Human Brain Mapping最新文献

The behavioural variant of frontotemporal dementia (bvFTD) is a younger-onset dementia syndrome characterised by early atrophy of frontoinsular cortices, manifesting in profound socioemotional disturbances. Converging evidence from correlational, data-driven, and computational approaches indicates large-scale network degeneration in bvFTD. While the insula is consistently implicated, it remains unclear whether insular atrophy causally impacts progressive large-scale structural network alterations in bvFTD. Eighty-two patients with clinically probable bvFTD were classified as very mild/mild (n = 35), moderate (n = 30), and severe (n = 17) using the CDR plus NACC FTLD. Grey matter volume comparison between the entire bvFTD group and a healthy control group matched for age and education identified the left anterior insula as the initial maximal site of atrophy in bvFTD. To determine potential causal effects of insular atrophy on network-based dysfunction in bvFTD, a voxel-wise causal structural covariance network (CaSCN) was constructed based on pseudo-time-series morphometric data using the left anterior insula as the seed region. Sex, age, years of education, total intracranial volume (TIV), and scanning site were included as covariates, along with the difference between the sum of boxes score for the CDR plus NACC FTLD across the two pseudo–time points. Finally, an event-based model (EBM) was applied to confirm the sequence of regional atrophy precipitated by left anterior insula atrophy, which emerged in the CaSCN analysis. BvFTD patients in the very mild/mild disease subgroup showed predominant atrophy of frontotemporal (e.g., insula, middle frontal gyrus), limbic (e.g., hippocampus, amygdala), and subcortical (e.g., putamen, nucleus accumbens) structures. Widespread grey matter atrophy was evident in the moderate bvFTD subgroup, extending to the middle cingulate, paracingulate gyri, and the thalamus, which progressed to posterior brain regions, including the fusiform gyrus and the cerebellum in the severe subgroup. Importantly, the CaSCN and event-based model analysis reinforced the disease-staging results by revealing progression of atrophy from the initial seed region of the left anterior insula to the orbitofrontal cortex, putamen/nucleus accumbens, anterior cingulate cortex, dorsolateral prefrontal cortex, inferior temporal gyrus, and supramarginal gyrus, before progressing posteriorly to the lingual gyrus. Using causal structural covariance network analysis and event-based modelling, our findings indicate a causal role for the left anterior insula in driving the spread of pathology in bvFTD through well-delineated functional brain networks known to support higher-order cognitive and socioemotional processing. By capturing the direction of atrophy progression, our findings hold utility for potentially monitoring and tracking the efficacy of novel therapeutics on brain function in bvFTD.

The loss of smell is common in older age, reducing quality of life and often precedes the onset of cognitive decline and dementia. While age-related olfactory loss has been linked to cortical thinning and volume reductions in key olfactory areas, associations between white-matter (WM) integrity and olfaction are poorly understood. Here, we studied individuals aged 25–85 years from a population-based cohort study with diffusion weighted imaging, together with self-reported olfactory impairment, odor identification and odor threshold measures at baseline (N = 248) and follow-up 5 years later (N = 192). Performance on the odor identification and threshold tests were lower in older adults and declined longitudinally. Older individuals also reported more olfaction complaints, and such complaints increased over time. Results from general linear models showed no cross-sectional associations between WM integrity and olfaction. However, results from non-competitive random forest models identified several tracts as significant contributors to odor identification and subjective olfactory impairment, including the fornix, cingulum and uncinate fasciculus. Moreover, longitudinal analyses showed that olfactory threshold decline was associated with decline in WM integrity in the body of corpus callosum. Taken together, the results support a link between white-matter integrity and olfaction and provide initial evidence for its interplay with age.

