SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosis

Obstructive sleep apnea (OSA) in children is a prevalent and serious respiratory condition linked to cardiovascular morbidity. Polysomnography, the standard diagnostic approach, faces challenges in accessibility and complexity, leading to underdiagnosis. To simplify OSA diagnosis, deep learning (DL)...

Descripción completa

Detalles Bibliográficos
Autores: García Vicente, Clara, Gutiérrez Tobal, Gonzalo César, Vaquerizo Villar, Fernando, Martín Montero, Adrián, Gozal, David, Hornero Sánchez, Roberto
Tipo de recurso: artículo
Fecha de publicación:2025
País:España
Institución:Universidad de Cantabria (UC)
Repositorio:UCrea Repositorio Abierto de la Universidad de Cantabria
Idioma:inglés
OAI Identifier:oai:repositorio.unican.es:10902/39108
Acceso en línea:https://hdl.handle.net/10902/39108
Access Level:acceso abierto
Palabra clave:Pediatric obstructive sleep apnea (OSA)
Deep learning (DL)
eXplainable artificial intelligence (XAI)
Electrocardiogram (ECG)
Gradient-weighted class activation mapping (Grad-CAM)
id ES_538781668bdcdbc841e6a3dba92c9b0f
oai_identifier_str oai:repositorio.unican.es:10902/39108
network_acronym_str ES
network_name_str España
repository_id_str
spelling SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosisGarcía Vicente, ClaraGutiérrez Tobal, Gonzalo CésarVaquerizo Villar, FernandoMartín Montero, AdriánGozal, DavidHornero Sánchez, RobertoPediatric obstructive sleep apnea (OSA)Deep learning (DL)eXplainable artificial intelligence (XAI)Electrocardiogram (ECG)Gradient-weighted class activation mapping (Grad-CAM)Obstructive sleep apnea (OSA) in children is a prevalent and serious respiratory condition linked to cardiovascular morbidity. Polysomnography, the standard diagnostic approach, faces challenges in accessibility and complexity, leading to underdiagnosis. To simplify OSA diagnosis, deep learning (DL) algorithms have been developed using cardiac signals, but they often lack interpretability. Our study introduces a novel interpretable DL approach (SleepECG-Net) for directly estimating OSA severity in at-risk children. A combination of convolutional and recurrent neural networks (CNN-RNN) was trained on overnight electrocardiogram (ECG) signals. Gradient-weighted Class Activation Mapping (Grad-CAM), an eXplainable Artificial Intelligence (XAI) algorithm, was applied to explain model decisions and extract ECG patterns relevant to pediatric OSA. Accordingly, ECG signals from the semi-public Childhood Adenotonsillectomy Trial (CHAT, n = 1610) and Cleveland Family Study (CFS,n = 64), and the private University of Chicago (UofC, n = 981) databases were used. OSA diagnostic performance reached 4-class Cohen's Kappa of 0.410, 0.335, and 0.249 in CHAT, UofC, and CFS, respectively. The proposal demonstrated improved performance with increased severity along with heightened cardiovascular risk. XAI findings highlighted the detection of established ECG features linked to OSA, such as bradycardia-tachycardia events and delayed ECG patterns during apnea/hypopnea occurrences, focusing on clusters of events. Furthermore, Grad-CAM heatmaps identified potential ECG patterns indicating cardiovascular risk, such as P, T, and U waves, QT intervals, and QRS complex variations. Hence, SleepECG-Net approach may improve pediatric OSA diagnosis by also offering cardiac risk factor information, thereby increasing clinician confidence in automated systems, and promoting their effective adoption in clinical practice.This work is part of the projects PID2020-115468RB-I00 and CPP2022- 009735, funded by MCIN/AEI/10.13039/501100011033 and the European Union “NextGenerationEU”/PRTR. This research was also co-funded by the European Union through the Interreg VI-A Spain-Portugal Program (POCTEP) 2021-2027 (0043_NET4SLEEP_2_E), and by “CIBER-Consorcio Centro de Investigación Biomédica en Red” (CB19/01/00012) through “Instituto de Salud Carlos III”, co-funded with European Regional Development Fund, as well as under the project TinyHeart from 2022 Early Stage call. The Childhood Adenotonsillectomy Trial (CHAT) was supported by the National Institutes of Health (HL083075, HL083129, UL1-RR-024134, UL1 RR024989). The Cleveland Family Study (CFS) was supported by grants from the National Institutes of Health (HL46380, M01 RR00080-39, T32-HL07567, RO1- 46380). The National Sleep Research Resource was supported by the National Heart, Lung, and Blood Institute (R24 HL114473, 75N92019R002).Institute of Electrical and Electronics Engineers, Inc.Universidad de Cantabria20252025-02-01journal articlehttp://purl.org/coar/resource_type/c_6501NAhttp://purl.org/coar/version/c_be7fb7dd8ff6fe43info:eu-repo/semantics/articlehttps://hdl.handle.net/10902/39108IEEE Journal of Biomedical and Health Informatics, 2025, 29(2), 1021-1034reponame:UCrea Repositorio Abierto de la Universidad de Cantabriainstname:Universidad de Cantabria (UC)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2info:eu-repo/semantics/openAccessoai:repositorio.unican.es:10902/391082026-06-02T12:39:31Z
