With our rigorously validated FastSurferVINN we distribute a rapid tool for morphometric neuroimage analysis. The VINN architecture, furthermore, represents an efficient resolution-independent segmentation method for wider application.

Circumventing the curse of dimensionality in magnetic resonance fingerprinting through a deep learning approach

Magnetic resonance fingerprinting (MRF) is a rapidly developing approach for fast quantitative MRI. A typical drawback of dictionary-based MRF is an explosion of the dictionary size as a function of the number of reconstructed parameters, according to the "curse of dimensionality", which determines an explosion of resource requirements. Neural networks (NNs) have been proposed as a feasible alternative, but this approach is still in its infancy. In this work, we design a deep learning approach to MRF using a fully connected network (FCN). In the first part we investigate, by means of simulations, how the NN performance scales with the number of parameters to be retrieved in comparison with the standard dictionary approach. Four MRF sequences were considered: IR-FISP, bSSFP, IR-FISP-B₁ , and IR-bSSFP-B₁ , the latter two designed to be more specific for B1+ parameter encoding. Estimation accuracy, memory usage, and computational time required to perform the estimation task were considered to compare the scalability capabilities of the dictionary-based and the NN approaches. In the second part we study optimal training procedures by including different data augmentation and preprocessing strategies during training to achieve better accuracy and robustness to noise and undersampling artifacts. The study is conducted using the IR-FISP MRF sequence exploiting both simulations and in vivo acquisitions. Results demonstrate that the NN approach outperforms the dictionary-based approach in terms of scalability capabilities. Results also allow us to heuristically determine the optimal training strategy to make an FCN able to predict T₁ ,  T₂ , and M₀ maps that are in good agreement with those obtained with the original dictionary approach. k-SVD denoising is proposed and found to be critical as a preprocessing step to handle undersampled data.

Simultaneous relaxometry and morphometry of human brain structures with 3D magnetic resonance fingerprinting: a multicenter, multiplatform, multifield-strength study

Relaxation times and morphological information are fundamental magnetic resonance imaging-derived metrics of the human brain that reflect the status of the underlying tissue. Magnetic resonance fingerprinting (MRF) enables simultaneous acquisition of T1 and T2 maps inherently aligned to the anatomy, allowing whole-brain relaxometry and morphometry in a single scan. In this study, we revealed the feasibility of 3D MRF for simultaneous brain structure-wise morphometry and relaxometry. Comprehensive test-retest scan analyses using five 1.5-T and three 3.0-T systems from a single vendor including different scanner types across 3 institutions demonstrated that 3D MRF-derived morphological information and relaxation times are highly repeatable at both 1.5 T and 3.0 T. Regional cortical thickness and subcortical volume values showed high agreement and low bias across different field strengths. The ability to acquire a set of regional T1, T2, thickness, and volume measurements of neuroanatomical structures with high repeatability and reproducibility facilitates the ability of longitudinal multicenter imaging studies to quantitatively monitor changes associated with underlying pathologies, disease progression, and treatments.

Deep robust residual network for super-resolution of 2D fetal brain MRI

Spatial resolution is a key factor of quantitatively evaluating the quality of magnetic resonance imagery (MRI). Super-resolution (SR) approaches can improve its spatial resolution by reconstructing high-resolution (HR) images from low-resolution (LR) ones to meet clinical and scientific requirements. To increase the quality of brain MRI, we study a robust residual-learning SR network (RRLSRN) to generate a sharp HR brain image from an LR input. Due to the Charbonnier loss can handle outliers well, and Gradient Difference Loss (GDL) can sharpen an image, we combined the Charbonnier loss and GDL to improve the robustness of the model and enhance the texture information of SR results. Two MRI datasets of adult brain, Kirby 21 and NAMIC, were used to train and verify the effectiveness of our model. To further verify the generalizability and robustness of the proposed model, we collected eight clinical fetal brain MRI 2D data for evaluation. The experimental results have shown that the proposed deep residual-learning network achieved superior performance and high efficiency over other compared methods.

