Frontiers in Neuroscience最新文献

Computing with coupled oscillators or oscillatory neural networks (ONNs) has recently attracted a lot of interest due to their potential for massive parallelism and energy-efficient computing. However, to date, ONNs have primarily been explored either analytically or through analog circuit implementations. This paper shifts the focus to the digital implementation of ONNs, examining various design architectures. We first report on an existing digital ONN design based on a recurrent architecture. The major challenge for scaling such recurrent architectures is the quadratic increase in coupling hardware with the network size. To overcome this challenge, we introduce a novel hybrid architecture that balances serialization and parallelism in the coupling elements that shows near-linear hardware scaling, on the order of about 1.2 with the network size. Furthermore, we evaluate the benefits and costs of these different digital ONN architectures in terms of time to solution and resource usage on field programmable gate array (FPGA) emulation. The proposed hybrid architecture allows for a 10.5 × increase in the number of oscillators while using 5-bits to represent the coupling weights and 4-bits to represent the oscillator phase on a Zynq-7020 FPGA board. The near-linear scaling is a major step toward implementing large scale ONN architectures. To the best of our knowledge, this work presents the largest fully connected digital ONN architecture implemented thus far with a total of 506 fully connected oscillators.

Background: Deep brain stimulation (DBS) of the subthalamic nucleus (STN) is an effective treatment for Parkinson's disease (PD) patients. However, postoperative outcomes vary with no reliable predictive method.

Methods: Our study involves 43 PD patients undergoing STN-DBS. Preoperative resting-state functional magnetic resonance imagings (rs-fMRI) were collected. The volume of tissue activated (VTA) was defined based on contact points and stimulation parameters. A model of the cortico-basal ganglia network was established using dynamic causal modeling. The correlation between the UPDRS-III and the network edges was determined through Pearson correlation analysis. Furthermore, a generalized linear model was employed to predict the post-DBS motor improvement.

Results: Individual STN-VTA intersections were found to be important to UPDRS-III improvement induced by DBS (R = 0.59, P = 0.001). STN-VTA intersections were related to the thalamic-primary motor cortex (M1) (R = 0.47, P = 0.005), and M1-STN (R = 0.40, P = 0.006) coupling strength. The coupling strength of Thal-M1 (R = 0.442, P = 0.009) and M1-STN (R = 0.481 P = 0.004) resulted in DBS-induced movement enhancement, particularly rigidity. The strength of effective connections within the STN-Thal-M1 pathway was found to predict improvements in UPDRS-III scores (P = 0.003).

Conclusion: Our study confirmed the relationship between clinical improvements in STN-DBS and target location as well as the stimulation parameters. By constructing personalized cortical-basal ganglia network models based on target location as well as the stimulation parameters, we discovered that the effective connection strength in STN-THA-M1 can predict motor improvement in PD patients undergoing STN-DBS.

Objective: To investigate the therapeutic mechanism of electroacupuncture (EA) in a mouse model of Parkinson's disease (PD) induced by 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP).

Methods: Motor functions were evaluated using open field test and pole tests. Tyrosine hydroxylase (TH) expression in the substantia nigra and striatum was detected by immunohistochemistry. Intestinal barrier integrity was assessed via immunofluorescence staining of tight junction proteins ZO-1 and Occludin. Gut microbiota composition was analyzed by 16S rRNA sequencing.

Results: EA treatment significantly improved motor deficits, restored TH expression in nigrostriatal regions, and enhanced colonic ZO-1 and Occludin levels. EA reversed MPTP-induced dysbiosis, notably normalizing the abundances of Dubosiella, Lactobacillus, Enterococcus, Desulfovibrio, Bacteroides, Allobaculum, and Parasutterella. Microbial co-occurrence network analysis revealed that EA simplified hyperconnected interactions and improved network stability.

Conclusion: EA treatment attenuated PD progression, which was associated with the remodeling of gut microbiota structure and restoration of microbial network stability. The concomitant protection of dopaminergic function suggests a potential link mediated by the gut-brain axis.

