A sulcal depth-based anatomical parcellation of the cerebral cortex

Brain structure, phenotypic and genetic correlates of reading performance

Reading is an evolutionarily recent development that recruits and tunes brain circuitry connecting primary- and language-processing regions. We investigated whether metrics of the brain's physical structure correlate with reading performance and whether genetic variants affect this relationship. To this aim, we used the Adolescent Brain Cognitive Development dataset (n = 9,013) of 9-10-year-olds and focused on 150 measures of cortical surface area (CSA) and thickness. Our results reveal that reading performance is associated with nine measures of brain structure including relevant regions of the reading network. Furthermore, we show that this relationship is partially mediated by genetic factors for two of these measures: the CSA of the entire left hemisphere and, specifically, of the left superior temporal gyrus CSA. These effects emphasize the complex and subtle interplay between genes, brain and reading, which is a partly heritable polygenic skill that relies on a distributed network.

Longitudinal evolution of cortical thickness signature reflecting Lewy body dementia in isolated REM sleep behavior disorder: a prospective cohort study

BACKGROUND

The isolated rapid-eye-movement sleep behavior disorder (iRBD) is a prodromal condition of Lewy body disease including Parkinson's disease and dementia with Lewy bodies (DLB). We aim to investigate the longitudinal evolution of DLB-related cortical thickness signature in a prospective iRBD cohort and evaluate the possible predictive value of the cortical signature index in predicting dementia-first phenoconversion in individuals with iRBD.

METHODS

We enrolled 22 DLB patients, 44 healthy controls, and 50 video polysomnography-proven iRBD patients. Participants underwent 3-T magnetic resonance imaging (MRI) and clinical/neuropsychological evaluations. We characterized DLB-related whole-brain cortical thickness spatial covariance pattern (DLB-pattern) using scaled subprofile model of principal components analysis that best differentiated DLB patients from age-matched controls. We analyzed clinical and neuropsychological correlates of the DLB-pattern expression scores and the mean values of the whole-brain cortical thickness in DLB and iRBD patients. With repeated MRI data during the follow-up in our prospective iRBD cohort, we investigated the longitudinal evolution of the cortical thickness signature toward Lewy body dementia. Finally, we analyzed the potential predictive value of cortical thickness signature as a biomarker of phenoconversion in iRBD cohort.

RESULTS

The DLB-pattern was characterized by thinning of the temporal, orbitofrontal, and insular cortices and relative preservation of the precentral and inferior parietal cortices. The DLB-pattern expression scores correlated with attentional and frontal executive dysfunction (Trail Making Test-A and B: R = - 0.55, P = 0.024 and R = - 0.56, P = 0.036, respectively) as well as visuospatial impairment (Rey-figure copy test: R = - 0.54, P = 0.0047). The longitudinal trajectory of DLB-pattern revealed an increasing pattern above the cut-off in the dementia-first phenoconverters (Pearson's correlation, R = 0.74, P = 6.8 × 10^-4) but no significant change in parkinsonism-first phenoconverters (R = 0.0063, P = 0.98). The mean value of the whole-brain cortical thickness predicted phenoconversion in iRBD patients with hazard ratio of 9.33 [1.16-74.12]. The increase in DLB-pattern expression score discriminated dementia-first from parkinsonism-first phenoconversions with 88.2% accuracy.

CONCLUSION

Cortical thickness signature can effectively reflect the longitudinal evolution of Lewy body dementia in the iRBD population. Replication studies would further validate the utility of this imaging marker in iRBD.

Relationships between cognitive performance, clinical insight and regional brain volumes in schizophrenia

