{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/chrononet-a-deep-recurrent-neural-network-for","title":"ChronoNet: A Deep Recurrent Neural Network for Abnormal EEG Identification","arxiv_id":"1802.00308","date":"2018-01-30","proceeding":null,"authors":["Subhrajit Roy","Isabell Kiral-Kornek","Stefan Harrer"],"abstract":"Brain-related disorders such as epilepsy can be diagnosed by analyzing\nelectroencephalograms (EEG). However, manual analysis of EEG data requires\nhighly trained clinicians, and is a procedure that is known to have relatively\nlow inter-rater agreement (IRA). Moreover, the volume of the data and the rate\nat which new data becomes available make manual interpretation a\ntime-consuming, resource-hungry, and expensive process. In contrast, automated\nanalysis of EEG data offers the potential to improve the quality of patient\ncare by shortening the time to diagnosis and reducing manual error. In this\npaper, we focus on one of the first steps in interpreting an EEG session -\nidentifying whether the brain activity is abnormal or normal. To solve this\ntask, we propose a novel recurrent neural network (RNN) architecture termed\nChronoNet which is inspired by recent developments from the field of image\nclassification and designed to work efficiently with EEG data. ChronoNet is\nformed by stacking multiple 1D convolution layers followed by deep gated\nrecurrent unit (GRU) layers where each 1D convolution layer uses multiple\nfilters of exponentially varying lengths and the stacked GRU layers are densely\nconnected in a feed-forward manner. We used the recently released TUH Abnormal\nEEG Corpus dataset for evaluating the performance of ChronoNet. Unlike previous\nstudies using this dataset, ChronoNet directly takes time-series EEG as input\nand learns meaningful representations of brain activity patterns. ChronoNet\noutperforms the previously reported best results by 7.79% thereby setting a new\nbenchmark for this dataset. Furthermore, we demonstrate the domain-independent\nnature of ChronoNet by successfully applying it to classify speech commands.","url_abs":"http://arxiv.org/abs/1802.00308v2","url_pdf":"http://arxiv.org/pdf/1802.00308v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"chrononet-a-deep-recurrent-neural-network-for","repo_url":"https://github.com/Sharad24/Epileptic-Seizure-Detection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"chrononet-a-deep-recurrent-neural-network-for","repo_url":"https://github.com/aguscerdo/EE239AS-Project","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"eeg-1","task_name":"EEG"},{"task_slug":"eeg","task_name":"Electroencephalogram (EEG)"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"time-series","task_name":"Time Series Analysis"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"gru","method_name":"GRU"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1802.00308","atlas_url":"https://app.syntology.ai/?focus=1802.00308","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}