Brain Topography最新文献

Investigating traumatic brain injuries (TBI) in the developing brain is a challenging task. The superposition of an injury to the normal development trajectory can lead to brain impairments which are not obvious at diagnosis. T1-weighted MRI, acquired routinely post-injury, has the potential to better inform diagnosis, but is limited by qualitative assessment by radiologists. Using T1-weighted volume images, we investigated the use of three-dimensional texture analysis (TA) on regions of the corpus callosum (CC) in children with TBI and typically developing controls (TDCs) in conjunction with analysis of diffusion weighted image (DWI)-derived metrics. Nineteen TDCs and 37 participants with TBI were included in the study. T1 textural metrics were extracted from the splenium, genu and body of the CC and assessed for differences between the groups. Textural skewness was found to be significantly higher in children with TBI than TDCs in the body of the CC (t-test: p < 0.004, effect size: g = 0.91) and significant differences were observed in the genu of the CC (grey level co-occurrence matrix and Grey-level run length matrix, p < 0.004, effect sizes > 0.6). Non-significant reductions in ADC were found between TBI and TDC groups in the body and the splenium of the CC. Interestingly, no differences were found between TDCs and the TBI sample using FA. The results suggest that TA can potentially be used to assess white matter integrity after paediatric TBI.

Alzheimer's disease (AD) is a prevalent neurodegenerative disorder affecting millions worldwide. Electroencephalography (EEG), a non-invasive, cost-effective, and safe diagnostic tool, is widely used for detecting neurological conditions. Existing EEG-based classification methods for AD diagnosis have limitations, particularly in adequately considering causal relationships between channels and implementing optimal feature selection, creating a need for highly interpretable feature screening mechanisms. This study presents EENet-RLA, a framework that integrates dynamical system theory with deep learning for AD classification, validated on the BrainLat EEG dataset. The framework operates in two stages, feature extraction and EEG classification, with the deep learning architecture serving primarily as a feature mapping and representation extractor. The core methodological contribution lies in the causal, stability-driven EEG channel selection strategy based on embedding entropy (EE), which quantifies nonlinear directional interactions between EEG channels. This strategy combines bootstrap resampling, multiple random seeds, and minimum connectivity thresholds to identify reproducible, informative channels under limited sample conditions. For classification, spatial and temporal EEG features are extracted using ResNet and LSTM respectively, then fused via a Multi-Head Attention mechanism to capture discriminative patterns. The proposed approach achieves 98.54% segment-level classification accuracy and perfect individual-level performance, demonstrating the discriminative potential of causality-informed feature selection in small-sample settings. While ensuring high accuracy, the method streamlines the analytical process and demonstrates the feasibility of causal-based EEG channel selection in AD characterization, with potential applicability to studying other neurological conditions with similar signal characteristics.

Migraine is a frequent and debilitating comorbidity in multiple sclerosis (MS). Migraine headache and concomitant symptoms might be just coincidental or due to inflammatory MS activity, which is highly relevant for patients. Headache in general has been shown to be attributed to inflammatory cerebral MS lesions in the central pain matrix. The question whether migraine headache is associated with a different lesion pattern and non-painful migraine symptoms are associated with specific brain lesions sites needs further clarification. This study aimed to assess the presence of specific lesion clusters in patients with MS and comorbid migraine via voxel-based lesion symptom mapping (VLSM). Patients with multiple sclerosis and headache were prospectively identified and included in a university neurological center. As a subgroup study, patients with migraine were identified. Demographic and clinical data were assessed, and lesion volumes calculated. Cerebral lesion sites were correlated voxel-wise with presence and absence of headache using non-parametric permutation tests. A cohort of multiple sclerosis patients served as controls for the VLSM-analysis. 22 multiple sclerosis patients with migraines were included, as well as 92 controls without headache. Clinical characteristics did not differ in both groups. The VLSM-analysis showed associations between migraine and lesion clusters in the left hippocampus and bilateral thalamus. Visual aura was associated with posterior brain lesions, whilst vertigo was related to cerebellar lesions. In patients with sensory disturbances, lesions in the bilateral basal ganglia were found. MS lesions in the left hippocampus and bilateral thalamus were associated with migraine in multiple sclerosis patients. The lesion pattern indicates that migraine in MS may be facilitated by lesions in the CNS pain processing network, hypothetically through disinhibition. Visual aura in migraineurs with MS was associated with posterior, vertigo with cerebellar lesions and sensory disturbances with lesions in the basal ganglia. Hence, our data indicates that different concomitant non-painful migraine symptoms are associated with lesion sites in the related brain regions of cerebral control of the respective neurological functions. Whether MS lesions might alter brain excitability and facilitate cortical spreading depression in migraine aura remains speculative.