Scan Once, Analyse Many: Using Large Open-Access Neuroimaging Datasets to Understand the Brain

We are now in a time of readily available brain imaging data. Not only are researchers now sharing data more than ever before, but additionally large-scale data collecting initiatives are underway with the vision that many future researchers will use the data for secondary analyses. Here I provide an overview of available datasets and some example use cases. Example use cases include examining individual differences, more robust findings, reproducibility-both in public input data and availability as a replication sample, and methods development. I further discuss a variety of considerations associated with using existing data and the opportunities associated with large datasets. Suggestions for further readings on general neuroimaging and topic-specific discussions are also provided.

Hypothesis Testing for Network Data with Power Enhancement

Comparing two population means of network data is of paramount importance in a wide range of scientific applications. Numerous existing network inference solutions focus on global testing of entire networks, without comparing individual network links. The observed data often take the form of vectors or matrices, and the problem is formulated as comparing two covariance or precision matrices under a normal or matrix normal distribution. Moreover, many tests suffer from a limited power under a small sample size. In this article, we tackle the problem of network comparison, both global and simultaneous inferences, when the data come in a different format, i.e., in the form of a collection of symmetric matrices, each of which encodes the network structure of an individual subject. Such data format commonly arises in applications such as brain connectivity analysis and clinical genomics. We no longer require the underlying data to follow a normal distribution, but instead impose some moment conditions that are easily satisfied for numerous types of network data. Furthermore, we propose a power enhancement procedure, and show that it can control the false discovery, while it has the potential to substantially enhance the power of the test. We investigate the efficacy of our testing procedure through both an asymptotic analysis and a simulation study under a finite sample size. We further illustrate our method with examples of brain connectivity analysis.

Fiber tractography bundle segmentation depends on scanner effects, vendor effects, acquisition resolution, diffusion sampling scheme, diffusion sensitization, and bundle segmentation workflow

When investigating connectivity and microstructure of white matter pathways of the brain using diffusion tractography bundle segmentation, it is important to understand potential confounds and sources of variation in the process. While cross-scanner and cross-protocol effects on diffusion microstructure measures are well described (in particular fractional anisotropy and mean diffusivity), it is unknown how potential sources of variation effect bundle segmentation results, which features of the bundle are most affected, where variability occurs, nor how these sources of variation depend upon the method used to reconstruct and segment bundles. In this study, we investigate six potential sources of variation, or confounds, for bundle segmentation: variation (1) across scan repeats, (2) across scanners, (3) across vendors (4) across acquisition resolution, (5) across diffusion schemes, and (6) across diffusion sensitization. We employ four different bundle segmentation workflows on two benchmark multi-subject cross-scanner and cross-protocol databases, and investigate reproducibility and biases in volume overlap, shape geometry features of fiber pathways, and microstructure features within the pathways. We find that the effects of acquisition protocol, in particular acquisition resolution, result in the lowest reproducibility of tractography and largest variation of features, followed by vendor-effects, scanner-effects, and finally diffusion scheme and b-value effects which had similar reproducibility as scan-rescan variation. However, confounds varied both across pathways and across segmentation workflows, with some bundle segmentation workflows more (or less) robust to sources of variation. Despite variability, bundle dissection is consistently able to recover the same location of pathways in the deep white matter, with variation at the gray matter/ white matter interface. Next, we show that differences due to the choice of bundle segmentation workflows are larger than any other studied confound, with low-to-moderate overlap of the same intended pathway when segmented using different methods. Finally, quantifying microstructure features within a pathway, we show that tractography adds variability over-and-above that which exists due to noise, scanner effects, and acquisition effects. Overall, these confounds need to be considered when harmonizing diffusion datasets, interpreting or combining data across sites, and when attempting to understand the successes and limitations of different methodologies in the design and development of new tractography or bundle segmentation methods.

