Peishan Dai , Zhuang He , Jialin Luo , Kaineng Huang , Ting Hu , Qiongpu Chen , Shenghui Liao , Xiaoping Yi , the REST-meta-MDD Consortium
{"title":"Using effective connectivity-based predictive modeling to predict MDD scale scores from multisite rs-fMRI data","authors":"Peishan Dai , Zhuang He , Jialin Luo , Kaineng Huang , Ting Hu , Qiongpu Chen , Shenghui Liao , Xiaoping Yi , the REST-meta-MDD Consortium","doi":"10.1016/j.jneumeth.2025.110406","DOIUrl":"10.1016/j.jneumeth.2025.110406","url":null,"abstract":"<div><h3>Background</h3><div>Major depressive disorder (MDD) is a severe mental illness, and the Hamilton Depression Rating Scale (HAMD) is commonly used to quantify its severity. Our aim is to develop a predictive model for MDD symptoms using machine learning techniques based on effective connectivity (EC) from resting-state functional magnetic resonance imaging (rs-fMRI).</div></div><div><h3>New method</h3><div>We obtained large-scale rs-fMRI data and HAMD scores from the multi-site REST-meta-MDD dataset. Average time series were extracted using different atlases. Brain EC features were computed using Granger causality analysis based on symbolic path coefficients, and a machine learning model based on EC was constructed to predict HAMD scores. Finally, the most predictive features were identified and visualized.</div></div><div><h3>Results</h3><div>Experimental results indicate that different brain atlases significantly impact predictive performance, with the Dosenbach atlas performing best. EC-based models outperformed functional connectivity, achieving the best predictive accuracy (r = 0.81, p < 0.001, Root Mean Squared Error=3.55). Among various machine learning methods, support vector regression demonstrated superior performance.</div></div><div><h3>Comparison with existing methods</h3><div>Current phenotype score prediction primarily relies on FC, which cannot indicate the direction of information flow within brain networks. Our method is based on EC, which contains more comprehensive brain network information and has been validated on large-scale multi-site data.</div></div><div><h3>Conclusions</h3><div>Brain network connectivity features effectively predict HAMD scores in MDD patients. The identified EC feature network may serve as a biomarker for predicting symptom severity. Our work may provide clinically significant insights for the early diagnosis of MDD, thereby facilitating the development of personalized diagnostic tools and therapeutic interventions.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"417 ","pages":"Article 110406"},"PeriodicalIF":2.7,"publicationDate":"2025-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143453812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Miriam Paola Pili , Livio Provenzi , Lucia Billeci , Valentina Riva , Maddalena Cassa , Eleonora Siri , Giorgia Procissi , Elisa Roberti , Elena Capelli
{"title":"Exploring the impact of manual and automatic EEG pre-processing methods on interpersonal neural synchrony measures in parent-infant hyperscanning studies","authors":"Miriam Paola Pili , Livio Provenzi , Lucia Billeci , Valentina Riva , Maddalena Cassa , Eleonora Siri , Giorgia Procissi , Elisa Roberti , Elena Capelli","doi":"10.1016/j.jneumeth.2025.110400","DOIUrl":"10.1016/j.jneumeth.2025.110400","url":null,"abstract":"<div><h3>Background</h3><div>Electroencephalograph (EEG) hyperscanning allows studying Interpersonal Neural Synchrony (INS) between two or more individuals across different social conditions, including parent-infant interactions. Signal pre-processing is crucial to optimize computation of INS estimates; however, few attempts have been made at comparing the impact of different dyadic EEG data pre-processing methods on INS estimates.</div></div><div><h3>New methods</h3><div>EEG data collected on 31 mother-infant dyads (8–10 months) engaged in a Face-to-Face Still-Face Procedure were pre-processed with two versions of the same pipeline, the “automated” and the “manual”. Cross-frequency PLV in the theta (3–5 Hz, 4–7 Hz) and alpha (6–9 Hz, 8–12 Hz) frequency bands were computed after automated and manual pre-processing and compared through Pearson’s correlations and Repeated Measures ANOVAs.</div></div><div><h3>Results</h3><div>PLVs computed in the theta, but not alpha, frequency band were significantly higher after automated pre-processing than after manual pre-processing. Moreover, the automated pipeline rejected a significantly lower percentage of ICs and epochs compared to the manual pipeline.</div></div><div><h3>Comparison with existing methods</h3><div>While no direct comparison with existing dyadic EEG data pre-processing pipelines was made, this is the first study assessing the impact of different methodological decisions, particularly of the degree of pre-processing automatization, on cross-frequency PLV computed on a dataset of parent-infant dyads.</div></div><div><h3>Conclusions</h3><div>Non-directional phase-based INS indexes such as the PLV seem to be affected by the degree of automatization of the pre-processing pipeline. Future research should strive for standardization of dyadic EEG pre-processing methods.