In this study, we aimed to investigate the intrinsic brain activity alterations in patients with disorders of consciousness (DOC) using multidimensional resting-state functional magnetic resonance imaging (rs-fMRI) metrics at ultra-high field (7 T) MRI. We enrolled 10 patients with DOC, including those with vegetative state/unresponsive wakefulness syndrome and minimally conscious state, and 11 healthy controls (HCs). We applied various rs-fMRI metrics ranging from neuronal activity to synchronization and coordination of whole-brain activity, including amplitude of low-frequency fluctuation (ALFF), fractional ALFF (fALFF), percent amplitude of fluctuation (PerAF), regional homogeneity (ReHo), and degree centrality (DC). Patients with DOC exhibited distinct brain activity patterns compared to HCs. The bilateral inferior temporal gyri showed enhanced activity across various metrics (right: ALFF, ReHo, DC; left: ALFF, fALFF, ReHo), while the right precuneus showed decreased activity in patients with DOC (ALFF, DC, PerAF), compared to HCs. Although an initial inverse relationship was observed between the left putamen and CRS-R total scores in DOC patients, this association did not survive multiple comparisons correction (Bonferroni-adjusted threshold: p < 0.0019). Our findings provide new insights into the neural mechanisms underlying DOC, highlighting the importance of the right precuneus and the bilateral inferior temporal gyri in consciousness level. These results can inform the development of diagnostic and therapeutic strategies for DOC.

To investigate a non-invasive magnetic resonance imaging (MRI)-based method for detecting amyloid-β (Aβ) protein deposition in different brain regions of patients with mild cognitive impairment (MCI) and Alzheimer's disease (AD). This study included 80 patients with MCI and 62 patients with AD, who were randomly divided into training and testing sets at an 8:2 ratio. All participants underwent 18 F-florbetapir positron emission tomography (PET) imaging and three-dimensional T1-weighted MRI. The interval between MRI and PET examinations did not exceed 30 days. A deep learning-based three-dimensional VB-Net model was developed for brain region segmentation. All PET images were registered to the corresponding MRI images, and standardized uptake ratios for 109 brain regions were calculated and averaged. Following radiomics feature extraction and selection using multiple methods, six machine learning algorithms were applied to establish regression models. In addition, a lightweight transformer-based deep learning model was constructed by improving the original transformer architecture. A total of 1,409 features were extracted from each brain region in patients with MCI and AD. After feature selection, 46, 16, 47, 59, 17, and 72 features were retained for the construction of stochastic gradient regression (SGR), GBR, random forest regression (RFR), support vector regression (SVR), extreme gradient boosting (XGB), and k-nearest neighbor (KNN) models, respectively. Delong test analysis demonstrated that the RFR model achieved the best performance, with mean absolute error (MAE), mean squared error (MSE), R² score (RS), and Pearson correlation coefficient (PCC) values of 0.13 ± 0.05, 0.03 ± 0.02, 0.77 ± 0.22, and 0.89 ± 0.05 in the training set, and 0.23 ± 0.10, 0.09 ± 0.08, 0.36 ± 0.12, and 0.65 ± 0.09 in the testing set, respectively. For the deep learning model, the MAE, MSE, RS, and PCC in the testing set were 0.41 ± 0.17, 0.25 ± 0.18, - 0.83 ± 0.42, and - 0.01 ± 0.17, respectively. An artificial intelligence-based approach was successfully developed to quantitatively detect Aβ protein accumulation in different brain regions of patients with AD and MCI using MRI. This method is convenient and non-invasive and does not require cerebrospinal fluid puncture or exposure to ionizing radiation.

Sleep critically influences socio-emotional functioning during interpersonal interactions; however, the relationship between poor sleep quality and susceptibility to social exclusion remains unclear. This study aimed to investigate this relationship and its underlying neural mechanisms. A total of 147 healthy sleepers (HS) and 105 individuals with poor sleep quality (PS) completed a social exclusion imagery task, followed by resting-state functional magnetic resonance imaging (fMRI). Negative feelings and reaction times during the task, as well as seed-based functional connectivity (FC) of the left ventral anterior cingulate cortex (vACC) and left inferior frontal gyrus (IFG), were compared between groups. Associations between FC showing group differences and behavioral measures were further examined. After controlling for depressive and anxiety symptoms, the PS group exhibited stronger negative feelings during the task and longer reaction times in neutral conditions. Seed-based FC analysis revealed increased connectivity between the left IFG and left temporal lobe (TL), alongside decreased connectivity between the left IFG and right precentral gyrus (PG) in the PS compared to the HS group. Moreover, FC between the IFG and PG was negatively correlated with negative affect in HS but not in PS. Poor sleep quality is associated with heightened susceptibility to social exclusion, potentially linked to altered functional connectivity between the IFG and PG. These findings underscore the protective role of healthy sleep in social functioning and suggest neural targets for interventions aimed at mitigating social impairments in individuals with poor sleep.

