{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,8]],"date-time":"2026-06-08T02:24:19Z","timestamp":1780885459372,"version":"3.54.1"},"reference-count":67,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2024,12,20]],"date-time":"2024-12-20T00:00:00Z","timestamp":1734652800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neuroinform."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p>Modeling multi-channel electroencephalographic (EEG) time-series is a challenging tasks, even for the most recent deep learning approaches. Particularly, in this work, we targeted our efforts to the high-fidelity reconstruction of this type of data, as this is of key relevance for several applications such as classification, anomaly detection, automatic labeling, and brain-computer interfaces.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We analyzed the most recent works finding that high-fidelity reconstruction is seriously challenged by the complex dynamics of the EEG signals and the large inter-subject variability. So far, previous works provided good results in either high-fidelity reconstruction of single-channel signals, or poor-quality reconstruction of multi-channel datasets. Therefore, in this paper, we present a novel deep learning model, called hvEEGNet, designed as a hierarchical variational autoencoder and trained with a new loss function. We tested it on the benchmark Dataset 2a (including 22-channel EEG data from 9 subjects).<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>We show that it is able to reconstruct all EEG channels with high-fidelity, fastly (in a few tens of epochs), and with high consistency across different subjects. We also investigated the relationship between reconstruction fidelity and the training duration and, using hvEEGNet as an anomaly detector, we spotted some data in the benchmark dataset that are corrupted and never highlighted before.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>Thus, hvEEGNet could be very useful in several applications where automatic labeling of large EEG dataset is needed and time-consuming. At the same time, this work opens new fundamental research questions about (1) the effectiveness of deep learning models training (for EEG data) and (2) the need for a systematic characterization of the input EEG data to ensure robust modeling.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fninf.2024.1459970","type":"journal-article","created":{"date-parts":[[2024,12,20]],"date-time":"2024-12-20T06:30:40Z","timestamp":1734676240000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":11,"title":["hvEEGNet: a novel deep learning model for high-fidelity EEG reconstruction"],"prefix":"10.3389","volume":"18","author":[{"given":"Giulia","family":"Cisotto","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alberto","family":"Zancanaro","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Italo F.","family":"Zoppis","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sara L.","family":"Manzoni","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2024,12,20]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"5320","DOI":"10.3390\/app11125320","article-title":"A review of machine learning and deep learning techniques for anomaly detection in IoT data","volume":"11","author":"Al-amri","year":"2021","journal-title":"Appl. Sci"},{"key":"B2","first-page":"370","article-title":"\u201cConvolutional autoencoder approach for EEG compression and reconstruction in m-health systems,\u201d","volume-title":"2018 14th International Wireless Communications & Mobile Computing Conference (IWCMC)","author":"Al-Marridi","year":"2018"},{"key":"B3","doi-asserted-by":"publisher","first-page":"106088","DOI":"10.1016\/j.compbiomed.2022.106088","article-title":"Wearable electroencephalography and multi-modal mental state classification: a systematic literature review","volume":"150","author":"Anders","year":"2022","journal-title":"Comp. Biol. Med"},{"key":"B4","doi-asserted-by":"publisher","first-page":"12814","DOI":"10.1016\/j.eswa.2012.05.012","article-title":"Correlation based dynamic time warping of multivariate time series","volume":"39","author":"Bank\u00f3","year":"2012","journal-title":"Expert Syst. Appl"},{"key":"B5","doi-asserted-by":"publisher","first-page":"400","DOI":"10.1109\/THMS.2021.3137035","article-title":"Shared