工程技术最新文献

{"title":"Convolutional autoencoder-based deep learning for intracerebral hemorrhage classification using brain CT images.","authors":"B Nageswara Rao, U Rajendra Acharya, Ru-San Tan, Pratyusa Dash, Manoranjan Mohapatra, Sukanta Sabut","doi":"10.1007/s11571-025-10259-5","DOIUrl":"10.1007/s11571-025-10259-5","url":null,"abstract":"<p><p>Intracerebral haemorrhage (ICH) is a common form of stroke that affects millions of people worldwide. The incidence is associated with a high rate of mortality and morbidity. Accurate diagnosis using brain non-contrast computed tomography (NCCT) is crucial for decision-making on potentially life-saving surgery. Limited access to expert readers and inter-observer variability imposes barriers to timeous and accurate ICH diagnosis. We proposed a hybrid deep learning model for automated ICH diagnosis using NCCT images, which comprises a convolutional autoencoder (CAE) to extract features with reduced data dimensionality and a dense neural network (DNN) for classification. In order to ensure that the model generalizes to new data, we trained it using tenfold cross-validation and holdout methods. Principal component analysis (PCA) based dimensionality reduction and classification is systematically implemented for comparison. The study dataset comprises 1645 (\"ICH\" class) and 1648 (\"Normal\" class belongs to patients with non-hemorrhagic stroke) labelled images obtained from 108 patients, who had undergone CT examination on a 64-slice computed tomography scanner at Kalinga Institute of Medical Sciences between 2020 and 2023. Our developed CAE-DNN hybrid model attained 99.84% accuracy, 99.69% sensitivity, 100% specificity, 100% precision, and 99.84% F1-score, which outperformed the comparator PCA-DNN model as well as the published results in the literature. In addition, using saliency maps, our CAE-DNN model can highlight areas on the images that are closely correlated with regions of ICH, which have been manually contoured by expert readers. The CAE-DNN model demonstrates the proof-of-concept for accurate ICH detection and localization, which can potentially be implemented to prioritize the treatment using NCCT images in clinical settings.</p>","PeriodicalId":10500,"journal":{"name":"Cognitive Neurodynamics","volume":"19 1","pages":"77"},"PeriodicalIF":3.1,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12089006/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144119113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Astrocytic signatures in neuronal activity: a machine learning-based identification approach.","authors":"João Pedro Pirola, Paige DeForest, Paulo R Protachevicz, Laura Fontenas, Ricardo F Ferreira, Rodrigo F O Pena","doi":"10.1007/s11571-025-10276-4","DOIUrl":"10.1007/s11571-025-10276-4","url":null,"abstract":"<p><p>This study investigates the expanding role of astrocytes, the predominant glial cells, in brain function, focusing on whether and how their presence influences neuronal network activity. We focus on particular network activities identified as synchronous and asynchronous. Using computational modeling to generate synthetic data, we examine these network states and find that astrocytes significantly affect synaptic communication, mainly in synchronous states. We use different methods of extracting data from a network and compare which is best for identifying glial cells, with mean firing rate emerging with higher accuracy. To reach the aforementioned conclusions, we applied various machine learning techniques, including Decision Trees, Random Forests, Bagging, Gradient Boosting, and Feedforward Neural Networks, the latter outperforming other models. Our findings reveal that glial cells play a crucial role in modulating synaptic activity, especially in synchronous networks, highlighting potential avenues for their detection with machine learning models through experimental accessible measures.</p><p><strong>Supplementary information: </strong>The online version contains supplementary material available at 10.1007/s11571-025-10276-4.</p>","PeriodicalId":10500,"journal":{"name":"Cognitive Neurodynamics","volume":"19 1","pages":"89"},"PeriodicalIF":3.1,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12158901/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144301260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TCANet: a temporal convolutional attention network for motor imagery EEG decoding.","authors":"Wei Zhao, Haodong Lu, Baocan Zhang, Xinwang Zheng, Wenfeng Wang, Haifeng Zhou","doi":"10.1007/s11571-025-10275-5","DOIUrl":"10.1007/s11571-025-10275-5","url":null,"abstract":"<p><p>Decoding motor imagery electroencephalogram (MI-EEG) signals is fundamental to the development of brain-computer interface (BCI) systems. However, robust decoding remains a challenge due to the inherent complexity and variability of MI-EEG signals. This study proposes the Temporal Convolutional Attention Network (TCANet), a novel end-to-end model that hierarchically captures spatiotemporal dependencies by progressively integrating local, fused, and global features. Specifically, TCANet employs a multi-scale convolutional module to extract local spatiotemporal representations across multiple temporal resolutions. A