International journal of neural systems最新文献_第5页

A Multimodal Distillation Framework for Sound-Based Sleep Quality Assessment. 基于声音的睡眠质量评估的多模态蒸馏框架。

IF 6.4

International journal of neural systems Pub Date : 2026-10-09 DOI: 10.1142/S0129065727500250

Ken-Ichi Fukui, Haoyu Lu, Takafumi Kato

{"title":"A Multimodal Distillation Framework for Sound-Based Sleep Quality Assessment.","authors":"Ken-Ichi Fukui, Haoyu Lu, Takafumi Kato","doi":"10.1142/S0129065727500250","DOIUrl":"https://doi.org/10.1142/S0129065727500250","url":null,"abstract":"Sleep is essential for health, yet polysomnography (PSG), the clinical gold standard for comprehensive sleep assessment, is intrusive and impractical for routine use. Sleep sounds offer a nonintrusive alternative, but acoustic signals alone may provide insufficient information for reliable assessment. To address this limitation, we propose a multimodal knowledge distillation (MKD) framework for estimating subjective sleep quality, defined in this study as participants' self-reported satisfaction with each night's sleep. The framework uses richer multimodal information only during training and deploys a sound-based student model at inference time. The teacher model combines sleep-stage features, PSG features, subjective factors, sound features, and demographic factors using hierarchical gated variable selection networks (GVSNs). The student model is restricted to sound features and demographic factors, and response-based distillation transfers the teacher's multimodal decision behavior to this practical input setting. Using 198 nights from 101 adults, we evaluate the framework under both night-wise and subject-wise protocols. Distillation improves the student over the nondistilled model in the repeated splits, with the larger observed gain occurring when test participants are unseen during training. Further analyses show that the preferred distillation weight, the teacher's learned modality weights, and the effect of removing each teacher modality differ between the evaluation protocols. In particular, the ablation results show that a modality's effect on standalone teacher performance does not necessarily match its effect on the distilled student's performance. These findings provide preliminary, dataset-specific evidence of multimodal distillation but do not establish clinical utility or the transfer of any specific modality.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2750025"},"PeriodicalIF":6.4,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149139978","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploiting Phase-Sliding Oscillations for Robust Asynchronous Brain-Computer Interface Control. 基于相位滑动振荡的鲁棒异步脑机接口控制。

IF 6.4

International journal of neural systems Pub Date : 2026-10-09 DOI: 10.1142/S0129065727500225

Jiaxin Li, Shuhuan Wen, Jincheng Hu, Qing Wang, Bin Zhang, Qingshu Zhou, Jing Zhao

{"title":"Exploiting Phase-Sliding Oscillations for Robust Asynchronous Brain-Computer Interface Control.","authors":"Jiaxin Li, Shuhuan Wen, Jincheng Hu, Qing Wang, Bin Zhang, Qingshu Zhou, Jing Zhao","doi":"10.1142/S0129065727500225","DOIUrl":"https://doi.org/10.1142/S0129065727500225","url":null,"abstract":"Asynchronous steady-state visual evoked potential (SSVEP)-based brain-computer interfaces (BCIs) show great potential for real-world control, yet their performance is often impaired by phase shifts between ongoing electroencephalogram (EEG) signals and fixed-phase templates. In this study, we observed and validated a phase-sliding oscillation (PSO) phenomenon: when a fixed-phase template is used, phase-sliding trajectories derived from sliding-window sequences with different initial phases exhibit stable periodic oscillations locked to the stimulus frequency. Based on this finding, we propose a PSO-based asynchronous classification (PSO-AC) method. Specifically, EEG signals are reconstructed into sequences with varying initial phases and transformed into stable time-frequency representations via the wavelet synchrosqueezing transform to extract phase-decoupled features. A probabilistic temporal model is then used to characterize temporal dynamics and distinguish control from noncontrol states. Offline experiments (n = 22) showed that PSO-AC achieved a mean control/noncontrol accuracy of 94.2% with a 2[Formula: see text]s data length, outperforming a state-of-the-art baseline. It also maintained superior performance in nine-class classification under different time delays. In online robotic-arm experiments, PSO-AC enabled more stable command triggering and higher control efficiency, reducing the command cost per successful trial from 4.30 to 1.14.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2750022"},"PeriodicalIF":6.4,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149139965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Recurrent Processing Dynamics in Occluded Object Recognition Revealed by Electroencephalography and Deep Neural Networks. 基于脑电图和深度神经网络的遮挡物识别的循环处理动态。