The accurate estimation of fiber orientation distribution functions (fODFs) in diffusion magnetic resonance imaging (MRI) is crucial for understanding early brain development and its potential disruptions. Although supervised deep learning (DL) models have shown promise in fODF estimation from neonatal diffusion MRI (dMRI) data, the out-of-domain (OOD) performance of these models remains largely unexplored, especially under diverse domain shift scenarios. This study evaluated the robustness of three state-of-the-art DL architectures: multilayer perceptron (MLP), transformer, and U-Net/convolutional neural network (CNN) on fODF predictions derived from dMRI data. Using 488 subjects from the developing Human Connectome Project (dHCP) and the Baby Connectome Project (BCP) datasets, we reconstructed reference fODFs from the full dMRI series using single-shell three-tissue constrained spherical deconvolution (SS3T-CSD) and multi-shell multi-tissue CSD (MSMT-CSD) to generate reference fODF reconstructions for model training, and systematically assessed the impact of age, scanner/protocol differences, and input dimensionality on model performance. Our findings reveal that U-Net consistently outperformed other models when fewer diffusion gradient directions were used, particularly with the SS3T-CSD-derived ground truth, which showed superior performance in capturing crossing fibers. However, as the number of input diffusion gradient directions increased, MLP and the transformer-based model exhibited steady gains in accuracy. Nevertheless, performance nearly plateaued from 28 to 45 input directions in all models. Age-related domain shifts showed asymmetric patterns, being less pronounced in late developmental stages (late neonates, and babies), with SS3T-CSD demonstrating greater robustness to variability compared to MSMT-CSD. To address inter-site domain shifts, we implemented two adaptation strategies: the Method of Moments (MoM) and fine-tuning. Both strategies achieved significant improvements () in over 95% of tested configurations, with fine-tuning consistently yielding superior results and U-Net benefiting the most from increased target subjects. This study represents the first systematic evaluation of OOD settings in DL applications to fODF estimation, providing critical insights into model robustness and adaptation strategies for diverse clinical and research applications.

Early life stress (ELS) has profound implications for developmental trajectories, yet the neural mechanisms underlying its long-term effects remain incompletely understood. In the present study, we examined whether interindividual similarity in ELS exposure aligns with similarity in neural representations and behavioral task performance in early adulthood. Leveraging a 20-year longitudinal dataset of Finnish families, we evaluated 87 young adults who underwent functional magnetic resonance imaging (fMRI) during an emotional go/no-go task. Intersubject representational similarity analysis (IS-RSA) was used to assess the associations between pairwise similarities in prospectively or retrospectively measured ELS, neural representations in 360 cortical regions, and task performance. We incorporated multidimensional scaling and Procrustes analysis to visualize interindividual differences in neural representational spaces. Prospective ELS—but not Retrospective ELS—was significantly associated with neural representational similarity across 40 cortical regions, including the anterior insula, frontal operculum, and anterior cingulate cortex. These findings highlight the systematic and chronic effects of more moderate ELS on brain development and emphasize the value of prospective measurements and advanced similarity analyses in capturing the nuanced influences of ELS. By integrating spatial and shape analytical techniques, the present study provides new insights into the long-term neurobiological correlates of ELS and introduces novel methodological tools for neurodevelopmental research.

Understanding the presymptomatic biology in those at high risk of developing amyotrophic lateral sclerosis (ALS) is essential for the development of preventative therapeutic interventions. Approximately 10% of ALS is associated with a C9orf72 expansion or pathogenic variants in SOD1. Magnetoencephalography (MEG), combined with machine learning algorithms, can model brain network dynamics in such at-risk populations to develop pathogenic biomarkers. Individuals with symptomatic ALS (symALS, n = 61), asymptomatic C9orf72 carriers (aC9, n = 16), or pathological SOD1 carriers (aSOD, n = 12), and healthy controls (n = 84) underwent resting-state MEG recordings. Extracted metrics included regional oscillatory power, connectivity, and spectral shape. ‘DyNeMo’ was trained to identify six functional dynamic brain networks. Metrics were compared between groups. A classifier was trained to distinguish asymptomatic gene carriers from controls. Compared to controls, beta frequency power was decreased in both symALS and aC9 groups. The aC9 group showed a marked slowing of frontal oscillatory activity, while the aSOD group showed a marked acceleration. Dynamic network coactivation was dramatically disrupted in aC9, more than in both symALS and aSOD. The classifier accurately distinguished genetically at-risk groups from controls (receiver-operator-characteristic area-under-curve 0.89). The cerebral network dynamics of aC9 are markedly different from both aSOD and symALS, supporting the concept of profoundly different upstream pathways in SOD1 ALS, sparing wider cortical pathology when compared to C9orf72 ALS. aC9 changes may reflect chronic adaptive changes relating to neurodevelopmental factors or underpin aspects of system vulnerability that define penetrance variability. MEG metrics might provide important biomarkers of prevention therapy efficacy and phenoconversion in at-risk populations.