Artificial Intelligence to Analyze the Cortical Thickness Through Age

In this study, Artificial Intelligence was used to analyze a dataset containing the cortical thickness from 1,100 healthy individuals. This dataset had the cortical thickness from 31 regions in the left hemisphere of the brain as well as from 31 regions in the right hemisphere. Then, 62 artificial neural networks were trained and validated to estimate the number of neurons in the hidden layer. These neural networks were used to create a model for the cortical thickness through age for each region in the brain. Using the artificial neural networks and kernels with seven points, numerical differentiation was used to compute the derivative of the cortical thickness with respect to age. The derivative was computed to estimate the cortical thickness speed. Finally, color bands were created for each region in the brain to identify a positive derivative, that is, a part of life with an increase in cortical thickness. Likewise, the color bands were used to identify a negative derivative, that is, a lifetime period with a cortical thickness reduction. Regions of the brain with similar derivatives were organized and displayed in clusters. Computer simulations showed that some regions exhibit abrupt changes in cortical thickness at specific periods of life. The simulations also illustrated that some regions in the left hemisphere do not follow the pattern of the same region in the right hemisphere. Finally, it was concluded that each region in the brain must be dynamically modeled. One advantage of using artificial neural networks is that they can learn and model non-linear and complex relationships. Also, artificial neural networks are immune to noise in the samples and can handle unseen data. That is, the models based on artificial neural networks can predict the behavior of samples that were not used for training. Furthermore, several studies have shown that artificial neural networks are capable of deriving information from imprecise data. Because of these advantages, the results obtained in this study by the artificial neural networks provide valuable information to analyze and model the cortical thickness.

Tractography dissection variability: What happens when 42 groups dissect 14 white matter bundles on the same dataset?

White matter bundle segmentation using diffusion MRI fiber tractography has become the method of choice to identify white matter fiber pathways in vivo in human brains. However, like other analyses of complex data, there is considerable variability in segmentation protocols and techniques. This can result in different reconstructions of the same intended white matter pathways, which directly affects tractography results, quantification, and interpretation. In this study, we aim to evaluate and quantify the variability that arises from different protocols for bundle segmentation. Through an open call to users of fiber tractography, including anatomists, clinicians, and algorithm developers, 42 independent teams were given processed sets of human whole-brain streamlines and asked to segment 14 white matter fascicles on six subjects. In total, we received 57 different bundle segmentation protocols, which enabled detailed volume-based and streamline-based analyses of agreement and disagreement among protocols for each fiber pathway. Results show that even when given the exact same sets of underlying streamlines, the variability across protocols for bundle segmentation is greater than all other sources of variability in the virtual dissection process, including variability within protocols and variability across subjects. In order to foster the use of tractography bundle dissection in routine clinical settings, and as a fundamental analytical tool, future endeavors must aim to resolve and reduce this heterogeneity. Although external validation is needed to verify the anatomical accuracy of bundle dissections, reducing heterogeneity is a step towards reproducible research and may be achieved through the use of standard nomenclature and definitions of white matter bundles and well-chosen constraints and decisions in the dissection process.

Functional Activities Detected in the Olfactory Bulb and Associated Olfactory Regions in the Human Brain Using T2-Prepared BOLD Functional MRI at 7T

Olfaction is a fundamental sense that plays a vital role in daily life in humans, and can be altered in neuropsychiatric and neurodegenerative diseases. Blood oxygenation level-dependent (BOLD) functional magnetic resonance imaging (fMRI) using conventional echo-planar-imaging (EPI) based sequences can be challenging in brain regions important for olfactory processing, such as the olfactory bulb (OB) and orbitofrontal cortex, mainly due to the signal dropout and distortion artifacts caused by large susceptibility effects from the sinonasal cavity and temporal bone. To date, few studies have demonstrated successful fMRI in the OB in humans. T2-prepared (T2prep) BOLD fMRI is an alternative approach developed especially for performing fMRI in regions affected by large susceptibility artifacts. The purpose of this technical study is to evaluate T2prep BOLD fMRI for olfactory functional experiments in humans. Olfactory fMRI scans were performed on 7T in 14 healthy participants. T2prep BOLD showed greater sensitivity than GRE EPI BOLD in the OB, orbitofrontal cortex and the temporal pole. Functional activation was detected using T2prep BOLD in the OB and associated olfactory regions. Habituation effects and a bi-phasic pattern of fMRI signal changes during olfactory stimulation were observed in all regions. Both positively and negatively activated regions were observed during olfactory stimulation. These signal characteristics are generally consistent with literature and showed a good intra-subject reproducibility comparable to previous human BOLD fMRI studies. In conclusion, the methodology demonstrated in this study holds promise for future olfactory fMRI studies in the OB and other brain regions that suffer from large susceptibility artifacts.