Functional near-infrared spectroscopy (fNIRS) offers a non-invasive method to monitor cerebral oxygenation and hemodynamic changes in real time. Transient hemodynamic instability during carotid artery stenting (CAS) may compromise blood-brain barrier (BBB) integrity and cerebral homeostasis, highlighting the need for sensitive intraoperative monitoring. Compared with traditional dual-channel NIRS systems limited to the frontal region, multi-channel fNIRS enables spatially resolved assessment of cortical hemodynamics and may provide a valuable adjunct for cerebroprotection research. We retrospectively analyzed five consecutive patients who underwent CAS under local anesthesia, with continuous multi-channel fNIRS monitoring before, during, and after the procedure. Forty-five cortical channels covering bilateral anterior circulation were recorded. Changes in oxyhemoglobin concentration (Δ[OxyHb]) and its standard deviation around key procedural events were analyzed to evaluate cortical oxygenation dynamics. All five procedures were technically successful, reducing residual stenosis to less than 30%. fNIRS consistently revealed increased cortical oxygenation following angioplasty and stent deployment, indicating improved cerebral perfusion. In four patients, transient fluctuations in cortical oxygenation corresponded to perioperative hemodynamic instability, such as bradycardia or hypotension induced by carotid sinus reflex. In one patient, marked declines in cortical oxygenation preceded transient neurological deficits, with recovery parallel to blood pressure normalization. These findings suggest that multi-channel fNIRS can sensitively capture both global and regional alterations in cerebral oxygenation during CAS, providing real-time insight into perfusion dynamics potentially linked to BBB function and cerebroprotection. Future studies integrating fNIRS with BBB-targeted markers may help refine intraoperative neuroprotection strategies in stroke and vascular interventions.

Background: Hemorrhagic transformation (HT) is a severe complication following acute ischemic stroke, associated with neurological deterioration and poor clinical outcomes. Deep learning represents a promising tool for HT prediction.

Methods: We conducted a retrospective analysis of 474 acute ischemic stroke cases (231 HT and 243 non-HT) admitted to Beijing Tiantan Hospital from April 2014 to November 2022. We constructed a dataset from this cohort and randomly partitioned it into training and validation sets. Subsequently, we developed a model utilizing convolutional neural networks (CNNs) and residual networks based on computed tomography (CT) scans to predict HT after ischemic stroke.

Results: The final dataset consisted of 613 CT scans. The model achieved an F1 score of 78.94% (95% CI, 67.7-86.4). The Area Under the Curve (AUC) was 0.842 (95% CI, 75.8-92.1), sensitivity was 71.55% (95% CI, 60.6%-85.0%), and accuracy was 74.52% (95% CI, 63.9%-83.2%).

Conclusion: By combining plain CT scans with deep learning methodologies, we developed a clinically applicable model with demonstrable interpretability. Primarily designed to predict HT after acute ischemic stroke, this model demonstrated significant performance advantages in testing compared to both clinical physicians and similar existing models.

Introduction: Being able to recognize the emotions in others is fundamental to social interaction, yet the precise temporal dynamics by which the brain integrates contextual cues with facial expressions remain unclear. This study used behavioral measures and event-related potentials (ERPs) to investigate how contextual congruency and emotional valence modulate facial emotion recognition in a neurotypical population.

Methods: Participants viewed emotional faces preceded by either congruent or incongruent bimodal cues, combining vocalizations and visual images.

Results: Behaviorally, participants responded faster and made fewer errors during congruent trials than in incongruent trials, indicating that context facilitates emotional processing. At the neural level, incongruent cues elicited a significantly larger P1 component, suggesting that the brain allocates increased early attentional resources to conflicting stimuli. Furthermore, the P3 component was significantly larger for negative stimuli compared to neutral ones, highlighting the role of emotional valence in later stages of cognitive processing.

Discussion: Together, these findings support a multi-stage model of emotional integration, where contextual incongruency impacts processing from early perceptual encoding to later cognitive evaluation. By integrating behavioral and neural evidence, this study clarifies the temporal course of contextual integration in multisensory emotion perception and provides new insights with implications for clinical and applied research.

The field of genetics has yet to elucidate the complex genetic underpinnings that influence sleep quality. Previous studies have conducted genome-wide association studies (GWAS) on different dimensions of sleep health, but have not directly analyzed the multivariate genetic structure of poor sleep quality (PSQ). To address this knowledge gap, we employed a multifaceted approach that incorporated Genomic Structural Equation Modeling (Genomic-SEM) and multiple Post-GWAS methods. This strategy enabled us to identify causal single nucleotide polymorphisms (SNPs) that contribute to the variability in poor sleep quality. Our study identified a total of 14 leading SNP loci (such as rs2820309) and 3 fine-mapping significant loci (such as KTN1: rs77168063). To further investigate the underlying mechanisms, we employed multiple whole-transcriptome association methods. These methods analyzed susceptible gene signal loci that exhibited strong correlation with poor sleep quality, as determined by tissue, cell layer, and genome component analysis, along with related component information. Subsequently, data on approximately 13,000 common diseases were evaluated to determine the associated predisposing factors for poor sleep quality, and the correlation between poor sleep quality and 20 common neurological diseases was assessed. Additionally, we utilized a polygenic score based on summary data to analyze evidence of risk for poor sleep quality across different chromosomes. This study offers a novel perspective on the genetic underpinnings of poor sleep quality by conducting a genome-wide association study for a phenotype that was not directly measured.