Impairments in cognitive performance are common in schizophrenia, and these contribute to poor awareness of symptoms and treatment (‘clinical insight’), which is an important predictor of functional outcome. Although relationships between cognitive impairment and reductions in regional brain volumes in patients are relatively well characterised, less is known about the brain structural correlates of clinical insight. To address this gap, we aimed to explore brain structural correlates of cognitive performance and clinical insight in the same sample. 108 patients with schizophrenia (SZH) and 94 age and gender-matched controls (CON) (from the Northwestern University Schizophrenia Data and Software Tool (NUSDAST) database) were included. SZH had smaller grey matter volume across most fronto-temporal regions and significantly poorer performance on all cognitive domains. Multiple regression showed that higher positive symptoms and poorer attention were significant predictors of insight in SZH; however, no significant correlations were seen between clinical insight and regional brain volumes. In contrast, symptomology did not contribute to cognitive performance, but robust positive relationships were found between regional grey matter volumes in fronto-temporal regions and cognitive performance (particularly executive function). Many of these appeared to be unique to SZH as they were not observed in CON. Findings suggest that while there exists a tight link between cognitive functioning and neuropathological processes affecting gross brain anatomy in SZH, this is not the case for clinical insight. Instead, clinical insight levels seem to be influenced by symptomology, attentional performance and other subject-specific variables.

Scalable Query Answering Under Uncertainty to Neuroscientific Ontological Knowledge: The NeuroLang Approach

Researchers in neuroscience have a growing number of datasets available to study the brain, which is made possible by recent technological advances. Given the extent to which the brain has been studied, there is also available ontological knowledge encoding the current state of the art regarding its different areas, activation patterns, keywords associated with studies, etc. Furthermore, there is inherent uncertainty associated with brain scans arising from the mapping between voxels-3D pixels-and actual points in different individual brains. Unfortunately, there is currently no unifying framework for accessing such collections of rich heterogeneous data under uncertainty, making it necessary for researchers to rely on ad hoc tools. In particular, one major weakness of current tools that attempt to address this task is that only very limited propositional query languages have been developed. In this paper we present NeuroLang, a probabilistic language based on first-order logic with existential rules, probabilistic uncertainty, ontologies integration under the open world assumption, and built-in mechanisms to guarantee tractable query answering over very large datasets. NeuroLang's primary objective is to provide a unified framework to seamlessly integrate heterogeneous data, such as ontologies, and map fine-grained cognitive domains to brain regions through a set of formal criteria, promoting shareable and highly reproducible research. After presenting the language and its general query answering architecture, we discuss real-world use cases showing how NeuroLang can be applied to practical scenarios.

Prediction of Obsessive-Compulsive Disorder: Importance of Neurobiology-Aided Feature Design and Cross-Diagnosis Transfer Learning

BACKGROUND

Machine learning applications using neuroimaging provide a multidimensional, data-driven approach that captures the level of complexity necessary for objectively aiding diagnosis and prognosis in psychiatry. However, models learned from small training samples often have limited generalizability, which continues to be a problem with automated diagnosis of mental illnesses such as obsessive-compulsive disorder (OCD). Earlier studies have shown that features incorporating prior neurobiological knowledge of brain function and combining brain parcellations from various sources can potentially improve the overall prediction. However, it is unknown whether such knowledge-driven methods can provide a performance that is comparable to state-of-the-art approaches based on neural networks.

METHODS

In this study, we apply a transparent and explainable multiparcellation ensemble learning framework EMPaSchiz (Ensemble algorithm with Multiple Parcellations for Schizophrenia prediction) to the task of predicting OCD, based on a resting-state functional magnetic resonance imaging dataset of 350 subjects. Furthermore, we apply transfer learning using the features found effective for schizophrenia to OCD to leverage the commonality in brain alterations across these psychiatric diagnoses.

RESULTS

We show that our knowledge-based approach leads to a prediction performance of 80.3% accuracy for OCD diagnosis that is better than domain-agnostic and automated feature design using neural networks. Furthermore, we show that a selection of reduced feature sets can be transferred from schizophrenia to the OCD prediction model without significant loss in prediction performance.

CONCLUSIONS

This study presents a machine learning framework for OCD prediction with neurobiology-aided feature design using resting-state functional magnetic resonance imaging that is generalizable and reasonably interpretable.