intelligence for robot teleoperation via BMI","volume":"52","author":"Beraldo","year":"2022","journal-title":"IEEE Transact. Hum. Mach. Syst"},{"key":"B6","doi-asserted-by":"crossref","first-page":"3150","DOI":"10.1109\/SMC53654.2022.9945517","article-title":"\u201cEEG2Vec: learning affective EEG representations via variational autoencoders,\u201d","volume-title":"2022 IEEE International Conference on Systems, Man, and Cybernetics (SMC)","author":"Bethge","year":"2022"},{"key":"B7","doi-asserted-by":"publisher","first-page":"539","DOI":"10.1016\/j.neuroimage.2007.01.051","article-title":"The non-invasive Berlin Brain-Computer Interface: fast acquisition of effective performance in untrained subjects","volume":"37","author":"Blankertz","year":"2007","journal-title":"Neuroimage"},{"key":"B8","doi-asserted-by":"publisher","first-page":"859","DOI":"10.1080\/01621459.2017.1285773","article-title":"Variational inference: a review for statisticians","volume":"112","author":"Blei","year":"2017","journal-title":"J. Am. Stat. Assoc"},{"key":"B9","doi-asserted-by":"publisher","first-page":"103","DOI":"10.3390\/fi13050103","article-title":"Deep learning-based classification of fine hand movements from low frequency EEG","volume":"13","author":"Bressan","year":"2021","journal-title":"Fut. Int"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1926","DOI":"10.1126\/science.1099745","article-title":"Neuronal oscillations in cortical networks","volume":"304","author":"Buzsaki","year":"2004","journal-title":"Science"},{"key":"B11","doi-asserted-by":"publisher","first-page":"104510","DOI":"10.1016\/j.ijmedinf.2021.104510","article-title":"The need to separate the wheat from the chaff in medical informatics: introducing a comprehensive checklist for the (self)-assessment of medical AI studies","volume":"153","author":"Cabitza","year":"2021","journal-title":"Int. J. Med. Inform"},{"key":"B12","doi-asserted-by":"publisher","first-page":"5353","DOI":"10.1109\/JSEN.2019.2906572","article-title":"Removal of muscle artifacts from the EEG: a review and recommendations","volume":"19","author":"Chen","year":"2019","journal-title":"IEEE Sens. J"},{"key":"B13","doi-asserted-by":"crossref","first-page":"1075","DOI":"10.1109\/ICASSP39728.2021.9414749","article-title":"\u201cREPAC: reliable estimation of phase-amplitude coupling in brain networks,\u201d","volume-title":"ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","author":"Cisotto","year":"2021"},{"key":"B14","first-page":"1","article-title":"\u201cReal-time detection of EEG electrode displacement for brain-computer interface applications,\u201d","volume-title":"Proceedings of 5th International Conference on Wireless Communications, Vehicular Technology, Information Theory and Aerospace & Electronic Systems (Wireless VITAE)","author":"Cisotto","year":"2015"},{"key":"B15","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1109\/TIT.1967.1053964","article-title":"Nearest neighbor pattern classification","volume":"13","author":"Cover","year":"1967","journal-title":"IEEE Transact. Inf. Theory"},{"key":"B16","article-title":"\u201cSoft-DTW: a differentiable loss function for time-series,\u201d","volume-title":"Proceedings of the 34th International Conference on Machine Learning (ICML)","author":"Cuturi","year":"2017"},{"key":"B17","doi-asserted-by":"publisher","first-page":"6152","DOI":"10.1007\/s00034-022-02071-x","article-title":"Joint ECG-EMG-EEG signal compression and reconstruction with incremental multimodal autoencoder approach","volume":"41","author":"Dasan","year":"2022","journal-title":"Circ. Syst. Signal Process"},{"key":"B18","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1111\/coin.12146","article-title":"Threaded ensembles of autoencoders for stream learning","volume":"34","author":"Dong","year":"2018","journal-title":"Comp. Intell"},{"key":"B19","doi-asserted-by":"publisher","first-page":"526","DOI":"10.1109\/TBME.2003.809476","article-title":"A simple system for detection of EEG artifacts in polysomnographic recordings","volume":"50","author":"Durka","year":"2003","journal-title":"IEEE Transact. Biomed. Eng"},{"key":"B20","doi-asserted-by":"publisher","first-page":"227","DOI":"10.1016\/j.compbiomed.2019.05.025","article-title":"Autoencoding of long-term scalp electroencephalogram to detect epileptic seizure for diagnosis