Prior-guided individualized thalamic parcellation based on local diffusion characteristics.

Comprising numerous subnuclei, the thalamus intricately interconnects the cortex and subcortex, orchestrating various facets of brain functions. Extracting personalized parcellation patterns for these subnuclei is crucial, as different thalamic nuclei play varying roles in cognition and serve as therapeutic targets for neuromodulation. However, accurately delineating the thalamic nuclei boundary at the individual level is challenging due to intersubject variability. In this study, we proposed a prior-guided parcellation (PG-par) method to achieve robust individualized thalamic parcellation based on a central-boundary prior. We first constructed probabilistic atlas of thalamic nuclei using high-quality diffusion MRI datasets based on the local diffusion characteristics. Subsequently, high-probability voxels in the probabilistic atlas were utilized as prior guidance to train unique multiple classification models for each subject based on a multilayer perceptron. Finally, we employed the trained model to predict the parcellation labels for thalamic voxels and construct individualized thalamic parcellation. Through a test-retest assessment, the proposed prior-guided individualized thalamic parcellation exhibited excellent reproducibility and the capacity to detect individual variability. Compared with group atlas registration and individual clustering parcellation, the proposed PG-par demonstrated superior parcellation performance under different scanning protocols and clinic settings. Furthermore, the prior-guided individualized parcellation exhibited better correspondence with the histological staining atlas. The proposed prior-guided individualized thalamic parcellation method contributes to the personalized modeling of brain parcellation.

Abnormal resting-state functional connectivity in subregions of amygdala in adults and adolescents with major depressive disorder.

BACKGROUND: The different symptoms of major depressive disorder (MDD) in adolescents compared to adults suggested there may be differences in the pathophysiology between adolescents and adults with MDD. However, despite the amygdala being considered critical in the pathophysiology, there was limited knowledge about the commonalities and differences in the resting-state functional connectivity (rsFC) of amygdala subregions in MDD patients of different age groups. METHODS: In the current study, 65 adolescents (46 with MDD and 19 controls) and 91 adults (35 with MDD and 56 controls) were included. A seed-based functional connectivity analysis was performed for each of the amygdala subregions. A 2 × 2 ANOVA was used to analyze the main effect of age, diagnosis, and their interaction on the rsFC of each subregion. RESULTS: A significant main effect of age was revealed in the rsFC of bilateral centromedial (CM) subregions and right laterobasal (LB) subregion with several brain regions in the limbic system and frontoparietal network. The significant main effect of diagnosis showed MDD patients of different ages showed higher connectivity than controls between the right LB and left middle frontal gyrus (MFG). CONCLUSIONS: The rsFC of specific amygdala subregions with brain regions in the limbic system and frontoparietal network is affected by age, indicating a distinct amygdala connectivity profile in adolescents. The decreased rsFC between the right LB and the left MFG in adolescents and adults with MDD could serve as a diagnostic biomarker and a target of nonpharmacological treatment for MDD.

Enhancement, integration, expansion: Activating representation of detailed features for occluded person re-identification.

A proposed method, Enhancement, integration, and Expansion, aims to activate the representation of detailed features for occluded person re-identification. Region and context are two important and complementary features, and integrating them in an occluded environment can effectively improve the robustness of the model. Firstly, a self-enhancement module is designed. Based on the constructed multi-stream architecture, rich and meaningful feature interference is introduced in the feature extraction stage to enhance the model's ability to perceive noise. Next, a collaborative integration module similar to cascading cross-attention is proposed. By studying the intrinsic interaction patterns of regional and contextual features, it adaptively fuses features across streams and enhances the diverse and complete representation of internal information. The module is not only robust to complex occlusions, but also mitigates the feature interference problem due to similar appearances or scenes. Finally, a matching expansion module that enhances feature discriminability and completeness is proposed. Providing more stable and accurate features for recognition. Compared with state-of-the-art methods on two occluded and holistic datasets, the proposed method is proved to be advanced and the effectiveness of the module is proved by extensive ablation studies.

Deformation depth decoupling network for point cloud domain adaptation.

Recently, point cloud domain adaptation (DA) practices have been implemented to improve the generalization ability of deep learning models on point cloud data. However, variations across domains often result in decreased performance of models trained on different distributed data sources. Previous studies have focused on output-level domain alignment to address this challenge. But this approach may increase the amount of errors experienced when aligning different domains, particularly for targets that would otherwise be predicted incorrectly. Therefore, in this study, we propose an input-level discretization-based matching to enhance the generalization ability of DA. Specifically, an efficient geometric deformation depth decoupling network (3DeNet) is implemented to learn the knowledge from the source domain and embed it into an implicit feature space, which facilitates the effective constraint of unsupervised predictions for downstream tasks. Secondly, we demonstrate that the sparsity within the implicit feature space varies between domains, rendering domain differences difficult to support. Consequently, we match sets of neighboring points with different densities and biases by differentiating the adaptive densities. Finally, inter-domain differences are aligned by constraining the loss originating from and between the target domains. We conduct experiments on point cloud DA datasets PointDA-10 and PointSegDA, achieving advanced results (over 1.2% and 1% on average).

