A systematic comparison of VBM pipelines and their application to age prediction

Sleep deprivation as a risk factor for cortical gray matter reduction in new medical residents

BACKGROUND AND PURPOSE

Sleep is an essential physiological condition for the proper functioning of humans, both physiologically, cognitively, and psychologically. Sleep deprivation leads to a loss of psychomotor skills in humans. It is important to evaluate the structural changes experienced by medical residents who are sleep-deprived due to extensive work shifts, including night shifts, assigned during their training program. Therefore, the main outcome was to evaluate the structural changes in the cortical gray matter and the hippocampus assessed by brain magnetic resonance imaging (MRI) in newly admitted medical residents four months after the start of the medical specialty.

MATERIAL AND METHODS

Forty-one newly admitted medical residents were enrolled, and an initial questionnaire was administered to assess sleep quality. All participants underwent a brain MRI study, utilizing an advanced MRI sequence: a 3D inversion recovery (IR)-prepped fast spoiled gradient-recalled (SPGR) high-resolution T1-weighted sequence. The images were then anonymized and reformatted, and volumetric analyses of gray matter and hippocampus were performed using an open-access platform for MRI brain analysis (volBrain). This process was repeated four months later with the acquisition of a new brain MRI study for each participant.

RESULTS

For gray matter volume, a baseline value of 728.04 ± 63.95 cm³ and a final value of 715.11 ± 59.38 cm³ were found (p < 0.01), and the frontal lobe showed the greatest reduction, with an initial value of 181.92 ± 15.58 cm3 and a final volume of 176.45 ± 17.35 cm3 (p = <0.001). We found an OR of 1.52 (95% CI 0.93-4.14, p = 0.01) between working night shifts and gray matter reduction.

CONCLUSIONS

The results of this study show a statistically significant reduction in gray matter volume in first-year residents after four months of shift work, with the greatest reduction in the frontal lobe.

A Case for Automated Segmentation of MRI Data in Neurodegenerative Diseases: Type II GM1 Gangliosidosis

BACKGROUND

Volumetric analysis and segmentation of magnetic resonance imaging (MRI) data is an important tool for evaluating neurological disease progression and neurodevelopment. Fully automated segmentation pipelines offer faster and more reproducible results. However, since these analysis pipelines were trained on or run based on atlases consisting of neurotypical controls, it is important to evaluate how accurate these methods are for neurodegenerative diseases. In this study, we compared five fully automated segmentation pipelines, including FSL, Freesurfer, volBrain, SPM12, and SimNIBS, with a manual segmentation process in GM1 gangliosidosis patients and neurotypical controls.

METHODS

We analyzed 45 MRI scans from 16 juvenile GM1 gangliosidosis patients, 11 MRI scans from 8 late-infantile GM1 gangliosidosis patients, and 19 MRI scans from 11 neurotypical controls. We compared the results for seven brain structures, including volumes of the total brain, bilateral thalamus, ventricles, bilateral caudate nucleus, bilateral lentiform nucleus, corpus callosum, and cerebellum.

RESULTS

We found volBrain's vol2Brain pipeline to have the strongest correlations with the manual segmentation process for the whole brain, ventricles, and thalamus. We also found Freesurfer's recon-all pipeline to have the strongest correlations with the manual segmentation process for the caudate nucleus. For the cerebellum, we found a combination of volBrain's vol2Brain and SimNIBS' headreco to have the strongest correlations, depending on the cohort. For the lentiform nucleus, we found a combination of recon-all and FSL's FIRST to give the strongest correlations, depending on the cohort. Lastly, we found segmentation of the corpus callosum to be highly variable.

CONCLUSIONS

Previous studies have considered automated segmentation techniques to be unreliable, particularly in neurodegenerative diseases. However, in our study, we produced results comparable to those obtained with a manual segmentation process. While manual segmentation processes conducted by neuroradiologists remain the gold standard, we present evidence to the capabilities and advantages of using an automated process that includes the ability to segment white matter throughout the brain or analyze large datasets, which pose feasibility issues to fully manual processes. Future investigations should consider the use of artificial intelligence-based segmentation pipelines to determine their accuracy in GM1 gangliosidosis, lysosomal storage disorders, and other neurodegenerative diseases.