IF 6.4

International journal of neural systems Pub Date : 2026-10-01 Epub Date: 2026-04-29 DOI: 10.1142/S0129065726500358

Ruiqi Li, Zhencai Liu, Shaoting Yan, Lipeng Zhang, Rui Zhang, Mingming Chen, Yuxia Hu, Li Shi

{"title":"Recurrent Processing Dynamics in Occluded Object Recognition Revealed by Electroencephalography and Deep Neural Networks.","authors":"Ruiqi Li, Zhencai Liu, Shaoting Yan, Lipeng Zhang, Rui Zhang, Mingming Chen, Yuxia Hu, Li Shi","doi":"10.1142/S0129065726500358","DOIUrl":"10.1142/S0129065726500358","url":null,"abstract":"The human visual system excels at recognizing occluded objects, yet the temporal dynamics of recurrent processing in this task remain unclear. Using high-temporal-resolution Electroencephalography (EEG), backward masking, and deep neural networks (DNNs), we employed a two-stage paradigm to investigate recurrent processing in occluded object recognition. In Experiment 1, we manipulated occlusion levels and applied multivariate pattern analysis (MVPA) and temporal generalization analysis (TGA) to investigate the neural differences in object recognition across varying degrees of occlusion. In Experiment 2, backward masking was used to dissociate feedforward and recurrent contributions, assessed via representational similarity analysis (RSA). Results revealed a distinct shift in processing mechanisms: While low occlusion primarily relied on a rapid feedforward sweep, higher occlusion necessitated the recruitment of additional processing. Further characterization of this processing based on TGA and RSA under mask conditions revealed a two-stage recurrent process: An early stage (200-300[Formula: see text]ms) associated with low-level features, and a late stage (300-500[Formula: see text]ms) involved mid- and high-level representations, reflecting cross-hierarchical recurrent interactions. The early mask condition disrupted this coordination, highlighting the essential role of recurrent processing. These findings clarify the temporal dynamics of recurrent processing in occluded object recognition and emphasize the critical role of recurrence in achieving robust biological vision.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2650035"},"PeriodicalIF":6.4,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147793103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Primary Visual Cortex Contributes to Color Constancy by Predicting Rather than Discounting the Illuminant. 初级视觉皮层通过预测而不是贬低光源来促进颜色的稳定性。

IF 6.4

International journal of neural systems Pub Date : 2026-10-01 Epub Date: 2026-04-13 DOI: 10.1142/S0129065726500346

Shaobing Gao, Yongjie Li

{"title":"Primary Visual Cortex Contributes to Color Constancy by Predicting Rather than Discounting the Illuminant.","authors":"Shaobing Gao, Yongjie Li","doi":"10.1142/S0129065726500346","DOIUrl":"10.1142/S0129065726500346","url":null,"abstract":"Color constancy (CC) is an important ability of the human visual system to stably perceive the colors of objects despite considerable changes in the color of the light illuminating them. While increasing evidence from the field of neuroscience supports that multiple levels of the visual system contribute to the realization of CC, how the primary visual cortex (V1) plays role in CC is not fully resolved. In specific, double-opponent (DO) neurons in V1 have been thought to contribute to realizing a degree of CC, but the computational mechanism is not clear. This work builds an electrophysiologically based V1 neural model to learn the color of the light source from a natural image dataset with the ground truth illuminants as the labels. Based on the qualitative and quantitative analysis of the responsive properties of the learned model neurons, this work found that both the spatial structures and color weights of the receptive fields of the learned model neurons are quite similar to those of the simple and DO neurons recorded in V1. Computationally, DO cells perform more robustly than the simple cells in V1 for illuminant prediction. Therefore, this work provides computational evidence supporting that V1 DO neurons serve to realize CC by encoding the illuminant, which provides a compelling alternative to the more common hypothesis that V1 contributes to CC by discounting the illuminant using its DO cells. This evidence is expected to not only help resolve the visual mechanisms of CC, but also provide inspiration to develop more effective computer vision algorithms.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2650034"},"PeriodicalIF":6.4,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147679726","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Network-Level Mechanisms of Sustained Recovery from Mental Fatigue Differentially Modulated by Acute Exercise and Rest. 急性运动和休息差异调节精神疲劳持续恢复的网络水平机制。