Contribution of resting-state functional connectivity of the subgenual anterior cingulate to prediction of antidepressant efficacy in patients with major depressive disorder.

This study investigated how resting-state functional connectivity (rsFC) of the subgenual anterior cingulate cortex (sgACC) predicts antidepressant response in patients with major depressive disorder (MDD). Eighty-seven medication-free MDD patients underwent baseline resting-state functional MRI scans. After 12 weeks of escitalopram treatment, patients were classified into remission depression (RD, n = 42) and nonremission depression (NRD, n = 45) groups. We conducted two analyses: a voxel-wise rsFC analysis using sgACC as a seed to identify group differences, and a prediction model based on the sgACC rsFC map to predict treatment efficacy. Haufe transformation was used to interpret the predictive rsFC features. The RD group showed significantly higher rsFC between the sgACC and regions in the fronto-parietal network (FPN), including the bilateral dorsolateral prefrontal cortex (DLPFC) and bilateral inferior parietal lobule (IPL), compared to the NRD group. These sgACC rsFC measures correlated positively with symptom improvement. Baseline sgACC rsFC also significantly predicted treatment response after 12 weeks, with a mean accuracy of 72.64% (p < 0.001), mean area under the curve of 0.74 (p < 0.001), mean specificity of 0.82, and mean sensitivity of 0.70 in 10-fold cross-validation. The predictive voxels were mainly within the FPN. The rsFC between the sgACC and FPN is a valuable predictor of antidepressant response in MDD patients. These findings enhance our understanding of the neurobiological mechanisms underlying treatment response and could help inform personalized treatment strategies for MDD.

Comparative Analysis of Human-Chimpanzee Divergence in Brain Connectivity and its Genetic Correlates.

Chimpanzees (Pan troglodytes) are humans' closest living relatives, making them the most directly relevant comparison point for understanding human brain evolution. Zeroing in on the differences in brain connectivity between humans and chimpanzees can provide key insights into the specific evolutionary changes that might have occured along the human lineage. However, conducting comparisons of brain connectivity between humans and chimpanzees remains challenging, as cross-species brain atlases established within the same framework are currently lacking. Without the availability of cross-species brain atlases, the region-wise connectivity patterns between humans and chimpanzees cannot be directly compared. To address this gap, we built the first Chimpanzee Brainnetome Atlas (ChimpBNA) by following a well-established connectivity-based parcellation framework. Leveraging this new resource, we found substantial divergence in connectivity patterns across most association cortices, notably in the lateral temporal and dorsolateral prefrontal cortex between the two species. Intriguingly, these patterns significantly deviate from the patterns of cortical expansion observed in humans compared to chimpanzees. Additionally, we identified regions displaying connectional asymmetries that differed between species, likely resulting from evolutionary divergence. Genes associated with these divergent connectivities were found to be enriched in cell types crucial for cortical projection circuits and synapse formation. These genes exhibited more pronounced differences in expression patterns in regions with higher connectivity divergence, suggesting a potential foundation for brain connectivity evolution. Therefore, our study not only provides a fine-scale brain atlas of chimpanzees but also highlights the connectivity divergence between humans and chimpanzees in a more rigorous and comparative manner and suggests potential genetic correlates for the observed divergence in brain connectivity patterns between the two species. This can help us better understand the origins and development of uniquely human cognitive capabilities.

Macaque Brainnetome Atlas: A multifaceted brain map with parcellation, connection, and histology.

The rhesus macaque (Macaca mulatta) is a crucial experimental animal that shares many genetic, brain organizational, and behavioral characteristics with humans. A macaque brain atlas is fundamental to biomedical and evolutionary research. However, even though connectivity is vital for understanding brain functions, a connectivity-based whole-brain atlas of the macaque has not previously been made. In this study, we created a new whole-brain map, the Macaque Brainnetome Atlas (MacBNA), based on the anatomical connectivity profiles provided by high angular and spatial resolution ex vivo diffusion MRI data. The new atlas consists of 248 cortical and 56 subcortical regions as well as their structural and functional connections. The parcellation and the diffusion-based tractography were evaluated with invasive neuronal-tracing and Nissl-stained images. As a demonstrative application, the structural connectivity divergence between macaque and human brains was mapped using the Brainnetome atlases of those two species to uncover the genetic underpinnings of the evolutionary changes in brain structure. The resulting resource includes: (1) the thoroughly delineated Macaque Brainnetome Atlas (MacBNA), (2) regional connectivity profiles, (3) the postmortem high-resolution macaque diffusion and T2-weighted MRI dataset (Brainnetome-8), and (4) multi-contrast MRI, neuronal-tracing, and histological images collected from a single macaque. MacBNA can serve as a common reference frame for mapping multifaceted features across modalities and spatial scales and for integrative investigation and characterization of brain organization and function. Therefore, it will enrich the collaborative resource platform for nonhuman primates and facilitate translational and comparative neuroscience research.