3D dense convolutional neural network for fast and accurate single MR image super-resolution

Super-resolution (SR) MR image reconstruction has shown to be a very promising direction to improve the spatial resolution of low-resolution (LR) MR images. In this paper, we presented a novel MR image SR method based on a dense convolutional neural network (DDSR), and its enhanced version called EDDSR. There are three major innovations: first, we re-designed dense modules to extract hierarchical features directly from LR images and propagate the extracted feature maps through dense connections. Therefore, unlike other CNN-based SR MR techniques that upsample LR patches in the initial phase, our methods take the original LR images or patches as input. This effectively reduces computational complexity and speeds up SR reconstruction. Second, a final deconvolution filter in our model automatically learns filters to fuse and upscale all hierarchical feature maps to generate HR MR images. Using this, EDDSR can perform SR reconstructions at different upscale factors using a single model with one stride fixed deconvolution operation. Third, to further improve SR reconstruction accuracy, we exploited a geometric self-ensemble strategy. Experimental results on three benchmark datasets demonstrate that our methods, DDSR and EDDSR, achieved superior performance compared to state-of-the-art MR image SR methods with less computational load and memory usage.

A deep learning approach for magnetic resonance fingerprinting: Scaling capabilities and good training practices investigated by simulations

MR fingerprinting (MRF) is an innovative approach to quantitative MRI. A typical disadvantage of dictionary-based MRF is the explosive growth of the dictionary as a function of the number of reconstructed parameters, an instance of the curse of dimensionality, which determines an explosion of resource requirements. In this work, we describe a deep learning approach for MRF parameter map reconstruction using a fully connected architecture. Employing simulations, we have investigated how the performance of the Neural Networks (NN) approach scales with the number of parameters to be retrieved, compared to the standard dictionary approach. We have also studied optimal training procedures by comparing different strategies for noise addition and parameter space sampling, to achieve better accuracy and robustness to noise. Four MRF sequences were considered: IR-FISP, bSSFP, IR-FISP-B₁, and IR-bSSFP-B₁. A comparison between NN and the dictionary approaches in reconstructing parameter maps as a function of the number of parameters to be retrieved was performed using a numerical brain phantom. Results demonstrated that training with random sampling and different levels of noise variance yielded the best performance. NN performance was at least as good as the dictionary-based approach in reconstructing parameter maps using Gaussian noise as a source of artifacts: the difference in performance increased with the number of estimated parameters because the dictionary method suffers from the coarse resolution of the parameter space sampling. The NN proved to be more efficient in memory usage and computational burden, and has great potential for solving large-scale MRF problems.

Temporal Dynamics of Brain White Matter Plasticity in Sighted Subjects during Tactile Braille Learning: A Longitudinal Diffusion Tensor Imaging Study

The white matter (WM) architecture of the human brain changes in response to training, though fine-grained temporal characteristics of training-induced white matter plasticity remain unexplored. We investigated white matter microstructural changes using diffusion tensor imaging at five different time points in 26 sighted female adults during 8 months of training on tactile braille reading. Our results show that training-induced white matter plasticity occurs both within and beyond the trained sensory modality, as reflected by fractional anisotropy (FA) increases in somatosensory and visual cortex, respectively. The observed changes followed distinct time courses, with gradual linear FA increase along the training in the somatosensory cortex and sudden visual cortex cross-modal plasticity occurring after braille input became linguistically meaningful. WM changes observed in these areas returned to baseline after the cessation of learning in line with the supply-demand model of plasticity. These results also indicate that the temporal dynamics of microstructural plasticity in different cortical regions might be modulated by the nature of computational demands. We provide additional evidence that observed FA training-induced changes are behaviorally relevant to tactile reading. Together, these results demonstrate that WM plasticity is a highly dynamic process modulated by the introduction of novel experiences.SIGNIFICANCE STATEMENT Throughout the lifetime the human brain is shaped by various experiences. Training-induced reorganization in white matter (WM) microstructure has been reported, but we know little about its temporal dynamics. To fill this gap, we scanned sighted subjects five times during tactile braille reading training. We observed different dynamics of WM plasticity in the somatosensory and visual cortices implicated in braille reading. The former showed a continuous increase in WM tissue anisotropy along with tactile training, while microstructural changes in the latter were observed only after the participants learned to read braille words. Our results confirm the supply-demand model of brain plasticity and provide evidence that WM reorganization depends on distinct computational demands and functional roles of regions involved in the trained skill.