temporal convolutional module then fuses and compresses these multi-scale features while modeling both short- and long-term dependencies. Subsequently, a stacked multi-head self-attention mechanism refines the global representations, followed by a fully connected layer that performs MI-EEG classification. The proposed model was systematically evaluated on the BCI IV-2a and IV-2b datasets under both subject-dependent and subject-independent settings. In subject-dependent classification, TCANet achieved accuracies of 83.06% and 88.52% on BCI IV-2a and IV-2b respectively, with corresponding Kappa values of 0.7742 and 0.7703, outperforming multiple representative baselines. In the more challenging subject-independent setting, TCANet achieved competitive performance on IV-2a and demonstrated potential for improvement on IV-2b. The code is available at https://github.com/snailpt/TCANet.</p>","PeriodicalId":10500,"journal":{"name":"Cognitive Neurodynamics","volume":"19 1","pages":"91"},"PeriodicalIF":3.1,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12167204/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144309661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Visual statistical learning based on a coupled shape-position recurrent neural network model.","authors":"Baolong Sun, Yihong Wang, Xuying Xu, Xiaochuan Pan","doi":"10.1007/s11571-025-10285-3","DOIUrl":"10.1007/s11571-025-10285-3","url":null,"abstract":"<p><p>The visual system has the ability to learn the statistical regularities (temporal and/or spatial) that characterize the visual scene automatically and implicitly. This ability is referred to as the visual statistical learning (VSL). The VSL could group several objects that have fixed statistical properties into a chunk. This complex process relies on the collaborative involvement of multiple brain regions that work together to learn the chunk. Although behavioral experiments have explored cognitive functions of the VSL, its computational mechanisms remain poorly understood. To address this issue, this study proposes a coupled shape-position recurrent neural network model based on the anatomical structure of the visual system to explain how chunk information is learned and represented in neural networks. The model comprises three core modules: the position network, which encodes object position information; the shape network, which encodes object shape information; and the decision network, which integrates the neuronal activity in the position and shape networks to make decisions. The model successfully simulates the results of a classic spatial VSL experiment. The distribution of neural firing rates in the decision network shows a significant difference between chunk and non-chunk conditions. Specifically, these neurons in the chunk condition exhibit stronger firing rates than those in the non-chunk condition. Furthermore, after the model learns a scene containing both chunk and non-chunk stimuli, neurons in the position network selectively encode far and near stimuli, respectively. In contrast, neurons in the shape network distinguish between chunk and non-chunk. The chunk encoding neurons selectively respond to specific chunks. These results indicate that the proposed model is able to learn spatial regularities of the stimuli to discriminate chunks from non-chunks, and neurons in the shape network selectively respond to chuck and non-chunk information. These findings offer important theoretical insights into the representation mechanisms of chunk information in neural networks and propose a new framework for modeling spatial VSL.</p>","PeriodicalId":10500,"journal":{"name":"Cognitive Neurodynamics","volume":"19 1","pages":"96"},"PeriodicalIF":3.1,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12174023/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144332590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Emotion recognition framework based on adaptive window selection and CA-KAN.","authors":"Xuefen Lin, Linhui Fan, Yifan Gu, Zhixian Wu","doi":"10.1007/s11571-025-10283-5","DOIUrl":"10.1007/s11571-025-10283-5","url":null,"abstract":"<p><p>In recent years, emotion recognition, particularly EEG-based emotion recognition, has found widespread application across various domains. Enhancing EEG data processing and emotion recognition models remains a key research focus in this field. This paper presents an emotion recognition framework combining the CUSUM algorithm-based adaptive window selection technique with the convolutional attention-enhanced Kolmogorov-Arnold Networks (CA-KAN). The improved CUSUM algorithm effectively extracts the most emotion-relevant segments from raw EEG data. Furthermore, by enhancing the KAN network, the CA-KAN model achieves both high accuracy and efficiency in emotion recognition. The proposed framework achieved peak classification accuracies of 94.63% and 94.73% on the SEED and SEED-IV datasets, respectively. Additionally, the framework offers a lightweight advantage, demonstrating significant potential for real-world applications, including medical emotion monitoring and driver emotion detection.