Automatic analysis of EEGs using big data and hybrid deep learning architectures
dc.creator | Golmohammadi, M | |
dc.creator | Harati Nejad Torbati, AH | |
dc.creator | Lopez de Diego, S | |
dc.creator | Obeid, I | |
dc.creator | Picone, J | |
dc.date.accessioned | 2020-12-16T18:10:56Z | |
dc.date.available | 2020-12-16T18:10:56Z | |
dc.date.issued | 2019-02-01 | |
dc.identifier.issn | 1662-5161 | |
dc.identifier.issn | 1662-5161 | |
dc.identifier.doi | http://dx.doi.org/10.34944/dspace/4558 | |
dc.identifier.other | HO5AU (isidoc) | |
dc.identifier.other | 30914936 (pubmed) | |
dc.identifier.uri | http://hdl.handle.net/20.500.12613/4576 | |
dc.description.abstract | © 2019 Golmohammadi, Harati Nejad Torbati, Lopez de Diego, Obeid and Picone. Brain monitoring combined with automatic analysis of EEGs provides a clinical decision support tool that can reduce time to diagnosis and assist clinicians in real-time monitoring applications (e.g., neurological intensive care units). Clinicians have indicated that a sensitivity of 95% with specificity below 5% was the minimum requirement for clinical acceptance. In this study, a high-performance automated EEG analysis system based on principles of machine learning and big data is proposed. This hybrid architecture integrates hidden Markov models (HMMs) for sequential decoding of EEG events with deep learning-based post-processing that incorporates temporal and spatial context. These algorithms are trained and evaluated using the Temple University Hospital EEG, which is the largest publicly available corpus of clinical EEG recordings in the world. This system automatically processes EEG records and classifies three patterns of clinical interest in brain activity that might be useful in diagnosing brain disorders: (1) spike and/or sharp waves, (2) generalized periodic epileptiform discharges, (3) periodic lateralized epileptiform discharges. It also classifies three patterns used to model the background EEG activity: (1) eye movement, (2) artifacts, and (3) background. Our approach delivers a sensitivity above 90% while maintaining a specificity below 5%. We also demonstrate that this system delivers a low false alarm rate, which is critical for any spike detection application. | |
dc.format.extent | 76- | |
dc.language.iso | eng | |
dc.relation.haspart | Frontiers in Human Neuroscience | |
dc.relation.isreferencedby | Frontiers Media SA | |
dc.rights | CC BY | |
dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | |
dc.subject | electroencephalography | |
dc.subject | EEG | |
dc.subject | hidden markov models | |
dc.subject | HMM | |
dc.subject | deep learning | |
dc.subject | stochastic denoising autoencoders | |
dc.subject | SdA | |
dc.subject | automatic detection | |
dc.title | Automatic analysis of EEGs using big data and hybrid deep learning architectures | |
dc.type | Article | |
dc.type.genre | Journal Article | |
dc.relation.doi | 10.3389/fnhum.2019.00076 | |
dc.ada.note | For Americans with Disabilities Act (ADA) accommodation, including help with reading this content, please contact scholarshare@temple.edu | |
dc.creator.orcid | Obeid, Iyad|0000-0002-5796-843X | |
dc.date.updated | 2020-12-16T18:10:52Z | |
refterms.dateFOA | 2020-12-16T18:10:57Z |