IF 6.4

International journal of neural systems Pub Date : 2026-10-01 Epub Date: 2026-05-13 DOI: 10.1142/S0129065726500383

Lingyun Gao, Li Zhu, Sujie Wang, Kuijun Wu, Yajing Pang, Rui Zhang, Yu Sun

{"title":"Network-Level Mechanisms of Sustained Recovery from Mental Fatigue Differentially Modulated by Acute Exercise and Rest.","authors":"Lingyun Gao, Li Zhu, Sujie Wang, Kuijun Wu, Yajing Pang, Rui Zhang, Yu Sun","doi":"10.1142/S0129065726500383","DOIUrl":"10.1142/S0129065726500383","url":null,"abstract":"Mental fatigue, a prevalent yet underestimated state, impairs cognitive performance and increases the risk of errors and accidents, creating persistent challenges in occupational and clinical contexts. While rest is commonly used for recovery, its benefits are context-dependent and typically short-lived. Acute aerobic exercise has shown promise in alleviating cognitive impairments, yet the neural mechanisms distinguishing exercise and rest, as well as their recovery effects during interventions, remain unclear. In this study, a within-subject design was employed with three prolonged psychomotor vigilance task (PVT) sessions: one with mid-task exercise, one with passive rest, and one as a no-intervention control. EEG-derived functional networks were constructed and analyzed to characterize immediate, carryover, and recovery effects across task and intervention periods. Both interventions elicited immediate behavioral improvements but induced divergent frequency-specific network reorganization. During task reengagement, exercise exhibited a state of functional integration characterized by reduced [Formula: see text]-band local efficiency and sustained [Formula: see text]-band global efficiency, whereas rest displayed network segregation. Recovery analyses revealed that exercise modulated distributed [Formula: see text]-band connectivity, with discriminative frontal features reliably distinguishing recovery pathways. Despite similar behavioral outcomes between exercise and rest, acute exercise uniquely sustained network integration distinct from the topological segregation observed during rest, highlighting its potential for fatigue regulation under cognitively demanding contexts from a brain network perspective.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2650038"},"PeriodicalIF":6.4,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147936012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An Interpretability Framework for Convolutional Neural Network-Based Electroencephalography Analysis Discovers New Spatial and Spectral Epileptic Biomarkers. 基于卷积神经网络的脑电图分析的可解释性框架发现了新的空间和频谱癫痫生物标志物。

IF 6.4

International journal of neural systems Pub Date : 2026-10-01 Epub Date: 2026-04-18 DOI: 10.1142/S0129065726500334

Vadim V Grubov, Oleg E Karpov, Sergei I Nazarikov, Semen A Kurkin, Nikita P Utyashev, Denis A Andrikov, Alexander E Hramov

{"title":"An Interpretability Framework for Convolutional Neural Network-Based Electroencephalography Analysis Discovers New Spatial and Spectral Epileptic Biomarkers.","authors":"Vadim V Grubov, Oleg E Karpov, Sergei I Nazarikov, Semen A Kurkin, Nikita P Utyashev, Denis A Andrikov, Alexander E Hramov","doi":"10.1142/S0129065726500334","DOIUrl":"10.1142/S0129065726500334","url":null,"abstract":"Deep learning (DL) models, particularly convolutional neural networks (CNNs), have shown promise in automated epileptic seizure detection from electroencephalogram (EEG). However, their \"black-box\" nature limits clinical adoption, as interpretability is critical for trust and validation in medical applications. A novel interpretability method for CNN-based seizure detection models, designed to uncover meaningful spatial and spectral EEG biomarkers, is proposed. The approach combines frequency- and spatial-domain interpretation to provide both global model behavior analysis and local, sample-specific explanations. It also accounts for the task-specific design, neurophysiological grounding and cross-framework validation - concepts often neglected by many state-of-the-art methods. Results are represented as heatmap matrices of feature importance (5 frequency bands * 5 brain regions) with important features determined through statistical testing. Interpretation is based on neurophysiological alignment of these features. The method is validated on three CNN architectures, demonstrating how each leverages distinct frequency bands and brain regions for seizure identification. Global interpretation reveals that the highest-performing model utilizes complementary biomarkers across multiple frequency bands, while local interpretation captures dynamic intra-seizure spectral shifts. The results align with known neurophysiological mechanisms, such as thalamocortical interactions (theta-band) and default mode network suppression (alpha/beta-bands), while also suggesting new biomarkers for seizure detection. The method bridges the gap between DL and clinical EEG analysis, offering a tool for model validation and discovery of electrophysiological signatures in epilepsy.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2650033"},"PeriodicalIF":6.4,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147725213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Dual-Stage Improvement with Domain Adaptation for Cross-Subject Epileptic Seizure Prediction. 跨学科癫痫发作预测的双阶段改进与领域适应。