Clustering of cognitive subtypes in schizophrenia patients and their siblings: relationship with regional brain volumes

Schizophrenia patients (SZH) often show impaired cognition and reduced brain structural volumes; these deficits are also detectable in healthy relatives of SZH. However, there is considerable heterogeneity: a sizable percentage of SZH are relatively cognitively intact; clustering strategies have proved useful for categorising into cognitive subgroups. We used a clustering strategy to investigate relationships between subgroup assignment and brain volumes, in 102 SZH (N = 102) and 32 siblings of SZH (SZH-SIB), alongside 92 controls (CON) and 48 of their siblings. SZH had poorer performance in all cognitive domains, and smaller brain volumes within prefrontal and temporal regions compared to controls. We identified three distinct cognitive clusters (‘neuropsychologically normal’, ‘intermediate’, ‘cognitively impaired’) based on age- and gender-adjusted cognitive domain scores. The majority of SZH (60.8%) were assigned to the cognitively impaired cluster, while the majority of SZH-SIB (65.6%) were placed in the intermediate cluster. Greater right middle temporal volume distinguished the normal cluster from the more impaired clusters. Importantly, the observed brain volume differences between SZH and controls disappeared after adjustment for cluster assignment. This suggests an intimate link between cognitive performance levels and regional brain volume differences in SZH. This highlights the importance of accounting for heterogeneity in cognitive performance within SZH populations when attempting to characterise the brain structural abnormalities associated with the disease.

Learnable Pooling in Graph Convolutional Networks for Brain Surface Analysis

Brain surface analysis is essential to neuroscience, however, the complex geometry of the brain cortex hinders computational methods for this task. The difficulty arises from a discrepancy between 3D imaging data, which is represented in Euclidean space, and the non-Euclidean geometry of the highly-convoluted brain surface. Recent advances in machine learning have enabled the use of neural networks for non-Euclidean spaces. These facilitate the learning of surface data, yet pooling strategies often remain constrained to a single fixed-graph. This paper proposes a new learnable graph pooling method for processing multiple surface-valued data to output subject-based information. The proposed method innovates by learning an intrinsic aggregation of graph nodes based on graph spectral embedding. We illustrate the advantages of our approach with in-depth experiments on two large-scale benchmark datasets. The ablation study in the paper illustrates the impact of various factors affecting our learnable pooling method. The flexibility of the pooling strategy is evaluated on four different prediction tasks, namely, subject-sex classification, regression of cortical region sizes, classification of Alzheimer's disease stages, and brain age regression. Our experiments demonstrate the superiority of our learnable pooling approach compared to other pooling techniques for graph convolutional networks, with results improving the state-of-the-art in brain surface analysis.

Highly Reproducible Whole Brain Parcellation in Individuals via Voxel Annotation with Fiber Clusters

A central goal in systems neuroscience is to parcellate the brain into discrete units that are neurobiologically coherent. Here, we propose a strategy for consistent whole-brain parcellation of white matter (WM) and gray matter (GM) in individuals. We parcellate the brain into coherent parcels using non-negative matrix factorization based on voxel annotation using fiber clusters. Tractography is performed using an algorithm that mitigates gyral bias, allowing full gyral and sulcal coverage for reliable parcellation of the cortical ribbon. Experimental results indicate that parcellation using our approach is highly reproducible with 100% test-retest parcel identification rate and is highly consistent with significantly lower inter-subject variability than FreeSurfer parcellation. This implies that reproducible parcellation can be obtained for subject-specific investigation of brain structure and function.

Cognitive impairment in schizophrenia: relationships with cortical thickness in fronto-temporal regions, and dissociability from symptom severity

Cognitive impairments are a core and persistent characteristic of schizophrenia with implications for daily functioning. These show only limited response to antipsychotic treatment and their neural basis is not well characterised. Previous studies point to relationships between cortical thickness and cognitive performance in fronto-temporal brain regions in schizophrenia patients (SZH). There is also evidence that these relationships might be independent of symptom severity, suggesting dissociable disease processes. We set out to explore these possibilities in a sample of 70 SZH and 72 age and gender-matched healthy controls (provided by the Center of Biomedical Research Excellence (COBRE)). Cortical thickness within fronto-temporal regions implicated by previous work was considered in relation to performance across various cognitive domains (from the MATRICS Cognitive Battery). Compared to controls, SZH had thinner cortices across most fronto-temporal regions and significantly lower performance on all cognitive domains. Robust relationships with cortical thickness were found: visual learning and attention performance correlated with bilateral superior and middle frontal thickness in SZH only. Correlations between attention performance and right transverse temporal thickness were also specific to SZH. Findings point to the importance of these regions for cognitive performance in SZH, possibly reflecting compensatory processes and/or aberrant connectivity. No links to symptom severity were observed in these regions, suggesting these relationships are dissociable from underlying psychotic symptomology. Findings enhance understanding of the brain structural underpinnings and possible aetiology of cognitive impairment in SZH.