</p>","PeriodicalId":10500,"journal":{"name":"Cognitive Neurodynamics","volume":"19 1","pages":"100"},"PeriodicalIF":3.1,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12187633/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144504983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Identification of macrophage-associated diagnostic biomarkers and molecular subtypes in gestational diabetes mellitus based on machine learning.","authors":"Kai Wei, Liyun Yuan, Yongsheng Ge, Han Yu, Guoping Zhao, Guoqing Zhang, Guohua Liu","doi":"10.1080/21691401.2025.2513893","DOIUrl":"https://doi.org/10.1080/21691401.2025.2513893","url":null,"abstract":"<p><p>Gestational diabetes mellitus (GDM) is a common metabolic disorder during pregnancy, involving multiple immune and inflammatory factors. Macrophages play a crucial role in its development. This study integrated scRNA-seq and RNA-seq data to explore macrophage-related diagnostic genes and GDM subtypes. For scRNA-seq data, cell clusters were annotated using the SingleR package and validated with marker gene expression profiles, while hdWGCNA analysis identified three gene modules related to macrophages. A diagnostic model for GDM derived from endothelial cell transcriptomes was constructed by employing a variety of machine learning ensemble algorithms, achieving an AUC of 0.887. The model identified five differentially expressed genes (ZEB2, MALAT1, HEBP1, AHSA1, and TTC3) as potential diagnostic biomarkers. The CB-DSNMF algorithm was proposed to identify two distinct GDM subtypes from RNA-seq data, revealing significant differences in biological behaviours. This algorithm outperformed other baselines in multiple clustering metrics. Mendelian randomisation analysis identified ZEB2 as a gene causally related to GDM risk. A transcription factor (TF)-gene regulatory network was constructed for these genes using the ENCODE database. The study highlights the importance of macrophages in GDM, provides a high-precision diagnostic model, and offers new insights into personalised treatment strategies, contributing to a better understanding of GDM pathophysiology.</p>","PeriodicalId":8736,"journal":{"name":"Artificial Cells, Nanomedicine, and Biotechnology","volume":"53 1","pages":"20-33"},"PeriodicalIF":4.5,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144233033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating the impact of improved maize varieties on agricultural productivity and technical efficiency among smallholder farmers in the Eastern Cape, South Africa: an empirical analysis.","authors":"Lelethu Mdoda, Nthabeleng Tamako, Lungile S Gidi, Denver Naidoo","doi":"10.1080/21645698.2025.2476667","DOIUrl":"10.1080/21645698.2025.2476667","url":null,"abstract":"<p><p>Agriculture is essential to South Africa's economy, and maize is a crucial crop for smallholder farmers in the Eastern Cape. Traditional maize varieties face challenges related to productivity and resilience, prompting the promotion of Improved Maize Varieties (IMVs) to enhance yields and efficiency. This study investigates the impact of IMV adoption on agricultural productivity and technical efficiency in the region, addressing a gap in empirical evidence. Using a multistage random sampling approach, data was collected from 150 smallholder maize farmers and analyzed using stochastic production frontier, endogenous switching regression models, and the stochastic meta-frontier model. The study results reveal that 62% of the farmers are male, averaging 53 years old, and manage about four hectares with a mean monthly income of ZAR 3,562.13. Challenges, such as rainfall shortages and limited access to credit, hinder IMV adoption, although high access to extension services and diverse input use positively affect productivity. The adopted IMVs by farmers, including open-pollinated, hybrid, and genetically modified (GM) varieties, significantly boost maize yields and farm returns - yielding an average increase of 1.92 kg/ha and returns of ZAR 468.01 per hectare. Key adoption factors are education, farm size, and access to seeds and extension services, whereas barriers include market distance and family size. Technical efficiency is generally high at 74%, with farm size, seed, pesticides, agrochemicals, and fertilizers positively impacting maize production, whereas family labor negatively affects it. Factors such as age, education, and access to services significantly reduce technical inefficiency, while herd size, off-farm income, and distance to the market have mixed effects. The stochastic meta-frontier approach reveals that smallholder farmers adopting improved technologies show higher mean technical efficiency, indicating that advanced methods contribute to better resource use and productivity than traditional systems. This study suggests that targeted support is needed for farmers, enhancing access to extension services, affordable seeds, financial support, and investing in infrastructure and education can further improve adoption rates, technical efficiency, and overall productivity. Promoting improved technologies such as maize varieties will enhance the technical efficiency of farms, regardless of their adoption status. It would be key to improving overall agricultural productivity and farm household incomes.