A deep cascade of ensemble of dual domain networks with gradient-based T1 assistance and perceptual refinement for fast MRI reconstruction

Deep learning networks have shown promising results in fast magnetic resonance imaging (MRI) reconstruction. In our work, we develop deep networks to further improve the quantitative and the perceptual quality of reconstruction. To begin with, we propose reconsynergynet (RSN), a network that combines the complementary benefits of independently operating on both the image and the Fourier domain. For a single-coil acquisition, we introduce deep cascade RSN (DC-RSN), a cascade of RSN blocks interleaved with data fidelity (DF) units. Secondly, we improve the structure recovery of DC-RSN for T2 weighted Imaging (T2WI) through assistance of T1 weighted imaging (T1WI), a sequence with short acquisition time. T1 assistance is provided to DC-RSN through a gradient of log feature (GOLF) fusion. Furthermore, we propose perceptual refinement network (PRN) to refine the reconstructions for better visual information fidelity (VIF), a metric highly correlated to radiologist's opinion on the image quality. Lastly, for multi-coil acquisition, we propose variable splitting RSN (VS-RSN), a deep cascade of blocks, each block containing RSN, multi-coil DF unit, and a weighted average module. We extensively validate our models DC-RSN and VS-RSN for single-coil and multi-coil acquisitions and report the state-of-the-art performance. We obtain a SSIM of 0.768, 0.923, and 0.878 for knee single-coil-4x, multi-coil-4x, and multi-coil-8x in fastMRI, respectively. We also conduct experiments to demonstrate the efficacy of GOLF based T1 assistance and PRN.

Quantitative comparison of subcortical and ventricular volumetry derived from MPRAGE and MP2RAGE images using different brain morphometry software

OBJECTIVE

In brain volume assessment with MR imaging, it is of interest to know the effects of the pulse sequence and software used, to determine whether they provide equivalent data. The aim of this study was to compare cross-sectional volumes of subcortical and ventricular structures and their repeatability derived from MP2RAGE and MPRAGE images using MorphoBox, and FIRST or ALVIN.

MATERIALS AND METHODS

MPRAGE and MP2RAGE T1-weighted images were obtained from 24 healthy volunteers. Back-to-back scans were performed in 12 of them. Volumes, coefficients of variation, concordance, and correlations were determined.

RESULTS

Significant differences were found for volumes derived from MorphoBox and FIRST. Ventricular volumes determined by MorphoBox and ALVIN were similar. Differences between volumes obtained using MPRAGE and MP2RAGE were significant for a few regions. Coefficients of variation, ranged from 0.2 to 9.1%, showed a significant inverse correlation with the mean volume. There was a correlation between volume measures, but agreement was rated as poor for most regions.

CONCLUSION

MP2RAGE sequences and MorphoBox are valid options for assessing subcortical and ventricular volumes, in the same way as MPRAGE and FIRST or ALVIN, accepted tools for clinical research. However, caution is needed when comparing volumes obtained with different tools.

Cortical Thickness Estimation: A Comparison of FreeSurfer and Three Voxel-Based Methods in a Test-Retest Analysis and a Clinical Application