Macro- and Microstructural Alterations in the Midbrain in Early Psychosis Associates with Clinical Symptom Scores

Early psychosis (EP) is a critical period for psychotic disorders during which the brain undergoes rapid and significant functional and structural changes ( Shinn et al., 2017). The Human Connectome Project (HCP) is a global effort to map the human brain's connectivity in health and disease. Here we focus on HCP-EP subjects (i.e., those within 5 years of the initial psychotic episode) to determine macro- and microstructural alterations in EP (HCP-EP sample, n = 179: EP, n = 123, controls, n = 56) and their association with clinical outcomes (i.e., symptoms severity) in HCP-EP. We carried out analyses of deformation-based morphometry (DBM), scalar indices from the diffusion tensor imaging (DTI), and tract-based spatial statistics (TBSS). Lastly, we conducted correlation analyses focused on the midbrain (DBM and DTI) to examine associations between its structure and clinical symptoms. Our results show that the midbrain displays robust alteration in its structure (DBM and DTI) in the voxel-based analysis. Complimentary alterations were also observed for the hippocampus and putamen. A seed-based analysis centered around the midbrain confirms the voxel-based analysis of DBM and DTI. TBSS displays structural differences within the midbrain and complementary alterations in the corticospinal tract and cingulum. Correlations between the midbrain structures and behavior showed that the quantified features correlate with cognition and clinical scores. Our findings contribute to understanding the midbrain-focused circuitry involvement in EP and provide a path for future investigations to inform specific brain-based biomarkers of EP.

A Case for Automated Segmentation of MRI Data in Milder Neurodegenerative Diseases

Background

Volumetric analysis and segmentation of magnetic resonance imaging (MRI) data is an important tool for evaluating neurological disease progression and neurodevelopment. Fully automated segmentation pipelines offer faster and more reproducible results. However, since these analysis pipelines were trained on or run based on atlases consisting of neurotypical controls, it is important to evaluate how accurate these methods are for neurodegenerative diseases. In this study, we compared 5 fully automated segmentation pipelines including FSL, Freesurfer, volBrain, SPM12, and SimNIBS with a manual segmentation process in GM1 gangliosidosis patients and neurotypical controls.

Methods

We analyzed 45 MRI scans from 16 juvenile GM1 gangliosidosis patients, 11 MRI scans from 8 late-infantile GM1 gangliosidosis patients, and 19 MRI scans from 11 neurotypical controls. We compared results for 7 brain structures including volumes of the total brain, bilateral thalamus, ventricles, bilateral caudate nucleus, bilateral lentiform nucleus, corpus callosum, and cerebellum.

Results

We found volBrain's vol2Brain pipeline to have the strongest correlations with the manual segmentation process for the whole brain, ventricles, and thalamus. We also found Freesurfer's recon-all pipeline to have the strongest correlations with the manual segmentation process for the caudate nucleus. For the cerebellum, we found a combination of volBrain's vol2Brain and SimNIBS' headreco to have the strongest correlations depending on the cohort. For the lentiform nucleus, we found a combination of recon-all and FSL's FIRST to give the strongest correlations depending on the cohort. Lastly, we found segmentation of the corpus callosum to be highly variable.

Conclusion

Previous studies have considered automated segmentation techniques to be unreliable, particularly in neurodegenerative diseases. However, in our study we produced results comparable to those obtained with a manual segmentation process. While manual segmentation processes conducted by neuroradiologists remain the gold standard, we present evidence to the capabilities and advantages of using an automated process including the ability to segment white matter throughout the brain or analyze large datasets, which pose feasibility issues to fully manual processes. Future investigations should consider the use of artificial intelligence-based segmentation pipelines to determine their accuracy in GM1 gangliosidosis, lysosomal storage disorders, and other neurodegenerative diseases.

Neuroanatomical Insights: Convergence and Divergence of Tinnitus with Normal or Mild Hearing Loss