Advances in fetal three-dimensional (3D) ultrasound (US) and magnetic resonance imaging (MRI) have revolutionized the study of fetal brain development, enabling detailed analysis of brain structures and growth. Despite their complementary capabilities, these modalities capture fundamentally different physical signals, potentially leading to systematic differences in image-derived phenotypes (IDPs). Here, we evaluate the agreement of IDPs between US and MRI by comparing the volumes of eight brain structures from 90 subjects derived using deep-learning algorithms from majority same-day imaging (days between scans: mean = 1.2, mode = 0 and max = 4). Excellent agreement (intra-class correlation coefficient, $��\mathrm{ICC}�>�0.75�$) was observed for the cerebellum, cavum septum pellucidum, thalamus, white matter and deep grey matter volumes, with significant correlations for most structures, except the ventricular system. Bland–Altman analysis revealed some systematic biases: intracranial and cortical plate volumes were larger on US than MRI, by an average of $��35��{\mathrm{cm}}^3�$ and $��4.1��{\mathrm{cm}}^3�$, respectively. Finally, we found the labels of the brainstem and ventricular system were not comparable between the modalities. These findings highlight the necessity of structure-specific adjustments when interpreting fetal brain IPDs across modalities and underscore the complementary roles of US and MRI in advancing fetal neuroimaging.

We previously reported that a unilateral dorsal column lesion (DCL) at the cervical C4 level primarily reduces inter-horn resting-state functional connectivity (rsFC) measured by functional Magnetic Resonance Imaging (fMRI) in segments below the lesion. This study compares changes in rsFC from fMRI with changes in local field potential (LFP) coherence over an extended post-injury period. High-resolution fMRI and LFP data were acquired bilaterally in healthy monkeys and at 3- and 6-months post-lesion. At 3 months post-injury, tactile-stimulus-evoked LFP power in the dorsal horn was significantly weaker than in the healthy cord and non-lesion side. LFP coherences increased on the lesion side for the dorsal-to-intermediate zone (D-IGM) and dorsal-to-ventral (D-V) pairs but decreased for the non-lesion side D-IGM. By 6 months, stimulus-evoked LFP power on the lesion side remained low. LFP coherences between dorsal-to-dorsal (D-D), ventral-to-ventral (V-V), and D-V pairs on both the lesion and non-lesion sides were significantly reduced relative to the healthy cord. Low-frequency (delta, theta, and alpha) D-IGM coherences on the lesion side, and high-frequency (beta and gamma) coherences on the non-lesion side, were also significantly weakened. Across specific inter-horn pairs and time points, changes in LFP coherences and rsFC measures were weakly correlated. Measurements of inter-horn correlations two segments caudal to the lesion level at C7 revealed distance-dependent intraspinal connectivity changes following DCL. Post-mortem histology confirmed a complete DCL in most animals (7/9). The extent of the disruption of ascending sensory afferents, as assessed histologically, did not appear to correlate with the degree of LFP power reduction or rsFC changes at post-injury time points. In summary, we observed temporally and spatially heterogeneous changes of fMRI correlations and LFP coherences within intraspinal circuits. fMRI rsFC and LFP coherences were not always concordant, with discrepancies depending on specific gray-matter horns and intermediate-zone pairs.

Aversive conditioning changes visuocortical responses to conditioned cues, and the generalization of these changes to perceptually similar cues may provide mechanistic insights into anxiety and fear disorders. Yet, as in many areas of cognitive neuroscience, testing hypotheses about trial-by-trial dynamics in conditioning paradigms is challenged by poor single-trial signal-to-noise ratios (SNR), missing trials, and inter-individual differences. The present technical report demonstrates how a state-of-the-art Bayesian workflow can overcome these issues, using a preliminary sample of simultaneously recorded EEG-fMRI data. A preliminary group of observers (N = 24) viewed circular gratings varying in orientation, with only one orientation paired with an aversive outcome (noxious electric pulse). Gratings were flickered at 15 Hz to evoke steady-state visual evoked potentials (ssVEPs), recorded with 31 channels of EEG in an MRI scanner. First, the benefits of a Bayesian multilevel structure are demonstrated on the fMRI data by improving a standard fMRI first-level multiple regression. Next, the Bayesian modeling approach is demonstrated by applying a theory-driven learning model to the EEG data. The multilevel structure of the Bayesian learning model informs and constrains estimates per participant, providing an interpretable generative model. In the example analysis provided in this report, it showed superior cross-validation accuracy and provided insights into participant-level learning dynamics. It also isolated the generalization effects of conditioning, providing improved statistical certainty. Lastly, missing trials were interpolated and weighted appropriately using the full model's structure. This is a critical aspect for single-trial analyses of simultaneously recorded physiological measures because each added measure will typically increase the number of trials missing a complete set of observations. The present report aims to illustrate the utility of this analytical framework. It shows how models may be iteratively built and compared in a modern Bayesian workflow. Future models may use different conceptualizations of learning, allow integration of clinically relevant factors, and enable the fusion of different simultaneous recordings such as EEG, autonomic, behavioral, and hemodynamic data.