support system","volume":"110","author":"Emami","year":"2019","journal-title":"Comput. Biol. Med"},{"key":"B21","first-page":"28","article-title":"\u201cA multi-artifact EEG denoising by frequency-based deep learning,\u201d","volume-title":"CEUR Workshop Proceedings, Vol. 3576 (CEUR-WS)","author":"Gabardi","year":"2023"},{"key":"B22","doi-asserted-by":"publisher","first-page":"312","DOI":"10.1097\/WNP.0b013e3181f534f4","article-title":"Automatic removal of various artifacts from EEG signals using combined methods","volume":"27","author":"Gao","year":"2010","journal-title":"J. Clin. Neurophysiol"},{"key":"B23","doi-asserted-by":"publisher","first-page":"932","DOI":"10.1002\/mrm.29014","article-title":"Training data distribution significantly impacts the estimation of tissue microstructure with machine learning","volume":"87","author":"Gyori","year":"2022","journal-title":"Magn. Reson. Med"},{"key":"B24","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1016\/0020-7101(85)90084-4","article-title":"EEG waveform analysis by means of dynamic time-warping","volume":"17","author":"Huang","year":"1985","journal-title":"Int. J. Biomed. Comput"},{"key":"B25","doi-asserted-by":"publisher","first-page":"066011","DOI":"10.1088\/1741-2552\/aadea0","article-title":"Moabb: trustworthy algorithm benchmarking for bcis","volume":"15","author":"Jayaram","year":"2018","journal-title":"J. Neural Eng"},{"key":"B26","doi-asserted-by":"publisher","first-page":"4112","DOI":"10.3390\/s23084112","article-title":"A shallow autoencoder framework for epileptic seizure detection in eeg signals","volume":"23","author":"Khan","year":"2023","journal-title":"Sensors"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1312.6114","article-title":"Auto-encoding variational bayes","author":"Kingma","year":"2013","journal-title":"arXiv"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1561\/9781680836233","article-title":"An introduction to variational autoencoders","author":"Kingma","year":"2019","journal-title":"arXiv"},{"key":"B29","doi-asserted-by":"publisher","first-page":"6573","DOI":"10.1093\/cercor\/bhac525","article-title":"Thirty-minute motor imagery exercise aided by EEG sensorimotor rhythm neurofeedback enhances morphing of sensorimotor cortices: a double-blind sham-controlled study","volume":"33","author":"Kodama","year":"2023","journal-title":"Cereb. Cortex"},{"key":"B30","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1109\/T-AFFC.2011.15","article-title":"Deap: A database for emotion analysis; using physiological signals","volume":"3","author":"Koelstra","year":"2012","journal-title":"IEEE Transact. Affect. Comp"},{"key":"B31","doi-asserted-by":"publisher","first-page":"aace8c","DOI":"10.1088\/1741-2552\/aace8c","article-title":"EEGNet: a compact convolutional network for EEG-based Brain-Computer Interfaces","volume":"15","author":"Lawhern","year":"2016","journal-title":"J. Neural Eng"},{"key":"B32","doi-asserted-by":"publisher","first-page":"448","DOI":"10.1016\/j.procs.2023.08.183","article-title":"Learning an autoencoder to compress EEG signals via a neural network based approximation of DTW","volume":"222","author":"Lerogeron","year":"","journal-title":"Proc. Comp. Sci"},{"key":"B33","doi-asserted-by":"publisher","first-page":"162","DOI":"10.1016\/j.patrec.2023.05.012","article-title":"Approximating dynamic time warping with a convolutional neural network on eeg data","volume":"171","author":"Lerogeron","year":"","journal-title":"Pattern Recognit. Lett"},{"key":"B34","doi-asserted-by":"publisher","first-page":"132720","DOI":"10.1109\/ACCESS.2019.2941867","article-title":"Densely feature fusion based on convolutional neural networks for motor imagery EEG classification","volume":"7","author":"Li","year":"2019","journal-title":"IEEE Access"},{"key":"B35","author":"Li","year":"2022","journal-title":"Constructing Large-Scale Real-World Benchmark Datasets for AIOPS"},{"key":"B36","doi-asserted-by":"publisher","first-page":"43","DOI":"10.3389\/fnsys.2020.00043","article-title":"EEG-based emotion classification using a deep neural network and sparse autoencoder","volume":"14","author":"Liu","year":"2020","journal-title":"Front. Syst. Neurosci"},{"key":"B37","doi-asserted-by":"publisher","first-page":"031005","DOI":"10.1088\/1741-2552\/aab2f2","article-title":"A review of classification algorithms