Objectives: To explore the neuroanatomical abnormalities in idiopathic tinnitus patients by voxel-based morphometry (VBM) and surface-based morphometry (SBM) techniques. To elucidate the central plasticity in tinnitus patients with normal or mild hearing loss from the neuroanatomical insights. Methods: A total of 74 patients with idiopathic tinnitus (43 with normal hearing and 31 with mild hearing loss) and 98 healthy subjects were enrolled. VBM and SBM were employed to analyze neuroimaging data and identify neuroanatomical differences. Results: Our analysis revealed a reduction in gray matter volume and a distinctive pattern of changes in cortical surface features in patients with idiopathic tinnitus, especially in brain regions closely related to the limbic system, such as the bilateral parahippocampal gyrus, bilateral entorhinal cortex, and insula. Tinnitus patients with mild hearing loss have more extensive gray matter volume reduction, and more complex changes in cortical surface features compared to tinnitus patients with normal hearing. In addition, we also found a significant correlation between the Self-Rating Anxiety Scale (SAS), the Self-Rating Depression Scale (SDS), and Montreal Cognitive Assessment (MoCA) scores of patients with idiopathic tinnitus and cortical characteristic parameters in the above brain regions. Conclusions: There are extensive neuroanatomical alterations in tinnitus patients. Mild hearing loss may aggravate the reduction of gray matter volume and change the surface characteristics of the cortex. Anxiety, depression, and cognitive impairment in patients with idiopathic tinnitus may be related to neuroanatomical alterations in specific brain regions.

Relationships of eating behaviors with psychopathology, brain maturation and genetic risk for obesity in an adolescent cohort study

Unhealthy eating, a risk factor for eating disorders (EDs) and obesity, often coexists with emotional and behavioral problems; however, the underlying neurobiological mechanisms are poorly understood. Analyzing data from the longitudinal IMAGEN adolescent cohort, we investigated associations between eating behaviors, genetic predispositions for high body mass index (BMI) using polygenic scores (PGSs), and trajectories (ages 14-23 years) of ED-related psychopathology and brain maturation. Clustering analyses at age 23 years (N = 996) identified 3 eating groups: restrictive, emotional/uncontrolled and healthy eaters. BMI PGS, trajectories of ED symptoms, internalizing and externalizing problems, and brain maturation distinguished these groups. Decreasing volumes and thickness in several brain regions were less pronounced in restrictive and emotional/uncontrolled eaters. Smaller cerebellar volume reductions uniquely mediated the effects of BMI PGS on restrictive eating, whereas smaller volumetric reductions across multiple brain regions mediated the relationship between elevated externalizing problems and emotional/uncontrolled eating, independently of BMI. These findings shed light on distinct contributions of genetic risk, protracted brain maturation and behaviors in ED symptomatology.

Neurostructural correlates of harm action/outcome aversion: The role of empathy

Harm aversion is essential for normal human functioning; however, the neuroanatomical mechanisms underlying harm aversion remain unclear. To explore this issue, we examined the brain structures associated with the two distinct dimensions of harm aversion (harm action/outcome aversion) and the potential mediating role of the four aspects of empathy: fantasy, perspective-taking, empathic concern, and personal distress. A sample of 214 healthy young adults underwent structural magnetic resonance imaging. Voxel-based morphometry was used to assess regional gray matter volume (rGMV) and regional gray matter density (rGMD). Whole-brain multiple regression analysis revealed significant correlations between harm action aversion and rGMV/rGMD in various brain regions, including the inferior frontal gyrus (IFG) and precuneus for both rGMV and rGMD, the cerebellum for rGMV, and the superior frontal gyrus for rGMD. The rGMV/rGMD in the IFG and the rGMD in the primary somatosensory cortex (S1) were correlated with harm outcome aversion. Utilizing 10-fold balanced cross-validation analysis, we confirmed the robustness of these significant associations between rGMV/rGMD in these brain regions and harm action/outcome aversion. Importantly, mediation analysis revealed that empathic concern mediated the relationship between rGMV/rGMD in the precuneus and harm action aversion. Additionally, empathic concern, personal distress, and total empathy mediated the relationship between rGMD in the S1 and harm outcome aversion. These findings enhance our understanding of the neural mechanism of harm aversion by integrating insights from the brain structure, harm aversion, and the personality hierarchy models while also extending the frontal asymmetry model of Emotion.

A perspective on brain-age estimation and its clinical promise

Brain-age estimation has gained increased attention in the neuroscientific community owing to its potential use as a biomarker of brain health. The difference between estimated and chronological age based on neuroimaging data enables a unique perspective on brain development and aging, with multiple open questions still remaining in the brain-age research field. This Perspective presents an overview of current advancements in the field and envisions the future evolution of the brain-age framework before its potential deployment in hospital settings.