dc.title.none.fl_str_mv SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosis
title SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosis
spellingShingle SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosis
García Vicente, Clara
Pediatric obstructive sleep apnea (OSA)
Deep learning (DL)
eXplainable artificial intelligence (XAI)
Electrocardiogram (ECG)
Gradient-weighted class activation mapping (Grad-CAM)
title_short SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosis
title_full SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosis
title_fullStr SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosis
title_full_unstemmed SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosis
title_sort SleepECG-Net: explainable deep learning approach with ECG for pediatric sleep apnea diagnosis
dc.creator.none.fl_str_mv García Vicente, Clara
Gutiérrez Tobal, Gonzalo César
Vaquerizo Villar, Fernando
Martín Montero, Adrián
Gozal, David
Hornero Sánchez, Roberto
author García Vicente, Clara
author_facet García Vicente, Clara
Gutiérrez Tobal, Gonzalo César
Vaquerizo Villar, Fernando
Martín Montero, Adrián
Gozal, David
Hornero Sánchez, Roberto
author_role author
author2 Gutiérrez Tobal, Gonzalo César
Vaquerizo Villar, Fernando
Martín Montero, Adrián
Gozal, David
Hornero Sánchez, Roberto
author2_role author
author
author
author
author
dc.contributor.none.fl_str_mv Universidad de Cantabria
dc.subject.none.fl_str_mv Pediatric obstructive sleep apnea (OSA)
Deep learning (DL)
eXplainable artificial intelligence (XAI)
Electrocardiogram (ECG)
Gradient-weighted class activation mapping (Grad-CAM)
topic Pediatric obstructive sleep apnea (OSA)
Deep learning (DL)
eXplainable artificial intelligence (XAI)
Electrocardiogram (ECG)
Gradient-weighted class activation mapping (Grad-CAM)
description Obstructive sleep apnea (OSA) in children is a prevalent and serious respiratory condition linked to cardiovascular morbidity. Polysomnography, the standard diagnostic approach, faces challenges in accessibility and complexity, leading to underdiagnosis. To simplify OSA diagnosis, deep learning (DL) algorithms have been developed using cardiac signals, but they often lack interpretability. Our study introduces a novel interpretable DL approach (SleepECG-Net) for directly estimating OSA severity in at-risk children. A combination of convolutional and recurrent neural networks (CNN-RNN) was trained on overnight electrocardiogram (ECG) signals. Gradient-weighted Class Activation Mapping (Grad-CAM), an eXplainable Artificial Intelligence (XAI) algorithm, was applied to explain model decisions and extract ECG patterns relevant to pediatric OSA. Accordingly, ECG signals from the semi-public Childhood Adenotonsillectomy Trial (CHAT, n = 1610) and Cleveland Family Study (CFS,n = 64), and the private University of Chicago (UofC, n = 981) databases were used. OSA diagnostic performance reached 4-class Cohen's Kappa of 0.410, 0.335, and 0.249 in CHAT, UofC, and CFS, respectively. The proposal demonstrated improved performance with increased severity along with heightened cardiovascular risk. XAI findings highlighted the detection of established ECG features linked to OSA, such as bradycardia-tachycardia events and delayed ECG patterns during apnea/hypopnea occurrences, focusing on clusters of events. Furthermore, Grad-CAM heatmaps identified potential ECG patterns indicating cardiovascular risk, such as P, T, and U waves, QT intervals, and QRS complex variations. Hence, SleepECG-Net approach may improve pediatric OSA diagnosis by also offering cardiac risk factor information, thereby increasing clinician confidence in automated systems, and promoting their effective adoption in clinical practice.
publishDate 2025
dc.date.none.fl_str_mv 2025
2025-02-01
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
NA
http://purl.org/coar/version/c_be7fb7dd8ff6fe43
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/10902/39108
url https://hdl.handle.net/10902/39108
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
eu_rights_str_mv openAccess
dc.publisher.none.fl_str_mv Institute of Electrical and Electronics Engineers, Inc.
publisher.none.fl_str_mv Institute of Electrical and Electronics Engineers, Inc.
dc.source.none.fl_str_mv IEEE Journal of Biomedical and Health Informatics, 2025, 29(2), 1021-1034
reponame:UCrea Repositorio Abierto de la Universidad de Cantabria
instname:Universidad de Cantabria (UC)
instname_str Universidad de Cantabria (UC)
reponame_str UCrea Repositorio Abierto de la Universidad de Cantabria
collection UCrea Repositorio Abierto de la Universidad de Cantabria
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869408122810925056
score 15.812455

Similares en LA Referencia