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"417 ","pages":"Article 110400"},"PeriodicalIF":2.7,"publicationDate":"2025-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143468257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Development of a non-invasive novel individual marmoset holder for evaluation by awake functional magnetic resonance brain imaging","authors":"Fumiko Seki , Terumi Yurimoto , Michiko Kamioka , Takashi Inoue , Yuji Komaki , Atsushi Iriki , Erika Sasaki , Yumiko Yamazaki","doi":"10.1016/j.jneumeth.2025.110390","DOIUrl":"10.1016/j.jneumeth.2025.110390","url":null,"abstract":"<div><h3>Background</h3><div>Although functional MRI (fMRI) in awake marmosets (<em>Callithrix jacchus</em>) is fascinating for functional brain mapping and evaluation of brain disease models, it is difficult to launch awake fMRI on scanners with bore sizes of less than 16 cm. A universal marmoset holder for the small-bore size MRI was designed, and it was evaluated whether this holder could conduct auditory stimulation fMRI in the awake state using 16 cm bore size MRI scanner.</div></div><div><h3>New method</h3><div>The marmoset holder was designed with an outer diameter of 71.9 mm. A holder was designed to allow adjustment according to the individual head shape, enabling the use of the holder universally. An awake fMRI study of auditory response was conducted to evaluate the practicality of the new holder. Whole-brain activation was investigated when marmosets heard the marmoset social communication “phee call” an artificial tone sound and reversed of those.</div></div><div><h3>Results</h3><div>The prefrontal cortex was significantly activated in response to phee calls, whereas only the auditory cortex was activated in response to pure tones. In contrast, the auditory response was decreased when marmosets heard phee call. Their stimulus-specific responses indicated they perceived and differentiated sound characteristics in the fMRI environment.</div></div><div><h3>Comparison with existing methods</h3><div>A holder does not require surgical intervention or a custom-made helmet to minimize head movement in a small space.</div></div><div><h3>Conclusion</h3><div>Our newly developed holder made it possible to perform longitudinal fMRI experiments on multiple marmosets in a less invasive manner.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"417 ","pages":"Article 110390"},"PeriodicalIF":2.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143433025","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Max A. van den Boom , Nicholas M. Gregg , Gabriela Ojeda Valencia , Brian N. Lundstrom , Kai J. Miller , Dorien van Blooijs , Geertjan J.M. Huiskamp , Frans S.S. Leijten , Gregory A. Worrell , Dora Hermes
{"title":"ER-detect: A pipeline for robust detection of early evoked responses in BIDS-iEEG electrical stimulation data","authors":"Max A. van den Boom , Nicholas M. Gregg , Gabriela Ojeda Valencia , Brian N. Lundstrom , Kai J. Miller , Dorien van Blooijs , Geertjan J.M. Huiskamp , Frans S.S. Leijten , Gregory A. Worrell , Dora Hermes","doi":"10.1016/j.jneumeth.2025.110389","DOIUrl":"10.1016/j.jneumeth.2025.110389","url":null,"abstract":"<div><h3>Background</h3><div>Human brain connectivity can be measured in different ways. Intracranial EEG (iEEG) measurements during single pulse electrical stimulation provide a unique way to assess the spread of electrical information with millisecond precision. However, the methods used for the detection of responses in cortico-cortical evoked potential (CCEP) data vary across studies, from visual inspection with manual annotation to a variety of automated methods.</div></div><div><h3>New method</h3><div>To provide a robust workflow to process CCEP data and detect early evoked responses in a fully automated and reproducible fashion, we developed the Early Response (ER)-detect toolbox. ER-detect is an open-source Python package and Docker application to preprocess BIDS structured iEEG data and detect early evoked CCEP responses. ER-detect can use three early response detection methods, which were validated against 14 manually annotated CCEP datasets from two different clinical sites by four independent raters.</div></div><div><h3>Results and comparison with existing methods</h3><div>ER-detect’s automated detection performed on par with the inter-rater reliability (Cohen’s Kappa of ∼0.6). Moreover, ER-detect was optimized for processing large CCEP datasets, to be used in conjunction with other connectomic investigations.</div></div><div><h3>Conclusion</h3><div>ER-detect provides a highly efficient standardized workflow such that iEEG-BIDS data can be processed in a consistent manner and enhance the reproducibility of CCEP based connectivity results for both research and clinical purposes.