Introduction: Myeloproliferative neoplasms (MPN) may contribute to cerebrovascular disease via cellular and endothelial pathology leading to impairment at the neurovascular unit (NVU) level. Studies targeting this patient cohort form a neuroscientific viewpoint are scarce.

Objective: We aimed at elucidating possible neuroimaging correlates of NVU alterations in MPNs patients.

Materials and methods: We initially included 187 patients with MPNs in this study, retaining 39 patients as per eligibility criteria (25.6% males, median age - 43 years), who were matched with a control group of 11 healthy subjects (36.4% males, median age - 41 years). Structural and task-based (motor paradigm) functional MRI were performed in both groups, along with the evaluation of baseline blood parameters (hemoglobin, hematocrit and platelet count), comorbidities (arterial hypertension, diabetes mellitus, atherosclerosis) and antiplatelet use: these factors were then used as covariates in statistical analysis.

Results: fMRI data analysis in the group of MPN patients revealed activation in the left primary sensorimotor cortex (pre- and post-central gyri); the right supramarginal gyrus showed significant activation (T = 5.99, p_FWEcorr = 0.015) in the MPN group only. Group fMRI data analysis in healthy volunteers showed two main clusters of activation in the left precentral gyrus and right hemisphere of the cerebellum during task execution. Second-level analysis of activation differences between MPN patients and healthy volunteers showed greater activation in the right primary sensorimotor cortex in MPN (Puncorr = 0.014 and <0.001 at cluster and peak level respectively).

Conclusion: Additional task-specific cortical activation in MPN patients may be potentially linked to NVU disturbance, even in otherwise unchanged cerebral activation patterns. Our findings also suggest that fMRI data in MPN may be confounded by higher blood cell count that needs to be controlled for in this cohort of patients.

This study characterizes the requirements of the kayak (kay) gene in Drosophila melanogaster adult eye biology by examining mutant phenotypes in photoreceptor development, external eye morphology, corneal and bristle ultrastructure, and visually guided behaviors, such as phototaxis, in kay strong loss-of-function homozygous mutant clones. Despite previous studies on kay, there is a dearth of phenotypic characterization of the morphological and behavioral consequences of kay loss-of-function alleles in the adult eye. We find that kay is expressed in developing ommatidia in eye discs. The kay mutant ommatidia are misaligned, lack photoreceptors, have malformed corneal surfaces, and have misshaped, misplaced, and fewer mechanosensory bristles. Corneal nipples, while present in mutant corneas on the corneal surface, are disorganized and malformed. With an average of 30% of the eye territory mutant, flies have a significantly lower response in a behavioral phototaxis assay. Altogether, kay function is required for multiple cell types in the adult retina, and this stands in stark contrast with other jun kinase genes, like the fly homologs of jun kinase and jun, genes not required for adult eye morphogenesis. This is consistent with Kayak functions that are independent of heterodimerizing with Jun proteins or requiring activation of the jun kinase pathway.

Neurodevelopmental conditions such as Attention-Deficit/Hyperactivity Disorder (ADHD) are usually framed as brain-based disorders driven by genetics and neurotransmitter imbalance. At the same time, converging evidence implicates the gut-brain axis and intestinal immunity in shaping cognition and behavior. In this Hypothesis and Theory article, I propose that a subset of ADHD and related neurodivergent profiles can be usefully conceptualized as neurointestinal syndromes, emerging from co-evolutionary interactions between the gut microbiota, intestinal parasites, and host immunity. Drawing on data from ADHD, autism spectrum conditions, and migraine, I synthesize evidence for altered microbiota, increased intestinal permeability, and low-grade inflammation in neurodivergent individuals, and discuss how these changes may bias tryptophan metabolism, vagal signaling, and large-scale brain networks. I then explore a speculative evolutionary scenario in which recurrent helminth exposure, historically ubiquitous, acted as a long-term ecological force shaping gut architecture, immunoregulation, and stress responsivity. Chronic parasitic pressure, combined with microbial metabolites and epigenetic imprinting, may have contributed to the emergence of attentional profiles characterized by hypervigilance, novelty seeking, and rapid switching-traits that could have been advantageous in ancestral, pathogen-rich environments but are often maladaptive in modern settings. This framework does not romanticize ADHD nor deny its frequent clinical burden. Rather, it reframes some ADHD phenotypes as possible mismatch syndromes involving the gut-brain axis, generated when an evolutionarily tuned intestinal and immune architecture is placed in sanitized, post-industrial ecologies. Clinically, this perspective supports continued use of established CNS-targeted treatments while motivating complementary research into microbial, barrier, and vagal interventions as potential adjuncts for carefully defined ADHD subgroups.