Macro- and Micro-Structural Alterations in the Midbrain in Early Psychosis

Introduction

Early psychosis (EP) is a critical period in the course of psychotic disorders during which the brain is thought to undergo rapid and significant functional and structural changes 1 . Growing evidence suggests that the advent of psychotic disorders is early alterations in the brain's functional connectivity and structure, leading to aberrant neural network organization. The Human Connectome Project (HCP) is a global effort to map the human brain's connectivity in healthy and disease populations; within HCP, there is a specific dataset that focuses on the EP subjects (i.e., those within five years of the initial psychotic episode) (HCP-EP), which is the focus of our study. Given the critically important role of the midbrain function and structure in psychotic disorders (cite), and EP in particular (cite), we specifically focused on the midbrain macro- and micro-structural alterations and their association with clinical outcomes in HCP-EP.

Methods

We examined macro- and micro-structural brain alterations in the HCP-EP sample (n=179: EP, n=123, Controls, n=56) as well as their associations with behavioral measures (i.e., symptoms severity) using a stepwise approach, incorporating a multimodal MRI analysis procedure. First, Deformation Based Morphometry (DBM) was carried out on the whole brain 3 Tesla T1w images to examine gross brain anatomy (i.e., seed-based and voxel-based volumes). Second, we extracted Fractional Anisotropy (FA), Axial Diffusivity (AD), and Mean Diffusivity (MD) indices from the Diffusion Tensor Imaging (DTI) data; a midbrain mask was created based on FreeSurfer v.6.0 atlas. Third, we employed Tract-Based Spatial Statistics (TBSS) to determine microstructural alterations in white matter tracts within the midbrain and broader regions. Finally, we conducted correlation analyses to examine associations between the DBM-, DTI- and TBSS-based outcomes and the Positive and Negative Syndrome Scale (PANSS) scores.

Results

DBM analysis showed alterations in the hippocampus, midbrain, and caudate/putamen. A DTI voxel-based analysis shows midbrain reductions in FA and AD and increases in MD; meanwhile, the hippocampus shows an increase in FA and a decrease in AD and MD. Several key brain regions also show alterations in DTI indices (e.g., insula, caudate, prefrontal cortex). A seed-based analysis centered around a midbrain region of interest obtained from freesurfer segmentation confirms the voxel-based analysis of DTI indices. TBSS successfully captured structural differences within the midbrain and complementary alterations in other main white matter tracts, such as the corticospinal tract and cingulum, suggesting early altered brain connectivity in EP. Correlations between these quantities in the EP group and behavioral scores (i.e., PANSS and CAINS tests) were explored. It was found that midbrain volume noticeably correlates with the Cognitive score of PA and all DTI metrics. FA correlates with the several dimensions of the PANSS, while AD and MD do not show many associations with PANSS or CAINS.

Conclusions

Our findings contribute to understanding the midbrain-focused circuitry involvement in EP and complimentary alteration in EP. Our work provides a path for future investigations to inform specific brain-based biomarkers of EP and their relationships to clinical manifestations of the psychosis course.

Magnetoencephalography Atlas Viewer for Dipole Localization and Viewing

Magnetoencephalography (MEG) is a noninvasive neuroimaging technique widely recognized for epilepsy and tumor mapping. MEG clinical reporting requires a multidisciplinary team, including expert input regarding each dipole's anatomic localization. Here, we introduce a novel tool, the "Magnetoencephalography Atlas Viewer" (MAV), which streamlines this anatomical analysis. The MAV normalizes the patient's Magnetic Resonance Imaging (MRI) to the Montreal Neurological Institute (MNI) space, reverse-normalizes MNI atlases to the native MRI, identifies MEG dipole files, and matches dipoles' coordinates to their spatial location in atlas files. It offers a user-friendly and interactive graphical user interface (GUI) for displaying individual dipoles, groups, coordinates, anatomical labels, and a tri-planar MRI view of the patient with dipole overlays. It evaluated over 273 dipoles obtained in clinical epilepsy subjects. Consensus-based ground truth was established by three neuroradiologists, with a minimum agreement threshold of two. The concordance between the ground truth and MAV labeling ranged from 79% to 84%, depending on the normalization method. Higher concordance rates were observed in subjects with minimal or no structural abnormalities on the MRI, ranging from 80% to 90%. The MAV provides a straightforward MEG dipole anatomic localization method, allowing a nonspecialist to prepopulate a report, thereby facilitating and reducing the time of clinical reporting.