</p>","PeriodicalId":54282,"journal":{"name":"Gm Crops & Food-Biotechnology in Agriculture and the Food Chain","volume":"16 1","pages":"272-304"},"PeriodicalIF":4.5,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11925116/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143659599","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The role of beta band phase resetting in audio-visual temporal order judgment.","authors":"Yueying Li, Yasuki Noguchi","doi":"10.1007/s11571-024-10183-0","DOIUrl":"10.1007/s11571-024-10183-0","url":null,"abstract":"<p><p>The integration of auditory and visual stimuli is essential for effective language processing and social perception. The present study aimed to elucidate the mechanisms underlying audio-visual (A-V) integration by investigating the temporal dynamics of multisensory regions in the human brain. Specifically, we evaluated inter-trial coherence (ITC), a neural index indicative of phase resetting, through scalp electroencephalography (EEG) while participants performed a temporal-order judgment task that involved auditory (beep, A) and visual (flash, V) stimuli. The results indicated that ITC phase resetting was greater for bimodal (A + V) stimuli compared to unimodal (A or V) stimuli in the posterior temporal region, which resembled the responses of A-V multisensory neurons reported in animal studies. Furthermore, the ITC got lager as the stimulus-onset asynchrony (SOA) between beep and flash approached 0 ms. This enhancement in ITC was most clearly seen in the beta band (13-30 Hz). Overall, these findings highlight the importance of beta rhythm activity in the posterior temporal cortex for the detection of synchronous audiovisual stimuli, as assessed through temporal order judgment tasks.</p><p><strong>Supplementary information: </strong>The online version contains supplementary material available at 10.1007/s11571-024-10183-0.</p>","PeriodicalId":10500,"journal":{"name":"Cognitive Neurodynamics","volume":"19 1","pages":"28"},"PeriodicalIF":3.1,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11735826/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143001022","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"<i>ZmNF-YB10</i>, a maize NF-Y transcription factor, positively regulates drought and salt stress response in <i>Arabidopsis thaliana</i>.","authors":"Yimeng Wang, Peng Jiao, Chenyang Wu, Chunlai Wang, Ke Shi, Xiaoqi Gao, Shuyan Guan, Yiyong Ma","doi":"10.1080/21645698.2024.2438421","DOIUrl":"https://doi.org/10.1080/21645698.2024.2438421","url":null,"abstract":"<p><p>Maize (<i>Zea mays</i> L.) is a major food and feed crop and an important raw material for energy, chemicals, and livestock. The NF-Y family of transcription factors in maize plays a crucial role in the regulation of plant development and response to environmental stress. In this study, we successfully cloned and characterized the maize NF-Y transcription factor gene <i>ZmNF-YB10</i>. We used bioinformatics, quantitative fluorescence PCR, and other techniques to analyze the basic properties of the gene, its tissue expression specificity, and its role in response to drought, salt, and other stresses. The results indicated that the gene was 1209 base pairs (bp) in length, with a coding sequence (CDS) region of 618 bp, encoding a polypeptide composed of 205 amino acid residues. This polypeptide has a theoretical isoelectric point of 5.85 and features a conserved structural domain unique to the NF-Y family. Quantitative fluorescence PCR results demonstrated that the <i>ZmNF-YB10</i> gene was differentially upregulated under drought and salt stress treatments but exhibited a negatively regulated expression pattern under alkali and cold stress treatments. Transgenic <i>Arabidopsis thaliana</i> subjected to drought and salt stress in soil showed greener leaves than wild-type <i>A. thaliana</i>. In addition, the overexpression lines showed reduced levels of hydrogen peroxide (H₂O₂), superoxide (O^2-), and malondialdehyde (MDA) and increased activities of peroxidase (POD), catalase (CAT), and superoxide dismutase (SOD). Western blot analysis revealed a distinct band at 21.8 kDa. Salt and drought tolerance analyses conducted in <i>E. coli</i> BL21 indicated a positive regulation. In yeast cells, <i>ZmNF-YB10</i> exhibited a biological function that enhances salt and drought tolerance. Protein interactions were observed among the <i>ZmNF-YB10</i>, <i>ZmNF-YC2</i>, and <i>ZmNF-YC4</i> genes. It is hypothesized that the <i>ZmNF-YB10, ZmNF-YC2</i>, and <i>ZmNF-YC4</i> genes may play a role in the response to abiotic stresses, such as drought and salt tolerance, in maize.</p>","PeriodicalId":54282,"journal":{"name":"Gm Crops & Food-Biotechnology in Agriculture and the Food Chain","volume":"16 1","pages":"28-45"},"PeriodicalIF":4.5,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142883647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction.","authors":"","doi":"10.1080/21691401.2025.2473244","DOIUrl":"https://doi.org/10.1080/21691401.2025.2473244","url":null,"abstract":"","PeriodicalId":8736,"journal":{"name":"Artificial Cells, Nanomedicine, and Biotechnology","volume":"53 1","pages":"87"},"PeriodicalIF":4.5,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143555639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}