International journal of neural systems最新文献

{"title":"Continual Learning by Contrastive Learning of Regularized Classes in Multivariate Gaussian Distributions.","authors":"Hyung-Jun Moon, Sung-Bae Cho","doi":"10.1142/S012906572550025X","DOIUrl":"10.1142/S012906572550025X","url":null,"abstract":"<p><p>Deep neural networks struggle with incremental updates due to catastrophic forgetting, where newly acquired knowledge interferes with the learned previously. Continual learning (CL) methods aim to overcome this limitation by effectively updating the model without losing previous knowledge, but they find it difficult to continuously maintain knowledge about previous tasks, resulting from overlapping stored information. In this paper, we propose a CL method that preserves previous knowledge as multivariate Gaussian distributions by independently storing the model's outputs per class and continually reproducing them for future tasks. We enhance the discriminability between classes and ensure the plasticity for future tasks by exploiting contrastive learning and representation regularization. The class-wise spatial means and covariances, distinguished in the latent space, are stored in memory, where the previous knowledge is effectively preserved and reproduced for incremental tasks. Extensive experiments on benchmark datasets such as CIFAR-10, CIFAR-100, and ImageNet-100 demonstrate that the proposed method achieves accuracies of 93.21%, 77.57%, and 78.15%, respectively, outperforming state-of-the-art CL methods by 2.34 %p, 2.1 %p, and 1.91 %p. Additionally, it achieves the lowest mean forgetting rates across all datasets.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2550025"},"PeriodicalIF":0.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143789446","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}

{"title":"Predicting Intraoperative Burst Suppression Using Preoperative EEG and Patient Characteristics.","authors":"Jingyi He, Joël M H Karel, Marcus L F Janssen, Erik D Gommer, Catherine J Vossen, Enrique Hortal","doi":"10.1142/S0129065725500339","DOIUrl":"10.1142/S0129065725500339","url":null,"abstract":"<p><p>Burst suppression (BS) is an electroencephalogram (EEG) pattern observed in patients undergoing general anesthesia. The occurrence of BS is associated with adverse outcomes such as postoperative delirium, extended recovery time, and increased postoperative mortality. The detection and prediction of BS can help expedite the evaluation of patient conditions, optimize anesthesia administration, and improve patient safety. This study explores the potential for automatic BS detection using intraoperative EEG and BS prediction using preoperative EEG signals and patient characteristics. A dataset comprising 287 patients who underwent carotid endarterectomy procedures at Maastricht University Medical Center+ was analyzed. An EEG toolbox developed by T. Zhan at the Massachusetts Institute of Technology was utilized for the automatic detection/annotation of BS, while five machine learning classifiers were employed to predict BS occurrence using preoperative data. Based on the 160 patients manually annotated by EEG experts (regarding the presence or absence of BS), the automatic detection tool demonstrated an accuracy of 0.75. For the BS prediction task, an initial subset of 120 patients was evaluated, showing modest performance, with the <i>K</i>-nearest neighbors ([Formula: see text]) classifier achieving the best results, with an accuracy of 0.72. Subsequent experiments indicated that increasing the number of patients (by using <i>Zhan's Toolbox</i> to annotate the unlabeled instances), applying SMOTE to balance the training set, and enriching the feature set was beneficial. The final experiment demonstrated a significant improvement, with Random Forest and Gradient Boosting outperforming other classifiers, achieving an accuracy of 0.86 and ROC-AUC of 0.94. Patient characteristics, including type of anesthetic agents, symptoms, age, mean absolute delta power, mean absolute theta power, and cognitive impairment, were identified by an xAI method as important features potentially indicating the predisposition to experience BS.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2550033"},"PeriodicalIF":0.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144033130","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}