The cortical thickness has been used as a biomarker to assess different cerebral conditions and to detect alterations in the cortical mantle. In this work, we compare methods from the FreeSurfer software, the Computational Anatomy Toolbox (CAT12), a Laplacian approach and a new method here proposed, based on the Euclidean Distance Transform (EDT), and its corresponding computational phantom designed to validate the calculation algorithm. At region of interest (ROI) level, within- and inter-method comparisons were carried out with a test-retest analysis, in a subset comprising 21 healthy subjects taken from the Multi-Modal MRI Reproducibility Resource (MMRR) dataset. From the Minimal Interval Resonance Imaging in Alzheimer's Disease (MIRIAD) data, classification methods were compared in their performance to detect cortical thickness differences between 23 healthy controls (HC) and 45 subjects with Alzheimer's disease (AD). The validation of the proposed EDT-based method showed a more accurate and precise distance measurement as voxel resolution increased. For the within-method comparisons, mean test-retest measures (percentages differences/intraclass correlation/Pearson correlation) were similar for FreeSurfer (1.80%/0.90/0.95), CAT12 (1.91%/0.83/0.91), Laplacian (1.27%/0.89/0.95) and EDT (2.20%/0.88/0.94). Inter-method correlations showed moderate to strong values (R > 0.77) and, in the AD comparison study, all methods were able to detect cortical alterations between groups. Surface- and voxel-based methods have advantages and drawbacks regarding computational demands and measurement precision, while thickness definition was mainly associated to the cortical thickness absolute differences among methods. However, for each method, measurements were reliable, followed similar trends along the cortex and allowed detection of cortical atrophies between HC and patients with AD.

Impaired response of cerebral oxygen metabolism to visual stimulation in Huntington's disease

Huntington's disease (HD) is a neurodegenerative disease caused by a CAG triplet repeat expansion in the Huntingtin gene. Metabolic and microvascular abnormalities in the brain may contribute to early physiological changes that subserve the functional impairments in HD. This study is intended to investigate potential abnormality in dynamic changes in cerebral blood volume (CBV) and cerebral blood flow (CBF), and cerebral metabolic rate of oxygen (CMRO2) in the brain in response to functional stimulation in premanifest and early manifest HD patients. A recently developed 3-D-TRiple-acquisition-after-Inversion-Preparation magnetic resonance imaging (MRI) approach was used to measure dynamic responses in CBV, CBF, and CMRO2 during visual stimulation in one single MRI scan. Experiments were conducted in 23 HD patients and 16 healthy controls. Decreased occipital cortex CMRO2 responses were observed in premanifest and early manifest HD patients compared to controls (P < 0.001), correlating with the CAG-Age Product scores in these patients (R2 = 0.4, P = 0.001). The results suggest the potential value of this reduced CMRO2 response during visual stimulation as a biomarker for HD and may illuminate the role of metabolic alterations in the pathophysiology of HD.

The ANTsX ecosystem for quantitative biological and medical imaging

The Advanced Normalizations Tools ecosystem, known as ANTsX, consists of multiple open-source software libraries which house top-performing algorithms used worldwide by scientific and research communities for processing and analyzing biological and medical imaging data. The base software library, ANTs, is built upon, and contributes to, the NIH-sponsored Insight Toolkit. Founded in 2008 with the highly regarded Symmetric Normalization image registration framework, the ANTs library has since grown to include additional functionality. Recent enhancements include statistical, visualization, and deep learning capabilities through interfacing with both the R statistical project (ANTsR) and Python (ANTsPy). Additionally, the corresponding deep learning extensions ANTsRNet and ANTsPyNet (built on the popular TensorFlow/Keras libraries) contain several popular network architectures and trained models for specific applications. One such comprehensive application is a deep learning analog for generating cortical thickness data from structural T1-weighted brain MRI, both cross-sectionally and longitudinally. These pipelines significantly improve computational efficiency and provide comparable-to-superior accuracy over multiple criteria relative to the existing ANTs workflows and simultaneously illustrate the importance of the comprehensive ANTsX approach as a framework for medical image analysis.

Quantity and quality: Normative open-access neuroimaging databases

The focus of this article is to compare twenty normative and open-access neuroimaging databases based on quantitative measures of image quality, namely, signal-to-noise (SNR) and contrast-to-noise ratios (CNR). We further the analysis through discussing to what extent these databases can be used for the visualization of deeper regions of the brain, such as the subcortex, as well as provide an overview of the types of inferences that can be drawn. A quantitative comparison of contrasts including T1-weighted (T1w) and T2-weighted (T2w) images are summarized, providing evidence for the benefit of ultra-high field MRI. Our analysis suggests a decline in SNR in the caudate nuclei with increasing age, in T1w, T2w, qT1 and qT2* contrasts, potentially indicative of complex structural age-dependent changes. A similar decline was found in the corpus callosum of the T1w, qT1 and qT2* contrasts, though this relationship is not as extensive as within the caudate nuclei. These declines were accompanied by a declining CNR over age in all image contrasts. A positive correlation was found between scan time and the estimated SNR as well as a negative correlation between scan time and spatial resolution. Image quality as well as the number and types of contrasts acquired by these databases are important factors to take into account when selecting structural data for reuse. This article highlights the opportunities and pitfalls associated with sampling existing databases, and provides a quantitative backing for their usage.