for EEG-based brain-computer interfaces: a 10 year update","volume":"15","author":"Lotte","year":"2018","journal-title":"J. Neural Eng"},{"key":"B38","volume-title":"Deep Recurrent Networks for Gesture Recognition and Synthesis","author":"Maghoumi","year":"2020"},{"key":"B39","first-page":"213","article-title":"\u201cDeepNAG: deep non-adversarial gesture generation,\u201d","author":"Maghoumi","year":"2021","journal-title":"26th International Conference on Intelligent User Interfaces"},{"key":"B40","doi-asserted-by":"publisher","first-page":"294","DOI":"10.1080\/24751839.2019.1565653","article-title":"Machine learning approaches for anomaly detection of water quality on a real-world data set","volume":"3","author":"Muharemi","year":"2019","journal-title":"J. Inf. Telecommun"},{"key":"B41","doi-asserted-by":"crossref","DOI":"10.1109\/GLOBECOM54140.2023.10437773","article-title":"\u201cLocal or edge\/cloud processing for data freshness,\u201d","volume-title":"IEEE Global Communications Conference, GLOBECOM","author":"Munari","year":"2023"},{"key":"B42","doi-asserted-by":"publisher","first-page":"7134","DOI":"10.1038\/s41598-019-43594-9","article-title":"Attempted arm and hand movements can be decoded from low-frequency eeg from persons with spinal cord injury","volume":"9","author":"Ofner","year":"2019","journal-title":"Sci. Rep"},{"key":"B43","doi-asserted-by":"publisher","first-page":"2050029","DOI":"10.1142\/S012906572050029X","article-title":"Dyslexia diagnosis by eeg temporal and spectral descriptors: an anomaly detection approach","volume":"30","author":"Ortiz","year":"2020","journal-title":"Int. J. Neural Syst"},{"key":"B44","doi-asserted-by":"publisher","first-page":"3439950","DOI":"10.1145\/3439950","article-title":"Deep learning for anomaly detection: a review","volume":"54","author":"Pang","year":"2021","journal-title":"ACM Comput. Surv"},{"key":"B45","doi-asserted-by":"publisher","first-page":"2825","DOI":"10.5555\/1953048.2078195","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res"},{"key":"B46","doi-asserted-by":"publisher","first-page":"181","DOI":"10.1016\/j.neuroimage.2019.05.026","article-title":"ICLabel: an automated electroencephalographic independent component classifier, dataset, and website","volume":"198","author":"Pion-Tonachini","year":"2019","journal-title":"Neuroimage"},{"key":"B47","doi-asserted-by":"publisher","DOI":"10.2139\/ssrn.4199295","article-title":"Diverse super-resolution with pretrained deep hiererarchical vaes","author":"Prost","year":"2022","journal-title":"arXiv"},{"key":"B48","doi-asserted-by":"publisher","first-page":"1350","DOI":"10.3390\/sym11111350","article-title":"KPI-TSAD: a time-series anomaly detector for KPI monitoring in cloud applications","volume":"11","author":"Qiu","year":"2019","journal-title":"Symmetry"},{"key":"B49","article-title":"\u201cGenerating diverse highfidelity images with VQ-VAE-2,\u201d","volume-title":"Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019","author":"Razavi","year":"2019"},{"key":"B50","doi-asserted-by":"publisher","first-page":"109037","DOI":"10.1016\/j.jneumeth.2020.109037","article-title":"MI-EEGNET: a novel convolutional neural network for motor imagery classification","volume":"353","author":"Riyad","year":"2021","journal-title":"J. Neurosci. Methods"},{"key":"B51","doi-asserted-by":"publisher","first-page":"5619","DOI":"10.1109\/TNNLS.2018.2789927","article-title":"Learning temporal information for brain-computer interface using convolutional neural networks","volume":"29","author":"Sakhavi","year":"2018","journal-title":"IEEE Transact. Neural Netw. Learn. Syst"},{"key":"B52","doi-asserted-by":"publisher","first-page":"159","DOI":"10.1016\/B978-0-08-051584-7.50016-4","article-title":"Dynamic programming algorithm optimization for spoken word recognition","volume":"26","author":"Sakoe","year":"1978","journal-title":"IEEE Trans. Acoust"},{"key":"B53","doi-asserted-by":"crossref","first-page":"166","DOI":"10.1109\/ICDCSW.2011.20","article-title":"\u201cFinding a \u201ckneedle? in a haystack: detecting knee points in system behavior,\u201d","volume-title":"2011 31st International Conference on Distributed Computing Systems Workshops","author":"Satopaa","year":"2011"},{"key":"B54","doi-asserted-by":"publisher","first-page":"5391","DOI":"10.1002\/hbm.23730","article-title":"Deep learning with convolutional neural networks for EEG decoding and visualization","volume":"38","author":"Schirrmeister","year":"2017","journal-title":"Hum. Brain Mapp."