{"title":"Directed Weighted EEG Connectogram Insights of One-to-One Causality for Identifying Developmental Dyslexia.","authors":"Ignacio Rodríguez-Rodríguez, José Ignacio Mateo-Trujillo, Andrés Ortiz, Nicolás J Gallego-Molina, Diego Castillo-Barnes, Juan L Luque","doi":"10.1142/S0129065725500327","DOIUrl":"10.1142/S0129065725500327","url":null,"abstract":"<p><p>Developmental dyslexia (DD) affects approximately 5-12% of learners, posing persistent challenges in reading and writing. This study presents a novel electroencephalography (EEG)-based methodology for identifying DD using two auditory stimuli modulated at 4.8[Formula: see text]Hz (prosodic) and 40[Formula: see text]Hz (phonemic). EEG signals were processed to estimate one-to-one Granger causality, yielding directed and weighted connectivity matrices. A novel Mutually Informed Correlation Coefficient (MICC) feature selection method was employed to identify the most relevant causal links, which were visualized using connectograms. Under the 4.8[Formula: see text]Hz stimulus, altered theta-band connectivity between frontal and occipital regions indicated compensatory frontal activation for prosodic processing and visual-auditory integration difficulties, while gamma-band anomalies between occipital and temporal regions suggested impaired visual-prosodic integration. Classification analysis under the 4.8[Formula: see text]Hz stimulus yielded area under the ROC curve (AUC) values of 0.92 (theta) and 0.91 (gamma band). Under the 40[Formula: see text]Hz stimulus, theta abnormalities reflected dysfunctions in integrating auditory phoneme signals with executive and motor regions, and gamma alterations indicated difficulties coordinating visual and auditory inputs for phonological decoding, with AUC values of 0.84 (theta) and 0.89 (gamma). These results support both the Temporal Sampling Framework and the Phonological Core Deficit Hypothesis. Future research should extend the range of stimuli frequencies and include more diverse cohorts to further validate these potential biomarkers.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2550032"},"PeriodicalIF":0.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144002111","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}

{"title":"Interactive EEG Emotion Recognition with Incremental Gaussian Processes.","authors":"Xiangle Ping, Wenhui Huang","doi":"10.1142/S0129065725500418","DOIUrl":"https://doi.org/10.1142/S0129065725500418","url":null,"abstract":"<p><p>Interactivity is crucial for enabling models to adjust and optimize based on user feedback, thereby enhancing overall performance. However, existing electroencephalogram (EEG)-based emotion recognition models rely on static training paradigms, lack interactivity, and struggle to effectively handle uncertainty in predictions. To address this issue, we propose a novel paradigm for interactive emotion recognition based on incremental Gaussian processes (GP). Unlike existing methods, our approach introduces an expert interaction mechanism to correct samples with high predictive uncertainty and incrementally update the model accordingly, thereby optimizing its performance. First, we model the emotion recognition task as a GP-based framework, utilizing the variance of the GP to quantify the model's uncertainty, thereby guiding experts in targeted interactions. Second, within the GP framework, we propose a novel incremental update strategy that allows the GP to incrementally update prediction results and uncertainties based only on new data obtained through expert interactions, without reprocessing all existing data. This effectively overcomes the shortcomings of traditional GP in updating efficiency. Third, to address the high computational complexity of GP, we use a sparse approximation strategy, selecting inducing points and performing variational inference to efficiently approximate the GP posterior, thereby reducing computational complexity. Subject-dependent and subject-independent experiments conducted on the DEAP and DREAMER datasets demonstrate that the proposed method exhibits significant advantages over state-of-the-art (SOTA) methods. In subject-dependent experiments, our method achieved the highest improvement (1.73%) in the Dominance dimension on the DREAMER dataset. In subject-independent experiments, it attained the largest performance improvement (2.96%) in the Arousal dimension on the DEAP dataset. These results further validate the proposed method's effectiveness.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2550041"},"PeriodicalIF":0.0,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144145249","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}

{"title":"Nonlinear Spiking Neural Systems for thermal Image Semantic Segmentation Networks.","authors":"Peng Wang, Minglong He, Hong Peng, Zhicai Liu","doi":"10.1142/S0129065725500388","DOIUrl":"https://doi.org/10.1142/S0129065725500388","url":null,"abstract":"<p><p>Thermal and RGB images exhibit significant differences in information representation, especially in low-light or nighttime environments. Thermal images provide temperature information, complementing the RGB images by restoring details and contextual information. However, the spatial discrepancy between different modalities in RGB-Thermal (RGB-T) semantic segmentation tasks complicates the process of multimodal feature fusion, leading to a loss of spatial contextual information and limited model performance. This paper proposes a channel-space fusion nonlinear spiking neural P system model network (CSPM-SNPNet) to address these challenges. This paper designs a novel color-thermal image fusion module to effectively integrate features from both modalities. During decoding, a nonlinear spiking neural P system is introduced to enhance multi-channel information extraction through the convolution of spiking neural P systems (ConvSNP) operations, fully restoring features learned in the encoder. Experimental results on public datasets MFNet and PST900 demonstrate that CSPM-SNPNet significantly improves segmentation performance. Compared with the existing methods, CSPM-SNPNet achieves a 0.5% improvement in mIOU on MFNet and 1.8% on PST900, showcasing its effectiveness in complex scenes.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2550038"},"PeriodicalIF":0.0,"publicationDate":"2025-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144096588","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}