Modeling dynamic neuronal activity within brain networks enables the precise tracking of rapid temporal fluctuations across different brain regions. However, current approaches in computational neuroscience fall short of capturing and representing the spatiotemporal dynamics within each brain network. We developed a novel weakly supervised spatiotemporal dense prediction model capable of generating personalized 4D dynamic brain networks from fMRI data, providing a more granular representation of brain activity over time. We developed a model that leverages the vision transformer (ViT) as its backbone, jointly encoding spatial and temporal information from fMRI inputs using two different configurations: space–time and sequential encoders. The model generates 4D brain network maps that evolve over time, capturing dynamic changes in both spatial and temporal dimensions. In the absence of ground-truth data, we used spatially constrained windowed independent component analysis (ICA) components derived from fMRI data as weak supervision to guide the training process. The model was evaluated using large-scale resting-state fMRI datasets, and statistical analyses were conducted to assess the effectiveness of the generated dynamic maps using various metrics. Our model effectively produced 4D brain maps that captured both inter-subject and temporal variations, offering a dynamic representation of evolving brain networks. Notably, the model demonstrated the ability to produce smooth maps from noisy priors, effectively denoising the resulting brain dynamics. Additionally, statistically significant differences were observed in the temporally averaged brain maps, as well as in the summation of absolute temporal gradient maps, between patients with schizophrenia and healthy controls. For example, within the Default Mode Network (DMN), significant differences emerged in the temporally averaged space–time configurations, particularly in the thalamus, where healthy controls exhibited higher activity levels compared to subjects with schizophrenia. These findings highlight the model's potential for differentiating between clinical populations. The proposed spatiotemporal dense prediction model offers an effective approach for generating dynamic brain maps by capturing significant spatiotemporal variations in brain activity. Leveraging weak supervision through ICA components enables the model to learn dynamic patterns without direct ground-truth data, making it a robust and efficient tool for brain mapping. Significance: This work presents an important new approach for dynamic brain mapping, potentially opening up new opportunities for studying brain dynamics within specific networks. By framing the problem as a spatiotemporal dense prediction task in computer vision, we leverage the spatiotemporal ViT architecture combined with weakly supervised learning techniques to efficiently and effectively estimate these maps.

The role of inflammation in the neuropathology of major depressive disorder (MDD) is evident. However, the neural correlates of peripheral inflammation in MDD and their transcriptomic architecture, neurochemical basis, and behavioral relevance have not been systematically investigated. We adopted functional and diffusion magnetic resonance imaging to assess gray matter function and white matter integrity, whose associations with serum C-reactive protein (CRP) levels were explored in a large sample of MDD patients. Further, we examined the spatial relationships of the identified neural correlates of CRP with transcriptome, neurotransmitter, and behavioral domain atlases. Higher serum CRP levels were associated with local gray matter function alterations and widespread white matter integrity changes in MDD patients, but not HC. Moreover, the gray matter functional correlates of CRP in MDD were spatially correlated with functional gene categories involving inflammatory signaling pathways (macrophage activation, NF-κB signaling, and JUN kinase activity), specific neurotransmitters (serotonin, GABA, and glutamate), and diverse behavioral domains (sensorimotor, cognition, emotion, and sleep). In addition, some neural correlates of CRP (anterior cingulate cortex function and superior corona radiata integrity) mediated the relationships of serum CRP with sustained attention and sleep structure in MDD patients. Our findings may not only confirm the role of inflammation in the neuropathology of MDD, but also inform a novel conceptualization of targeting inflammatory processes to treat this disorder.