Brain structural correlates of functional capacity in first-episode psychosis

Impaired functional capacity is a core feature of schizophrenia and presents even in first-episode psychosis (FEP) patients. Impairments in daily functioning tend to persist despite antipsychotic therapy but their neural basis is less clear. Previous studies suggest that volume loss in frontal cortex might be an important contributor, but findings are inconsistent. We aimed to comprehensively investigate the brain structural correlates of functional capacity in FEP using MRI and a reliable objective measure of functioning [University of California, San Diego Performance-Based Skills Assessment (UPSA)]. In a sample of FEP (n = 39) and a well-matched control group (n = 21), we measured cortical thickness, gray matter volume, and white matter tract integrity (fractional anisotropy, FA) within brain regions implicated by previous work. The FEP group had thinner cortex in various frontal regions and fusiform, and reduced FA in inferior longitudinal fasciculus (ILF). In FEP, poorer functional capacity correlated with reduced superior frontal volume and lower FA in left ILF. Importantly, frontal brain volumes and integrity of the ILF were identified as the structural correlates of functional capacity in FEP, controlling for other relevant factors. These findings enhance mechanistic understanding of functional capacity deficits in schizophrenia by specifying the underlying neural correlates. In future, this could help inform intervention strategies.

Neurovegetative symptom subtypes in young people with major depressive disorder and their structural brain correlates

Depression is a leading cause of burden of disease among young people. Current treatments are not uniformly effective, in part due to the heterogeneous nature of major depressive disorder (MDD). Refining MDD into more homogeneous subtypes is an important step towards identifying underlying pathophysiological mechanisms and improving treatment of young people. In adults, symptom-based subtypes of depression identified using data-driven methods mainly differed in patterns of neurovegetative symptoms (sleep and appetite/weight). These subtypes have been associated with differential biological mechanisms, including immuno-metabolic markers, genetics and brain alterations (mainly in the ventral striatum, medial orbitofrontal cortex, insular cortex, anterior cingulate cortex amygdala and hippocampus). K-means clustering was applied to individual depressive symptoms from the Quick Inventory of Depressive Symptoms (QIDS) in 275 young people (15-25 years old) with MDD to identify symptom-based subtypes, and in 244 young people from an independent dataset (a subsample of the STAR*D dataset). Cortical surface area and thickness and subcortical volume were compared between the subtypes and 100 healthy controls using structural MRI. Three subtypes were identified in the discovery dataset and replicated in the independent dataset; severe depression with increased appetite, severe depression with decreased appetite and severe insomnia, and moderate depression. The severe increased appetite subtype showed lower surface area in the anterior insula compared to both healthy controls. Our findings in young people replicate the previously identified symptom-based depression subtypes in adults. The structural alterations of the anterior insular cortex add to the existing evidence of different pathophysiological mechanisms involved in this subtype.

iEEGview: an open-source multifunction GUI-based Matlab toolbox for localization and visualization of human intracranial electrodes

OBJECTIVE

The precise localization of intracranial electrodes is a fundamental step relevant to the analysis of intracranial electroencephalography (iEEG) recordings in various fields. With the increasing development of iEEG studies in human neuroscience, higher requirements have been posed on the localization process, resulting in urgent demand for more integrated, easy-operation and versatile tools for electrode localization and visualization. With the aim of addressing this need, we develop an easy-to-use and multifunction toolbox called iEEGview, which can be used for the localization and visualization of human intracranial electrodes.