IF 6.4

International journal of neural systems Pub Date : 2026-10-01 Epub Date: 2026-05-23 DOI: 10.1142/S0129065726500322

Chenchen Cheng, Wanjin Song, Bo You, Haichao Wu, Yan Liu

{"title":"Dual-Stage Improvement with Domain Adaptation for Cross-Subject Epileptic Seizure Prediction.","authors":"Chenchen Cheng, Wanjin Song, Bo You, Haichao Wu, Yan Liu","doi":"10.1142/S0129065726500322","DOIUrl":"10.1142/S0129065726500322","url":null,"abstract":"Electroencephalogram (EEG)-based epileptic seizure prediction is a critical adjunctive treatment strategy for drug-resistant epilepsy. To mitigate cross-subject EEG variability and improve generalization performance, existing methods use domain adaptation (DA) to minimize distribution discrepancy between source and target domains. However, these methods overlook the joint impact of data imbalance and noisy pseudo-labels. This results in biased attention toward imbalanced source samples and degraded understanding of the true target distribution, thereby impairing the model's generalization performance. To address these issues, a cross-subject seizure prediction method based on DA with dual-stage improvement is proposed to integrate data balancing and pseudo-label optimization. Specifically, a global context-aware generative network is introduced to generate preictal samples with global contextual consistency to rectify class imbalance between preictal and interictal samples in the source domain. A common spatial pattern clustering filter is then introduced, which uses a confidence-guided covariance weighting strategy to integrate classification confidence from source-domain labels and target-domain pseudo-labels, optimizing spatial filters to maximize inter-class discrimination. To mitigate the accumulation and propagation of noisy pseudo-labels during training, a dual-filtering mechanism iteratively eliminates them. Experimental evaluation on the CHB-MIT dataset demonstrates the effectiveness of the proposed method in mitigating interpatient domain discrepancies and achieving superior performance over recent leading approaches.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2650032"},"PeriodicalIF":6.4,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148038704","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An Avian-Inspired Computational Model Driven by Texture-Color Prioritization Strategies in Visual Processing. 视觉处理中纹理-颜色优先策略驱动的鸟类启发计算模型。

IF 6.4

International journal of neural systems Pub Date : 2026-09-17 DOI: 10.1142/S0129065727500237

Yanyan Peng, Yonghao Han, Xiaoke Niu, Zhifang Wang, Xiaoteng Zhang, Li Shi, Shan Lu

{"title":"An Avian-Inspired Computational Model Driven by Texture-Color Prioritization Strategies in Visual Processing.","authors":"Yanyan Peng, Yonghao Han, Xiaoke Niu, Zhifang Wang, Xiaoteng Zhang, Li Shi, Shan Lu","doi":"10.1142/S0129065727500237","DOIUrl":"https://doi.org/10.1142/S0129065727500237","url":null,"abstract":"Birds have evolved distinctive visual processing strategies to adapt to complex natural environments. Pigeons can discriminate visual objects using multiple cues, including local texture, color, and shape. However, natural objects contain these cues simultaneously, and it remains unclear how the pigeon visual system balances these visual features when they co-occur. Whether such biological feature-prioritization principles can inform computational recognition models also remains unknown. Here, we combined behavioral delayed matching-to-sample tasks, in vivo electrophysiological recordings from the entopallium, and spectral-coherence-based network analysis to investigate multidimensional visual processing in pigeons. Behavioral and electrophysiological results revealed a stable feature-prioritization pattern dominated by color and texture, with shape playing a subordinate role. Functional connectivity further indicated that network organization became sparser under information-rich conditions, suggesting that pigeons selectively extract nonredundant cues rather than processing all dimensions equally. Guided by these biological findings, the present research developed the Avian-inspired Feature Gating Model (AFGM). By incorporating independent channel gating and sparse regularization, AFGM formalizes this biological feature-weighting strategy. The model achieved a modest performance advantage over the HVE baseline on fine-grained recognition tasks. Together, these results reveal a consistent texture-color prioritization pattern in pigeons and provide a biologically inspired framework for feature-weighted machine vision.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2750023"},"PeriodicalIF":6.4,"publicationDate":"2026-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148993209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Neural Stability Enhancement Through Dynamic Lipschitz Constraints in Generative Adversarial Networks. 生成对抗网络中基于动态Lipschitz约束的神经稳定性增强。