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"418 ","pages":"Article 110389"},"PeriodicalIF":2.7,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143425460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Cross prior Bayesian attention with correlated inception and residual learning for brain tumor classification using MR images (CB-CIRL Net)","authors":"B. Vijayalakshmi , S. Anand","doi":"10.1016/j.jneumeth.2025.110392","DOIUrl":"10.1016/j.jneumeth.2025.110392","url":null,"abstract":"<div><h3>Background</h3><div>Brain tumor classification from magnetic resonance (MR) images is crucial for early diagnosis and effective treatment planning. However, the homogeneity of tumors across different categories poses a challenge. Although, attention-based convolutional neural networks (CNNs) approaches have shown promising results in brain tumor classification, simultaneous consideration of both spatial and channel-specific features remains limited.</div></div><div><h3>Methods</h3><div>This study proposes a novel model that integrates Bi-FocusNet with correlated learning and CB-Attention. Bi-FocusNet is designed to concentrate on both spatial and channel-wise tumor features by using two complementary learning methodologies: correlated spatial inception learning and correlated channel residual learning. These learnings extract richer and more diverse feature representations from tumor lesions of varied sizes, significantly enhancing the model’s learning capacity. The CB-Attention mechanism works as a cross-learning module, facilitating interaction between the two learning methods to capture the missing information across spatial and channel-wise features.</div></div><div><h3>Results</h3><div>Ablation studies and experiments were conducted using the BT-large-2c, Figshare, and Kaggle datasets. The proposed model outperformed existing classification methods in accuracy and other metrics, demonstrating enhanced performance on all three datasets with accuracies of 99.02 %, 97.06 %, and 96.44 %, respectively. Additionally, the BT-Merged-4c dataset was used to assess the ability to handle class variation, and 96.28 % accuracy was achieved.</div></div><div><h3>Conclusion</h3><div>The CB-CIRL Net improves the extraction of spatial and channel-wise features through the utilization of Bi-FocusNet with correlated learning and CB-Attention, resulting in enhanced classification accuracy across various datasets. The model's outstanding performance demonstrates its capacity to improve brain tumor diagnosis and clinical application.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"417 ","pages":"Article 110392"},"PeriodicalIF":2.7,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143414309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Comparison of methods for the preparation of single-cell suspensions of rat retina and immunophenotyping by flow cytometry","authors":"Qingyao Wang , Xiandan Zhu , Xing Zhang , Yong Xia , Xuesong Tian","doi":"10.1016/j.jneumeth.2025.110384","DOIUrl":"10.1016/j.jneumeth.2025.110384","url":null,"abstract":"<div><h3>Background</h3><div>Although the eye has traditionally been considered an immune-privileged organ, progressive studies have re-evaluated the role of the immune response in retinopathy. The application of flow cytometry for immunophenotyping analysis of retinal single-cell suspensions has gradually attracted attention.</div></div><div><h3>New methods</h3><div>We dissociated the retinal tissue using trypsin digestion, papain digestion, mechanical grinding treatment and liberase + DNase Ⅰ digestion, respectively. Subsequently we assessed the quality of cell suspension (clumping rate, concentration and viability) with two different dyes, Trypan Blue (TPB) and Acridine Orange/Propidium Iodide (AO/PI). We distinguished T cells and their sub-populations by flow cytometry. Furthermore, we analyzed their immune responses to evaluate those four methods for preparing retinal single-cell suspension.</div></div><div><h3>Results</h3><div>The AO/PI staining method enables a rapid and precise evaluation of cell quality of retinal single cell suspensions, while TPB staining has limitations. Flow cytometric analysis showed that single cell suspensions dispersed with papain and trypsin exhibited reduced cell adhesion. However, trypsin digestion may affect antibody binding. The mechanical grinding treatment reduced cell yield and was prone to double-positivity. The number of cells within the cell-circle gate is significantly limited in the Liberase + DNase I digestion method.</div></div><div><h3>Comparison with existing methods</h3><div>The AO/PI staining method was employed to assess the quality of retinal single-cell suspensions. T cells and their subpopulations in retinal tissues were analyzed by flow cytometry. These results were integrated to evaluate the optimal preparation protocol for retinal single-cell suspensions.</div></div><div><h3>Conclusions</h3><div>Papain digestion is a superior method for preparing retinal single-cell suspensions.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"417 ","pages":"Article 110384"},"PeriodicalIF":2.7,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143382630","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mallory Karr , Arti Patel , William Klipec , Christopher L. Kliethermes