{"title":"Enhanced Graph Attention Network by Integrating Transformer for Epileptic EEG Identification.","authors":"Zhenhua Xie, Jian Lian, Dong Wang","doi":"10.1142/S0129065725500376","DOIUrl":"https://doi.org/10.1142/S0129065725500376","url":null,"abstract":"<p><p>Electroencephalography signal classification is essential for the diagnosis and monitoring of neurological disorders, with significant implications for patient treatment. Despite the progress made, existing methods face challenges such as capturing the complex dynamics of Electroencephalogram (EEG) signals and generalizing across diverse patient populations. In this study, the graph attention network and the transformer model are integrated for EEG signal classification, leveraging the enhanced capability to dynamically compute attention weights and adapt to the variable relevance of brain regions. The proposed approach is capable of modeling the intricate relationships within EEG activities by learning context-dependent attention scores. We conducted a comprehensive evaluation of the proposed approach comparing with the state-of-the-art algorithms. Experimental outcomes show that it surpasses the competing models. The superior performance is attributed to the proposed approach's dynamic attention mechanism, which better captures the nuanced patterns in EEG signals across different subjects and seizure types. In the experiments, the CHB-MIT dataset was exploited, which served as a benchmark for evaluating the performance of the proposed framework in distinguishing interictal, ictal, and normal EEG patterns. The results prove the usefulness of our work in advancing EEG signal classification. The findings suggest that the combination of graph attention and self-attention mechanisms is a promising approach for improving the accuracy and reliability of EEG-based diagnostics, potentially improving the management of neurological disorders.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2550037"},"PeriodicalIF":0.0,"publicationDate":"2025-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144012151","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}

{"title":"Self-Supervised Image Segmentation Using Meta-Learning and Multi-Backbone Feature Fusion.","authors":"Muhammad Shahroz Ajmal, Guohua Geng, Xiaofeng Wang, Mohsin Ashraf","doi":"10.1142/S0129065725500121","DOIUrl":"10.1142/S0129065725500121","url":null,"abstract":"<p><p>Few-shot segmentation (FSS) aims to reduce the need for manual annotation, which is both expensive and time-consuming. While FSS enhances model generalization to new concepts with only limited test samples, it still relies on a substantial amount of labeled training data for base classes. To address these issues, we propose a multi-backbone few shot segmentation (MBFSS) method. This self-supervised FSS technique utilizes unsupervised saliency for pseudo-labeling, allowing the model to be trained on unlabeled data. In addition, it integrates features from multiple backbones (ResNet, ResNeXt, and PVT v2) to generate a richer feature representation than a single backbone. Through extensive experimentation on PASCAL-5i and COCO-20i, our method achieves 54.3% and 25.1% on one-shot segmentation, exceeding the baseline methods by 13.5% and 4%, respectively. These improvements significantly enhance the model's performance in real-world applications with negligible labeling effort.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2550012"},"PeriodicalIF":0.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143191645","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}

{"title":"Coherence-Based Graph Convolution Network to Assess Brain Reorganization in Spinal Cord Injury Patients.","authors":"Jiancai Leng, Jiaqi Zhao, Yongjian Wu, Chengyan Lv, Zhixiao Lun, Yanzi Li, Chao Zhang, Bin Zhang, Yang Zhang, Fangzhou Xu, Changsong Yi, Tzyy-Ping Jung","doi":"10.1142/S0129065725500212","DOIUrl":"10.1142/S0129065725500212","url":null,"abstract":"<p><p>Motor imagery (MI) engages a broad network of brain regions to imagine a specific action. Investigating the mechanism of brain network reorganization during MI after spinal cord injury (SCI) is crucial because it reflects overall brain activity. Using electroencephalogram (EEG) data from SCI patients, we conducted EEG-based coherence analysis to examine different brain network reorganizations across different frequency bands, from resting to MI. Furthermore, we introduced a consistency calculation-based residual graph convolution (C-ResGCN) classification algorithm. The results show that the [Formula: see text]- and [Formula: see text]-band connectivity weakens, and brain activity decreases during the MI task compared to the resting state. In contrast, the [Formula: see text]-band connectivity increases in motor regions while the default mode network activity declines during MI. Our C-ResGCN algorithm showed excellent performance, achieving a maximum classification accuracy of 96.25%, highlighting its reliability and stability. These findings suggest that brain reorganization in SCI patients reallocates relevant brain resources from the resting state to MI, and effective network reorganization correlates with improved MI performance. This study offers new insights into the mechanisms of MI and potential biomarkers for evaluating rehabilitation outcomes in patients with SCI.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2550021"},"PeriodicalIF":0.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143639865","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}