Panic disorder (PD) and generalized anxiety disorder (GAD) are among the most prevalent anxiety disorders (ADs), yet their neural mechanisms remain unclear. This study aimed to characterize EEG microstate patterns and their functional connectivity (FC) in patients with GAD and PD to explore the neural mechanisms underlying anxiety symptoms. Resting-state EEG was collected from 35 patients with PD, 31 patients with GAD, and 39 healthy controls (HCs). Four microstate classes (A-D) were selected to calculate the parameters, including the mean duration, time coverage, occurrence, mean global field power (GFP), and transitions. Furthermore, the FC patterns underlying each microstate class were analyzed. Correlation analyses were performed between anxiety symptoms and microstate metrics. Compared with HCs, ADs presented increased duration of microstate D and decreased time coverage of microstate A, suggesting altered neural dynamics in ADs, characterized by impaired sensory processing and executive functioning.The correlation analysis revealed that the features of microstate C (associated with self-referential processing) were positively correlated with anxiety symptoms. In contrast, the features of microstates A and B (involved in sensory network functioning) showed consistent negative correlations with anxiety symptoms. Furthermore, PD and GAD groups exhibited distinct FC patterns within microstate A. These FC differences in microstate A demonstrated potential value in distinguishing between GAD and PD.

To explore the relationship between alterations in cerebral white matter microstructure and cerebral blood perfusion underlying cognitive decline in patients with type 2 diabetes mellitus (T2DM). This cross-sectional study enrolled 47 T2DM patients (26 with mild cognitive impairment [T2DM-MCI] and 21 without [T2DM-nMCI]) and 23 healthy controls. All participants underwent multi-post-labeling delay arterial spin labeling and diffusion tensor imaging to assess cerebral blood flow (CBF) and white matter integrity. Group differences in imaging parameters and their correlations with cognitive scores were analyzed. Mediation analysis explored pathways between fractional anisotropy (FA), CBF, and cognition. The T2DM-MCI group showed significantly reduced CBF in the bilateral frontal lobes and lower FA in multiple tracts (e.g., superior/inferior longitudinal fasciculus, cingulum, corpus callosum) compared to both control groups (all P < 0.05). MoCA scores positively correlated with FA in several tracts and right frontal CBF. Crucially, mediation analysis revealed that cerebral hypoperfusion accounted for 24.83% of the effect of white matter damage on MCI (β = 0.016, 95% CI: 0.000-0.040). T2DM-MCI is characterized by co-occurring white matter microstructural damage and cerebral hypoperfusion. Our findings identify cerebral hypoperfusion as a significant mediator linking white matter injury to cognitive impairment, providing new mechanistic insights into diabetic cognitive decline.

Electroencephalography (EEG) is widely used in both research and clinical settings, yet its accuracy can be significantly impacted by subject-specific anatomical anomalies such as brain lesions and skull defects. This study investigates the effects of glioma-related brain lesions and craniotomy-induced bone discontinuities on scalp-recorded EEG signals. To do this, single- and multi-source simulations were created using individualized forward models with and without these structural anomalies. We assessed changes in signal amplitude and topography, and identified the most affected electrodes. Furthermore, real EEG recordings were also analyzed longitudinally to evaluate how these anomalies influence the topography and source localization of early auditory evoked responses (P1 and N1 ERP components). Both single- and multi-source simulations showed that the distortions in the EEG signals depend on the location of the neural source in relation to the location of the lesion. Electrode-level analyses showed that these distortions were most pronounced at the electrodes near the bone flap, and thus near the lesions. Real ERP data supported these findings: a subject with a lesion near the auditory cortex showed notable topographic deviations longitudinally for the P1 and N1 ERP components, while a subject with a frontal lesion showed minimal changes in the scalp EEG. These results highlight the need to include detailed brain and skull anatomy in EEG models, especially in studies that track longitudinal changes in clinical populations.