{"title":"Assessment of voluntary drug and alcohol intake in Drosophila melanogaster using a modified one-tube capillary feeding assay","authors":"Mallory Karr , Arti Patel , William Klipec , Christopher L. Kliethermes","doi":"10.1016/j.jneumeth.2025.110388","DOIUrl":"10.1016/j.jneumeth.2025.110388","url":null,"abstract":"<div><h3>Background</h3><div>The capillary feeding assay measures intakes of solutions containing ethanol and other drugs of abuse in the fruit fly <em>Drosophila melanogaster</em>. Existing single-tube and two-choice variations of the assay conflate drug intake with nutrient intake and utilize conditions that decrease lifespan of flies, suggesting these conditions might serve as a physiological stressor.</div></div><div><h3>New method</h3><div>The current experiments used a novel variation of a single-tube capillary feeding assay with flies maintained on standard, semi-soft food throughout the experiment, and offered a drug in a moderately preferred sucrose solution. Preference or aversion for a drug-containing solution was assessed relative to a control sucrose solution.</div></div><div><h3>Results</h3><div>We found concentration-dependent preferences for solutions containing ethanol or fencamfamine, aversions to solutions containing amphetamine, caffeine, muscimol, nicotine, pentobarbital, or picrotoxin, and a nominal aversion to solutions containing methamphetamine. Some aversions were found only in male flies, although low overall intake by female flies likely contributed to this apparent sex-specific effect.</div></div><div><h3>Comparison with existing method(s)</h3><div>The continuous availability of semi-soft food in our assay to decouple the need for nutrients from intake of the drug-containing solution. In addition, our assay uses only a single capillary tube per vial of flies, making it less resource intensive than two-choice capillary feeding assays.</div></div><div><h3>Conclusions</h3><div>The availability of standard food in our modified, one-tube capillary feeding assay should prove to be useful modification of the capillary feeding assay for studies of drug intake.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"416 ","pages":"Article 110388"},"PeriodicalIF":2.7,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143267532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Optimization of permeabilized brain tissue preparation to improve the analysis of mitochondrial oxidative capacities in specific subregions of the rat brain","authors":"Léa Dorémus , Emilie Dugast , Arnaud Delafenêtre , Morgane Delouche , Thomas Aupy , Olivier Bernard , Stéphane Sebille , Nathalie Thiriet , Jérôme Piquereau","doi":"10.1016/j.jneumeth.2025.110387","DOIUrl":"10.1016/j.jneumeth.2025.110387","url":null,"abstract":"<div><h3>Background</h3><div>As the major energy producer of cerebral tissue, mitochondria play key roles in brain physiology and physiopathology. Yet, the fine details of the functioning of mitochondrial oxidative phosphorylation in this organ are still scattered with grey area. This is partly due to the heterogeneity of this tissue that challenges our abilities to study specific cerebral subregions. In the last decades, cerebral mitochondria have largely been studied as a single entity by isolating mitochondria from large sections of brain. Given the evidence that these organelles must adapt to brain areas functions, it seems crucial to develop technologies enabling study of the mitochondria in given subregions.</div></div><div><h3>New method</h3><div>A few years ago, a method allowing the investigation of mitochondrial functions in permeabilized brain subregions have been proposed by Holloway’s team. Although this protocol represented a significant advance, we propose improvements in the tissue permeabilization procedure and in the conditions for measuring oxidative capacity.</div></div><div><h3>Results and comparison with existing methods</h3><div>The present study demonstrates that adjustments enabled obtention of higher respiration values than Holloway’s protocol and might allow the detection of slight mitochondrial alterations. In a second part of this study, we showed that cortex, striatum, hippocampus and cerebellum displayed similar maximal oxidative capacities (under pyruvate, malate and succinate) while complex IV-driven respiration is significantly lower in cerebellum compared to cortex. These observations were supported by the measurement of citrate synthase and cytochrome oxidase activities.</div></div><div><h3>Conclusion</h3><div>The developed procedure improves the investigations of mitochondrial electron transfer chain in specific cerebral regions.