APPROACH

iEEGview is written in Matlab scripts and implemented with a GUI. From the GUI, by taking only pre-implant MRI and post-implant CT images as input, users can directly run the full localization pipeline including brain segmentation, image co-registration, electrode reconstruction, anatomical information identification, activation map generation and electrode projection from native brain space into common brain space for group analysis. Additionally, iEEGview implements methods for brain shift correction, visual location inspection on MRI slices and computation of certainty index in anatomical label assignment.

MAIN RESULTS

All the introduced functions of iEEGview work reliably and successfully, and are tested by images from 28 human subjects implanted with depth and/or subdural electrodes.

SIGNIFICANCE

iEEGview is the first public Matlab GUI-based software for intracranial electrode localization and visualization that holds integrated capabilities together within one pipeline. iEEGview promotes convenience and efficiency for the localization process, provides rich localization information for further analysis and offers solutions for addressing raised technical challenges. Therefore, it can serve as a useful tool in facilitating iEEG studies.

Towards artificial intelligence in mental health by improving schizophrenia prediction with multiple brain parcellation ensemble-learning

In the literature, there are substantial machine learning attempts to classify schizophrenia based on alterations in resting-state (RS) brain patterns using functional magnetic resonance imaging (fMRI). Most earlier studies modelled patients undergoing treatment, entailing confounding with drug effects on brain activity, and making them less applicable to real-world diagnosis at the point of first medical contact. Further, most studies with classification accuracies >80% are based on small sample datasets, which may be insufficient to capture the heterogeneity of schizophrenia, limiting generalization to unseen cases. In this study, we used RS fMRI data collected from a cohort of antipsychotic drug treatment-naive patients meeting DSM IV criteria for schizophrenia (N = 81) as well as age- and sex-matched healthy controls (N = 93). We present an ensemble model -- EMPaSchiz (read as ‘Emphasis’; standing for ‘Ensemble algorithm with Multiple Parcellations for Schizophrenia prediction’) that stacks predictions from several ‘single-source’ models, each based on features of regional activity and functional connectivity, over a range of different a priori parcellation schemes. EMPaSchiz yielded a classification accuracy of 87% (vs. chance accuracy of 53%), which out-performs earlier machine learning models built for diagnosing schizophrenia using RS fMRI measures modelled on large samples (N > 100). To our knowledge, EMPaSchiz is first to be reported that has been trained and validated exclusively on data from drug-naive patients diagnosed with schizophrenia. The method relies on a single modality of MRI acquisition and can be readily scaled-up without needing to rebuild parcellation maps from incoming training images.

Diffusion tensor imaging and voxel-based morphometry reveal corticospinal tract involvement in the motor dysfunction of adult-onset myotonic dystrophy type 1

Magnetic resonance imaging (MRI) studies have demonstrated that patients with myotonic dystrophy type 1 (DM1) exhibit gray and white matter abnormalities that are correlated with various genetic and neuropsychological measures. However, few MRI studies have focused on the correlations between brain abnormalities and overall motor function including gait performance. Here, we investigated the correlations between brain abnormalities, as assessed with MRI including diffusion tensor imaging (DTI), and motor performance, as assessed with the Medical Research Council sum score (MRCSS), 6-minute walk test (6MWT), and hand grip power, in patients with DM1. Eighteen patients with DM1 and twenty healthy controls participated in this study. The MRCSS and 6MWT reflect patients’ general motor performance, particularly gait, while hand grip reflects the presence of myotonia. We found significant relationships between DTI parameters in the corticospinal tract (CST) and genetic factors and motor performance in patients with DM1. These findings suggest that CST involvement reflecting deterioration of the motor tracts may play a significant role in clinical myotonia. Further, a direct relationship between the cortical gray matter volume and DTI measures in the CST suggests that white matter abnormalities in the CST are strongly associated with volume reductions in the sensorimotor cortex of patients with DM1.

A Multi-Tissue Global Estimation Framework for Asymmetric Fiber Orientation Distributions

In connectomics, tractography involves tracing connections across gray-white matter boundaries in gyral blades of complex cortical convolutions. To date, most tractography algorithms exhibit gyral bias with fiber streamlines preferentially terminating at gyral crowns rather than sulcal banks or fundi. In this work, we will demonstrate that a multi-tissue global estimation framework of the asymmetric fiber orientation distribution function (AFODF) will mitigate the effects of gyral bias and will allow fiber streamlines at gyral blades to make sharper turns into the cortical gray matter. This is validated using in-vivo data from the Human Connectome Project (HCP), showing that, in a typical gyral blade with high curvature, the fiber streamlines estimated using AFODFs bend more naturally into the cortex than FODFs. Furthermore, we show that AFODF tractography results in better cortico-cortical connectivity.