Evaluating phase synchronization methods in fMRI: A comparison study and new approaches

In recent years there has been growing interest in measuring time-varying functional connectivity between different brain regions using resting-state functional magnetic resonance imaging (rs-fMRI) data. One way to assess the relationship between signals from different brain regions is to measure their phase synchronization (PS) across time. There are several ways to perform such analyses, and we compare methods that utilize a PS metric together with a sliding window, referred to here as windowed phase synchronization (WPS), with those that directly measure the instantaneous phase synchronization (IPS). In particular, IPS has recently gained popularity as it offers single time-point resolution of time-resolved fMRI connectivity. In this paper, we discuss the underlying assumptions required for performing PS analyses and emphasize the importance of band-pass filtering the data to obtain valid results. Further, we contrast this approach with the use of Empirical Mode Decomposition (EMD) to achieve similar goals. We review various methods for evaluating PS and introduce a new approach within the IPS framework denoted the cosine of the relative phase (CRP). We contrast methods through a series of simulations and application to rs-fMRI data. Our results indicate that CRP outperforms other tested methods and overcomes issues related to undetected temporal transitions from positive to negative associations common in IPS analysis. Further, in contrast to phase coherence, CRP unfolds the distribution of PS measures, which benefits subsequent clustering of PS matrices into recurring brain states.

Three dimensional MRF obtains highly repeatable and reproducible multi-parametric estimations in the healthy human brain at 1.5T and 3T

Magnetic resonance fingerprinting (MRF) is highly promising as a quantitative MRI technique due to its accuracy, robustness, and efficiency. Previous studies have found high repeatability and reproducibility of 2D MRF acquisitions in the brain. Here, we have extended our investigations to 3D MRF acquisitions covering the whole brain using spiral projection k-space trajectories. Our travelling head study acquired test/retest data from the brains of 12 healthy volunteers and 8 MRI systems (3 systems at 3 T and 5 at 1.5 T, all from a single vendor), using a study design not requiring all subjects to be scanned at all sites. The pulse sequence and reconstruction algorithm were the same for all acquisitions. After registration of the MRF-derived PD T1 and T2 maps to an anatomical atlas, coefficients of variation (CVs) were computed to assess test/retest repeatability and inter-site reproducibility in each voxel, while a General Linear Model (GLM) was used to determine the voxel-wise variability between all confounders, which included test/retest, subject, field strength and site. Our analysis demonstrated a high repeatability (CVs 0.7-1.3% for T1, 2.0-7.8% for T2, 1.4-2.5% for normalized PD) and reproducibility (CVs of 2.0-5.8% for T1, 7.4-10.2% for T2, 5.2-9.2% for normalized PD) in gray and white matter. Both repeatability and reproducibility improved when compared to similar experiments using 2D acquisitions. Three-dimensional MRF obtains highly repeatable and reproducible estimations of T1 and T2, supporting the translation of MRF-based fast quantitative imaging into clinical applications.

Synthetic atrophy for longitudinal surface-based cortical thickness measurement

Difficulty in validating accuracy remains a substantial setback in the field of surface-based cortical thickness (CT) measurement due to the lack of experimental validation against ground truth. Although methods have been developed to create synthetic datasets for this purpose, none provide a robust mechanism for measuring exact thickness changes with surface-based approaches. This work presents a registration-based technique for inducing synthetic cortical atrophy to create a longitudinal, ground truth dataset specifically designed for accuracy validation of surface-based CT measurements. Across the entire brain, we show our method can induce up to between 0.6 and 2.6 mm of localized cortical atrophy in a given gyrus depending on the region's original thickness. By calculating the image deformation to induce this atrophy at 400% of the original resolution in each direction, we can induce a sub-voxel resolution amount of atrophy while minimizing partial volume effects. We also show that our method can be extended beyond application to CT measurements for the accuracy validation of longitudinal cortical segmentation and surface reconstruction pipelines when measuring accuracy against cortical landmarks. Importantly, our method relies exclusively on publicly available software and datasets.