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"416 ","pages":"Article 110387"},"PeriodicalIF":2.7,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143267536","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Discrete variational autoencoders BERT model-based transcranial focused ultrasound for Alzheimer's disease detection","authors":"Kaushika Reddy Thipparthy , Archana Kollu , Chaitanya Kulkarni , Ashit Kumar Dutta , Hardik Doshi , Aditya Kashyap , Kumari Priyanka Sinha , Suresh Babu Kondaveeti , Rupesh Gupta","doi":"10.1016/j.jneumeth.2025.110386","DOIUrl":"10.1016/j.jneumeth.2025.110386","url":null,"abstract":"<div><h3>Research background</h3><div>Alzheimer's Disease (AD) is a neurodegenerative condition marked by symptoms including aphasia and diminished verbal fluency. Researchers have employed phonetic attributes, fluency, pauses, and various paralinguistic traits, or derived aspects from transcribed text, to identify Alzheimer's disease.</div></div><div><h3>Methods and methodology</h3><div>Nevertheless, conventional acoustic feature-based detection techniques are constrained in their ability to capture semantic information, and the process of transcribing speech into text is both time-consuming and labour-intensive. Non-invasive brain stimulation (NBS), encompassing methods such as transcranial magnetic stimulation (TMS) and Transcranial focused ultrasound (tFUS), has been investigated as a potential intervention to enhance cognitive functions and communication in Alzheimer's patients, demonstrating efficacy in modulating brain activity and promoting neuroplasticity. This research utilises Discrete Variational Autoencoders to transform speech into pseudo-phoneme sequences, subsequently applying the BERT (Bidirectional Encoder Representations from Transformers) model to analyse the relationships among these pseudo-phoneme sequences. This research proposes a tFUS-BERT model to encapsulate the linguistic representations of audio.</div></div><div><h3>Result analysis</h3><div>The proposed tFUS-BERT model demonstrated its effectiveness with an accuracy of 76.06 % when combined with Wav2vec 2.0 and 71.83 % with Hu-BERT, outperforming the baseline by 5.63 % on the ADReSSo dataset. Additionally, the model exhibited superior performance in capturing linguistic representations compared to traditional acoustic methods, showcasing its potential for accurate and scalable Alzheimer's detection.</div></div><div><h3>Comparison with previous studies</h3><div>The model attains an accuracy of 70.42 % on the ADReSSo (Alzheimer's Dementia Recognition through Spontaneous Speech Only) dataset, reflecting a 5.63 % enhancement compared to the baseline system.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"416 ","pages":"Article 110386"},"PeriodicalIF":2.7,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143167454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"EEG-based fatigue state evaluation by combining complex network and frequency-spatial features","authors":"Kefa Wang, Xiaoqian Mao, Yuebin Song, Qiuyu Chen","doi":"10.1016/j.jneumeth.2025.110385","DOIUrl":"10.1016/j.jneumeth.2025.110385","url":null,"abstract":"<div><h3>Background</h3><div>The proportion of traffic accidents caused by fatigue driving is increasing year by year, which has aroused wide concerns for researchers. In order to rapidly and accurately detect drivers' fatigue, this paper proposed an electroencephalogram (EEG)-based fatigue state evaluation method by combining complex network and frequency-spatial features.</div></div><div><h3>New method</h3><div>First, this paper constructed a complex network model based on the relative wavelet entropy to characterize the correlation strength information between channels. Then, the differential entropy and symmetry quotient were respectively calculated to extract frequency and spatial features. Then, the brain heat map combined the complex network and frequency-spatial features with different dimensions together as the fusion features. Finally, a convolutional neural network-long short-term memory (CNN-LSTM) neural network was used to evaluate the three-class fatigue states of the EEG data in the Shanghai Jiao Tong University (SJTU) Emotion EEG Dataset (SEED)-VIG dataset, and it was validated on the dataset on the Mendeley Data website.</div></div><div><h3>Results</h3><div>The experimental results of SEED-VIG dataset show that the average classification accuracy of three-class fatigue states, namely, awake, tired and drowsy, reaches 96.57 %. The average classification accuracy on the dataset on the Mendeley Data website reaches 99.23 %.</div></div><div><h3>Comparison with existing methods</h3><div>This method has a best evaluation performance compared with the state-of-the-art methods for the three-class fatigue states recognition.</div></div><div><h3>Conclusions</h3><div>The experiment results validated the feasibility of the fatigue state evaluation method based on the correlations between channels and the frequency-spatial features, which is of great significance for developing a driver fatigue detection system.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"416 ","pages":"Article 110385"},"PeriodicalIF":2.7,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143255909","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}