Associations between Verbal Learning Slope and Neuroimaging Markers across the Cognitive Aging Spectrum

A symptom of mild cognitive impairment (MCI) and Alzheimer's disease (AD) is a flat learning profile. Learning slope calculation methods vary, and the optimal method for capturing neuroanatomical changes associated with MCI and early AD pathology is unclear. This study cross-sectionally compared four different learning slope measures from the Rey Auditory Verbal Learning Test (simple slope, regression-based slope, two-slope method, peak slope) to structural neuroimaging markers of early AD neurodegeneration (hippocampal volume, cortical thickness in parahippocampal gyrus, precuneus, and lateral prefrontal cortex) across the cognitive aging spectrum [normal control (NC); (n=198; age=76±5), MCI (n=370; age=75±7), and AD (n=171; age=76±7)] in ADNI. Within diagnostic group, general linear models related slope methods individually to neuroimaging variables, adjusting for age, sex, education, and APOE4 status. Among MCI, better learning performance on simple slope, regression-based slope, and late slope (Trial 2-5) from the two-slope method related to larger parahippocampal thickness (all p-values<.01) and hippocampal volume (p<.01). Better regression-based slope (p<.01) and late slope (p<.01) were related to larger ventrolateral prefrontal cortex in MCI. No significant associations emerged between any slope and neuroimaging variables for NC (p-values ≥.05) or AD (p-values ≥.02). Better learning performances related to larger medial temporal lobe (i.e., hippocampal volume, parahippocampal gyrus thickness) and ventrolateral prefrontal cortex in MCI only. Regression-based and late slope were most highly correlated with neuroimaging markers and explained more variance above and beyond other common memory indices, such as total learning. Simple slope may offer an acceptable alternative given its ease of calculation.

Brain complex network analysis by means of resting state fMRI and graph analysis: will it be helpful in clinical epilepsy?

Functional magnetic resonance imaging (fMRI) has just completed 20 years of existence. It currently serves as a research tool in a broad range of human brain studies in normal and pathological conditions, as is the case of epilepsy. To date, most fMRI studies aimed at characterizing brain activity in response to various active paradigms. More recently, a number of strategies have been used to characterize the low-frequency oscillations of the ongoing fMRI signals when individuals are at rest. These datasets have been largely analyzed in the context of functional connectivity, which inspects the covariance of fMRI signals from different areas of the brain. In addition, resting state fMRI is progressively being used to evaluate complex network features of the brain. These strategies have been applied to a number of different problems in neuroscience, which include diseases such as Alzheimer's, schizophrenia, and epilepsy. Hence, we herein aimed at introducing the subject of complex network and how to use it for the analysis of fMRI data. This appears to be a promising strategy to be used in clinical epilepsy. Therefore, we also review the recent literature that has applied these ideas to the analysis of fMRI data in patients with epilepsy.

Hemispheric asymmetries of cortical volume in the human brain

Hemispheric asymmetry represents a cardinal feature of cerebral organization, but the nature of structural and functional differences between the hemispheres is far from fully understood. Using Magnetic Resonance Imaging morphometry, we identified several volumetric differences between the two hemispheres of the human brain. Heteromodal inferoparietal and lateral prefrontal cortices are more extensive in the right than left hemisphere, as is visual cortex. Heteromodal mesial and orbital prefrontal and cingulate cortices are more extensive in the left than right hemisphere, as are somatosensory, parts of motor, and auditory cortices. Thus, heteromodal association cortices are more extensively represented on the lateral aspect of the right than in the left hemisphere, and modality-specific cortices are more extensively represented on the lateral aspect of the left than in the right hemisphere. On the mesial aspect heteromodal association cortices are more extensively represented in the left than right hemisphere.