{"title":"Understanding the Spatio-Temporal Coupling of Spikes and Spindles in Focal Epilepsy Through a Network-Level Computational Model.","authors":"Min Pan, Qiang Li, Jiangling Song, Bo Wang, Wenhua Wang, Rui Zhang","doi":"10.1142/S0129065725500182","DOIUrl":"10.1142/S0129065725500182","url":null,"abstract":"<p><p>The electrophysiological findings have shown that epileptiform spikes triggering sleep spindles within 1[Formula: see text]s across multiple channels are commonly observed during sleep in focal epilepsy (FE). Such spatio-temporal couplings of spikes and spindles (STCSSs) are defined as a kind of pathological waves, and frequent emergence of them may cause the degradation of cognitive function for FE patients. However, the neural mechanisms underlying STCSSs are not well understood. To this end, this work first develops a neural mass network model for focal epilepsy (FE-NMNM) with multiple thalamocortical columns being its nodes and the long-range synaptic interactions of thalamocortical columns being its edges, where each thalamocortical column is extended on the basis of Costa model and then they are connected through excitatory synapses between pyramidal cells. Then, how the cortico-cortical connectivity affects the evolution of STCSSs across the network is especially discussed by simulations in two cases, where the inter-ictal state and the ictal state are considered separately. Simulation results demonstrate that: (1) the more STCSSs occur in a more extensive area when the cortico-cortical connectivity becomes stronger, and the significant increase of coupling discharges is attributed to the presence of abundant spikes; (2) when the connectivity is excessively strong, the cortical hyperexcitability will happen, thereby inducing massive spike discharges which may further inhibit the occurrence of spindles, and hence, resulting in the disappearance of STCSSs. The obtained results provide a mechanistic insight into STCSSs, and suggest that such coupling patterns could reflect widespread network dysfunction in FE, thereby potentially advancing therapeutic strategies for FE.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":" ","pages":"2550018"},"PeriodicalIF":0.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143625844","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}

{"title":"Adaptive Dynamic Surface Control of Epileptor Model Based on Nonlinear Luenberger State Observer.","authors":"Mahdi Kamali Dolatabadi, Marzieh Kamali, Farzaneh Shayegh","doi":"10.1142/S0129065725500224","DOIUrl":"10.1142/S0129065725500224","url":null,"abstract":"<p><p>Epilepsy is a prevalent neurological disorder characterized by recurrent seizures, which are sudden bursts of electrical activity in the brain. The Epileptor model is a computational model specifically created to replicate the complex dynamics of epileptic seizures. The parameters of the Epileptor model can be adjusted to simulate activities associated with some seizure classes seen in patients. Due to the closeness of this model to nonlinear systems with nonstrict feedback form and the existence of uncertainties in the model, an adaptive dynamic surface controller is chosen for control of the system. Considering that the states in the Epileptor model are not measurable and the only measurable output is the Local Field Potentials signal, a nonlinear Luenberger state observer is developed to estimate the system states. It is the first time that the Luenberger state observer is used for the Epileptor model. In this approach, Radial Basis Neural Networks are utilized to estimate the system's nonlinear dynamics. The stability of our proposed controller along with the observer is proved, and the performance is shown using simulation. Simulation results show that by using the suggested method, the output and states of the, system track their reference, value with an acceptable error.</p>","PeriodicalId":94052,"journal":{"name":"International journal of neural systems","volume":"35 5","pages":"2550022"},"PeriodicalIF":0.0,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143766138","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}