},{"key":"B55","doi-asserted-by":"publisher","first-page":"066054","DOI":"10.1088\/1741-2552\/ac42b5","article-title":"Neural tracking to go: auditory attention decoding and saliency detection with mobile EEG","volume":"18","author":"Straetmans","year":"2022","journal-title":"J. Neural Eng"},{"key":"B56","first-page":"1","article-title":"\u201cFundamentals of EEG measurement,\u201d","author":"Teplan","year":"2002","journal-title":"IEEE Measurement Science Review, Vol. 2"},{"key":"B57","article-title":"\u201cNVAE: a deep hierarchical variational autoencoder,\u201d","author":"Vahdat","year":"2021","journal-title":"Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020"},{"key":"B58","doi-asserted-by":"publisher","first-page":"106405","DOI":"10.1016\/j.knosys.2020.106405","article-title":"Long-range forecasting in feature-evolving data streams","volume":"206","author":"Wambura","year":"2020","journal-title":"Knowl. Based Syst"},{"key":"B59","doi-asserted-by":"publisher","first-page":"105916","DOI":"10.1016\/j.bspc.2023.105916","article-title":"Multifractal organization of eeg signals in multiple sclerosis","volume":"91","author":"Watorek","year":"2024","journal-title":"Biomed. Signal Process. Control"},{"key":"B60","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1109\/TAU.1967.1161901","article-title":"The use of fast fourier transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms","volume":"15","author":"Welch","year":"1967","journal-title":"IEEE Transact. Audio Electroacoust"},{"key":"B61","doi-asserted-by":"publisher","DOI":"10.1007\/s00521-019-04288-5","article-title":"Identifying data streams anomalies by evolving spiking restricted boltzmann machines","author":"Xing","year":"2020","journal-title":"Neural Comp. Appl"},{"key":"B62","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-62753-8_7","article-title":"\u201cvEEGNet: learning latent representations to reconstruct EEG raw data via variational autoencoders,\u201d","volume-title":"Springer Communications in Computer and Information Science Series","author":"Zancanaro","year":""},{"key":"B63","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/CIBCB49929.2021.9562821","article-title":"\u201cCNN-based approaches for cross-subject classification in motor imagery: from the state-ofthe-art to DynamicNet,\u201d","volume-title":"2021 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB)","author":"Zancanaro","year":"2021"},{"key":"B64","first-page":"245","article-title":"\u201cvEEGNet: a new deep learning model to classify and generate EEG,\u201d","volume-title":"Proceedings of the 9th International Conference on Information and Communication Technologies for Ageing Well and e-Health - Volume 1: ICT4AWE, INSTICC","author":"Zancanaro","year":""},{"key":"B65","doi-asserted-by":"publisher","first-page":"162","DOI":"10.1109\/TAMD.2015.2431497","article-title":"Investigating critical frequency bands and channels for EEG-based emotion recognition with deep neural networks","volume":"7","author":"Zheng","year":"2015","journal-title":"IEEE Trans. Auton. Ment. Dev"},{"key":"B66","doi-asserted-by":"publisher","first-page":"0162657","DOI":"10.1371\/journal.pone.0162657","article-title":"A fully automated trial selection method for optimization of motor imagery based brain-computer interface","volume":"11","author":"Zhou","year":"2016","journal-title":"PLoS ONE"},{"key":"B67","first-page":"665","article-title":"\u201cAnomaly detection with robust deep autoencoders,\u201d","author":"Zhou","year":"2017"}],"container-title":["Frontiers in Neuroinformatics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fninf.2024.1459970\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,12,20]],"date-time":"2024-12-20T06:30:49Z","timestamp":1734676249000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fninf.2024.1459970\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,12,20]]},"references-count":67,"alternative-id":["10.3389\/fninf.2024.1459970"],"URL":"https:\/\/doi.org\/10.3389\/fninf.2024.1459970","relation":{},"ISSN":["1662-5196"],"issn-type":[{"value":"1662-5196","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,12,20]]},"article-number":"1459970"}}