Replicability, Repeatability, and Long-term Reproducibility of Cerebellar Morphometry

To identify robust and reproducible methods of cerebellar morphometry that can be used in future large-scale structural MRI studies, we investigated the replicability, repeatability, and long-term reproducibility of three fully automated software tools: FreeSurfer, CEREbellum Segmentation (CERES), and automatic cerebellum anatomical parcellation using U-Net with locally constrained optimization (ACAPULCO). Replicability was defined as computational replicability, determined by comparing two analyses of the same high-resolution MRI data set performed with identical analysis software and computer hardware. Repeatability was determined by comparing the analyses of two MRI scans of the same participant taken during two independent MRI sessions on the same day for the Kirby-21 study. Long-term reproducibility was assessed by analyzing two MRI scans of the same participant in the longitudinal OASIS-2 study. We determined percent difference, the image intraclass correlation coefficient, the coefficient of variation, and the intraclass correlation coefficient between two analyses. Our results show that CERES and ACAPULCO use stochastic algorithms that result in surprisingly high differences between identical analyses for ACAPULCO and small differences for CERES. Changes between two consecutive scans from the Kirby-21 study were less than ± 5% in most cases for FreeSurfer and CERES (i.e., demonstrating high repeatability). As expected, long-term reproducibility was lower than repeatability for all software tools. In summary, CERES is an accurate, as demonstrated before, and reproducible tool for fully automated segmentation and parcellation of the cerebellum. We conclude with recommendations for the assessment of replicability, repeatability, and long-term reproducibility in future studies on cerebellar structure.

Apolipoprotein E allele 4 effects on Single-Subject Gray Matter Networks in Mild Cognitive Impairment

There is evidence that gray matter networks are disrupted in Mild Cognitive Impairment (MCI) and associated with cognitive impairment and faster disease progression. However, it remains unknown how these alterations are related to the presence of Apolipoprotein E isoform E4 (ApoE4), the most prominent genetic risk factor for late-onset Alzheimer's disease (AD). To investigate this topic at the individual level, we explore the impact of ApoE4 and the disease progression on the Single-Subject Gray Matter Networks (SSGMNets) using the graph theory approach. Our data sample comprised 200 MCI patients selected from the ADNI database, classified as non-Converters and Converters (will progress into AD). Each group included 50 ApoE4-positive ('Carriers', ApoE4 + ) and 50 ApoE4-negative ('non-Carriers', ApoE4-). The SSGMNets were estimated from structural MRIs at two-time points: baseline and conversion. We investigated whether altered network topological measures at baseline and their rate of change (RoC) between baseline and conversion time points were associated with ApoE4 and disease progression. We also explored the correlation of SSGMNets attributes with general cognition score (MMSE), memory (ADNI-MEM), and CSF-derived biomarkers of AD (Aβ42, T-tau, and P-tau). Our results showed that ApoE4 and the disease progression modulated the global topological network properties independently but not in their RoC. MCI converters showed a lower clustering index in several regions associated with neurodegeneration in AD. The SSGMNets' topological organization was revealed to be able to predict cognitive and memory measures. The findings presented here suggest that SSGMNets could indeed be used to identify MCI ApoE4 Carriers with a high risk for AD progression.

Matchability of heterogeneous networks pairs

We consider the problem of graph matchability in non-identically distributed networks. In a general class of edge-independent networks, we demonstrate that graph matchability can be lost with high probability when matching the networks directly. We further demonstrate that under mild model assumptions, matchability is almost perfectly recovered by centering the networks using universal singular value thresholding before matching. These theoretical results are then demonstrated in both real and synthetic simulation settings. We also recover analogous core-matchability results in a very general core-junk network model, wherein some vertices do not correspond between the graph pair.