2026-07-15T20:19:57Zhttps://uvadoc.uva.es/oai/request

oai:uvadoc.uva.es:10324/615072025-02-20T11:39:39Zcom_10324_23459com_10324_954com_10324_894col_10324_23460

Vaquerizo Villar, Fernando 6d198fa6faee87eb 600 0000-0002-5898-2006 Gutierrez Tobal, Gonzalo César ab1f68670e20dbd2 600 Calvo, Eva 3784c178-4b71-423e-b673-d44ecf923f14 Álvarez González, Daniel 6c685c5708ff8b75 600 0000-0003-1027-2395 Kheirandish Gozal, Leila f810daa8-4fa5-448e-9f05-ea00aa529bfe 600 Campo Matias, Félix del ca3583181f1300f5 600 0000-0002-4554-2167 Hornero Sánchez, Roberto f6af2dd4a94089d7 600 0000-0001-9915-2570 2023-09-11T12:54:50Z 2023-09-11T12:54:50Z 2023 Computers in Biology and Medicine, 2023, vol. 165, 107419 0010-4825 https://uvadoc.uva.es/handle/10324/61507 10.1016/j.compbiomed.2023.107419 107419 Computers in Biology and Medicine 165 Producción Científica Automatic deep-learning models used for sleep scoring in children with obstructive sleep apnea (OSA) are perceived as black boxes, limiting their implementation in clinical settings. Accordingly, we aimed to develop an accurate and interpretable deep-learning model for sleep staging in children using single-channel electroencephalogram (EEG) recordings. We used EEG signals from the Childhood Adenotonsillectomy Trial (CHAT) dataset (n = 1637) and a clinical sleep database (n = 980). Three distinct deep-learning architectures were explored to automatically classify sleep stages from a single-channel EEG data. Gradient-weighted Class Activation Mapping (Grad-CAM), an explainable artificial intelligence (XAI) algorithm, was then applied to provide an interpretation of the singular EEG patterns contributing to each predicted sleep stage. Among the tested architectures, a standard convolutional neural network (CNN) demonstrated the highest performance for automated sleep stage detection in the CHAT test set (accuracy = 86.9% and five-class kappa = 0.827). Furthermore, the CNN-based estimation of total sleep time exhibited strong agreement in the clinical dataset (intra-class correlation coefficient = 0.772). Our XAI approach using Grad-CAM effectively highlighted the EEG features associated with each sleep stage, emphasizing their influence on the CNN's decision-making process in both datasets. Grad-CAM heatmaps also allowed to identify and analyze epochs within a recording with a highly likelihood to be misclassified, revealing mixed features from different sleep stages within these epochs. Finally, Grad-CAM heatmaps unveiled novel features contributing to sleep scoring using a single EEG channel. Consequently, integrating an explainable CNN-based deep-learning model in the clinical environment could enable automatic sleep staging in pediatric sleep apnea tests. Ministerio de Ciencia e Innovación- Agencia Estatal de Investigación- FEDER-EU y NextGenerationEU/PRTR (PID2020-115468RB-I00 y PDC2021-120775-I00) Sociedad Española de Neumología y Cirugía Torácica (SEPAR) (649/2018) CIBER -Consorcio Centro de Investigación Biomédica en Red- Instituto de Salud Carlos III (CB19/01/00012) Institutos Nacionales de Salud (HL083075, HL083129, UL1-RR-024134, UL1 RR024989) Instituto Nacional del Corazón, los Pulmones y la Sangre (R24 HL114473, 75N92019R002) Ministerio de Ciencia e Innovación - Agencia Estatal de Investigación- “Ramón y Cajal” grant (RYC2019-028566-I) application/pdf eng Elsevier info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by-nc-nd/4.0/ © 2023 The Authors Attribution-NonCommercial-NoDerivatives 4.0 Internacional Sueño, trastornos del Pediatría Deep learning Electroencephalogram (EEG) Pediatric obstructive sleep apnea (OSA) Aprendizaje profundo Electroencefalograma (EEG) Apnea obstructiva del sueño pediátrica (AOS) 3201.10 Pediatría An explainable deep-learning model to stage sleep states in children and propose novel EEG-related patterns in sleep apnea info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion https://www.sciencedirect.com/science/article/pii/S0010482523008843?via%3Dihub SI THUMBNAIL An-explainable-deep-learning.pdf.jpg An-explainable-deep-learning.pdf.jpg IM Thumbnail image/jpeg 2187 https://uvadoc.uva.es/bitstream/10324/61507/4/An-explainable-deep-learning.pdf.jpg b9773cca2e5ff1921cda2ce31b49457c MD5 4 LICENSE license.txt license.txt text/plain 3929 https://uvadoc.uva.es/bitstream/10324/61507/3/license.txt 289d98c1665ee0d20312360f0e8643c8 MD5 3 CC-LICENSE license_rdf license_rdf application/rdf+xml; charset=utf-8 817 https://uvadoc.uva.es/bitstream/10324/61507/2/license_rdf 975848b171c53b753f7e445c8b5bc0e1 MD5 2 ORIGINAL An-explainable-deep-learning.pdf An-explainable-deep-learning.pdf application/pdf 13127212 https://uvadoc.uva.es/bitstream/10324/61507/1/An-explainable-deep-learning.pdf d5fa6b68e9df31778a0833d702e96ccb MD5 1 10324/61507 oai:uvadoc.uva.es:10324/61507 2025-02-20 12:39:39.594 UVaDOC repositorio@uva.es 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