IF 6.4

International journal of neural systems Pub Date : 2026-09-08 DOI: 10.1142/S0129065727500195

Juexin Zhang, Ying Weng, Xueping Zhao, Xiangjian He

{"title":"Neural Stability Enhancement Through Dynamic Lipschitz Constraints in Generative Adversarial Networks.","authors":"Juexin Zhang, Ying Weng, Xueping Zhao, Xiangjian He","doi":"10.1142/S0129065727500195","DOIUrl":"https://doi.org/10.1142/S0129065727500195","url":null,"abstract":"Generative adversarial networks (GANs) are widely used for generative modeling in neural networks, yet their practical use is hindered by training instability and mode collapse. Existing stabilization approaches predominantly rely on enforcing fixed 1-Lipschitz constraints on the discriminator through techniques such as gradient penalty, spectral normalization, and gradient normalization. However, these methods fail to account for the dynamic nature of neural network training, where the appropriate Lipschitz constant target may vary with evolving system dynamics. This work challenges the conventional fixed-constraint paradigm by formalizing how the effective constraint strength and a bound-implied Lipschitz constant target can depend on architecture and training state. We provide theoretical analysis to motivate adaptive Lipschitz constraint (ALC) control and introduce ALC, a modular mechanism that adjusts the Lipschitz constant target rather than introducing another standalone regularization term. Experiments across multiple stable GAN architectures show improvements in generation quality and training stability in several evaluated settings. Our contributions advance the understanding of neural system stability in adversarial learning and provide a practical approach to ALC control in adversarial objectives.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2750019"},"PeriodicalIF":6.4,"publicationDate":"2026-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148922366","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An Interpretable Functional-Dynamic Synaptic Graph Neural Network for Major Depressive Disorder Diagnosis from rs-fMRI. 基于rs-fMRI的可解释功能动态突触图神经网络诊断重度抑郁症。

IF 6.4

International journal of neural systems Pub Date : 2026-09-01 Epub Date: 2026-02-28 DOI: 10.1142/S0129065726500243

Zhihong Chen, Jiayi Peng, Xiaorui Han, Mengfan Wang, Jiang Wu, Xinhua Wei, Zhengze Gong, Dezhong Yao, Li Pu, Hongmei Yan

{"title":"An Interpretable Functional-Dynamic Synaptic Graph Neural Network for Major Depressive Disorder Diagnosis from rs-fMRI.","authors":"Zhihong Chen, Jiayi Peng, Xiaorui Han, Mengfan Wang, Jiang Wu, Xinhua Wei, Zhengze Gong, Dezhong Yao, Li Pu, Hongmei Yan","doi":"10.1142/S0129065726500243","DOIUrl":"10.1142/S0129065726500243","url":null,"abstract":"Major depressive disorder (MDD) is a serious, complex psychiatric condition that affects millions of people worldwide. Early diagnosis and biomarker identification are critical for personalized treatment and effective disease monitoring. While resting-state functional magnetic resonance imaging (rs-fMRI) combined with deep learning has facilitated MDD prediction, existing methods often overlook the dynamic temporal characteristics of blood oxygen level-dependent (BOLD) signals and ignore the strength of inter-regional connections, resulting in brain region updates devoid of biological specificity. To this end, a functional-dynamic synaptic graph neural network (FDSyn-GNN) is proposed, which integrates a bidirectional gated recurrent unit (Bi-GRU) timestamp encoding (BGTE) module for modeling dynamic BOLD signals and a synaptic graph Transformer (SGT) module for connection-aware brain region updates. FDSyn-GNN is validated on two large-scale MDD datasets collected across multiple sites, where it outperforms 12 state-of-the-art (SOTA) baseline methods. In addition, extensive ablation and interpretability analyses highlight its potential for biomarker discovery, offering insights into the neural mechanisms underlying MDD. The code is publicly available at https://github.com/ZHChen-294/FDSyn-GNN.","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2650024"},"PeriodicalIF":6.4,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147319189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0