Semi-supervised active transfer learning for fetal ECG arrhythmia detection

Deep learning has demonstrated excellent results for ECG anomaly detection, wherein most approaches used supervised learning. The requirement of thousands of manually annotated samples is a concern for state-of-the-art anomaly detection systems, especially for fetal ECG (FECG), and currently, there...

Descripción completa

Detalles Bibliográficos
Autores: Mohebbian, Mohammad Reza, Marateb, Hamid Reza|||0000-0003-4408-2397, Wahid, Khan A.
Tipo de recurso: artículo
Fecha de publicación:2023
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/410000
Acceso en línea:https://hdl.handle.net/2117/410000
https://dx.doi.org/10.1016/j.cmpbup.2023.100096
Access Level:acceso abierto
Palabra clave:Fetal monitoring
Electrocardiography
FECG
Arrhythmia
Transfer learning
Active learning
Anomaly
Monitoratge fetal
Electrocardiografia
Àrees temàtiques de la UPC::Enginyeria biomèdica::Electrònica biomèdica
id ES_a42e2a49ef58a1aedc1cefbd28d620f1
oai_identifier_str oai:upcommons.upc.edu:2117/410000
network_acronym_str ES
network_name_str España
repository_id_str
spelling Semi-supervised active transfer learning for fetal ECG arrhythmia detectionMohebbian, Mohammad RezaMarateb, Hamid Reza|||0000-0003-4408-2397Wahid, Khan A.Fetal monitoringElectrocardiographyFECGArrhythmiaTransfer learningActive learningAnomalyMonitoratge fetalElectrocardiografiaÀrees temàtiques de la UPC::Enginyeria biomèdica::Electrònica biomèdicaDeep learning has demonstrated excellent results for ECG anomaly detection, wherein most approaches used supervised learning. The requirement of thousands of manually annotated samples is a concern for state-of-the-art anomaly detection systems, especially for fetal ECG (FECG), and currently, there is not a publicly available FECG dataset annotated for each FECG beat. In this paper, we offer a modified active learning technique based on transfer learning, calibration probability, and autoencoder-based sampling to reduce number of samples requires to annotate. In this regard, we used 25,000 s of recording from 47 patients from the MIT-BIH Arrhythmia Database to train a deep learning model to detect anomalies in non-fetus subjects. Then we used the unlabeled Non-Invasive Fetal ECG Arrhythmia Database (NIFEA DB) of 26 subjects to fine-tune the trained model to fine-tune the trained model based on active learning to detect anomalies in binary form for fetal. A variational autoencoder is trained on all data (adult and fetal ECG), and clustering is applied to latent features extracted from data after dimension reduction. Then, the sampling process of active learning selected samples from different clusters with low confidence to cover all data distribution. Moreover, a probability calibration based on mc-dropout and isotonic regression is used to calibrate confidences, helping to select reliable low-confidence samples. Various ablation studies were performed to show the influence of autoencoder-based sampling, calibration, and transfer learning, which showed that the proposed method could achieve 92% accuracy using 399 training samples. In contrast, other methods required more training samples to reach the same level of accuracy without calibration or an autoencoder and clustering approach or training without active learning. The study also found that transfer learning significantly impacted faster convergence and that the proposed active learning approach was more effective than traditional methods.The authors thank Dr Samani for annotating data and acknowledge funding from the Natural Sciences and Engineering Research Council of Canada (NSERC) to support the work.Peer ReviewedElsevier20232023-01-0120242024-06-14journal articlehttp://purl.org/coar/resource_type/c_6501VoRhttp://purl.org/coar/version/c_970fb48d4fbd8a85info:eu-repo/semantics/articleapplication/pdfhttps://hdl.handle.net/2117/410000https://dx.doi.org/10.1016/j.cmpbup.2023.100096reponame:UPCommons. Portal del coneixement obert de la UPCinstname:Universitat Politècnica de Catalunya (UPC)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2Attribution 4.0 Internationalhttp://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccessoai:upcommons.upc.edu:2117/4100002026-05-27T15:37:01Z
dc.title.none.fl_str_mv Semi-supervised active transfer learning for fetal ECG arrhythmia detection
title Semi-supervised active transfer learning for fetal ECG arrhythmia detection
spellingShingle Semi-supervised active transfer learning for fetal ECG arrhythmia detection
Mohebbian, Mohammad Reza
Fetal monitoring
Electrocardiography
FECG
Arrhythmia
Transfer learning
Active learning
Anomaly
Monitoratge fetal
Electrocardiografia
Àrees temàtiques de la UPC::Enginyeria biomèdica::Electrònica biomèdica
title_short Semi-supervised active transfer learning for fetal ECG arrhythmia detection
title_full Semi-supervised active transfer learning for fetal ECG arrhythmia detection
title_fullStr Semi-supervised active transfer learning for fetal ECG arrhythmia detection
title_full_unstemmed Semi-supervised active transfer learning for fetal ECG arrhythmia detection
title_sort Semi-supervised active transfer learning for fetal ECG arrhythmia detection
dc.creator.none.fl_str_mv Mohebbian, Mohammad Reza
Marateb, Hamid Reza|||0000-0003-4408-2397
Wahid, Khan A.
author Mohebbian, Mohammad Reza
author_facet Mohebbian, Mohammad Reza
Marateb, Hamid Reza|||0000-0003-4408-2397
Wahid, Khan A.
author_role author
author2 Marateb, Hamid Reza|||0000-0003-4408-2397
Wahid, Khan A.
author2_role author
author
dc.subject.none.fl_str_mv Fetal monitoring
Electrocardiography
FECG
Arrhythmia
Transfer learning
Active learning
Anomaly
Monitoratge fetal
Electrocardiografia
Àrees temàtiques de la UPC::Enginyeria biomèdica::Electrònica biomèdica
topic Fetal monitoring
Electrocardiography
FECG
Arrhythmia
Transfer learning
Active learning
Anomaly
Monitoratge fetal
Electrocardiografia
Àrees temàtiques de la UPC::Enginyeria biomèdica::Electrònica biomèdica
description Deep learning has demonstrated excellent results for ECG anomaly detection, wherein most approaches used supervised learning. The requirement of thousands of manually annotated samples is a concern for state-of-the-art anomaly detection systems, especially for fetal ECG (FECG), and currently, there is not a publicly available FECG dataset annotated for each FECG beat. In this paper, we offer a modified active learning technique based on transfer learning, calibration probability, and autoencoder-based sampling to reduce number of samples requires to annotate. In this regard, we used 25,000 s of recording from 47 patients from the MIT-BIH Arrhythmia Database to train a deep learning model to detect anomalies in non-fetus subjects. Then we used the unlabeled Non-Invasive Fetal ECG Arrhythmia Database (NIFEA DB) of 26 subjects to fine-tune the trained model to fine-tune the trained model based on active learning to detect anomalies in binary form for fetal. A variational autoencoder is trained on all data (adult and fetal ECG), and clustering is applied to latent features extracted from data after dimension reduction. Then, the sampling process of active learning selected samples from different clusters with low confidence to cover all data distribution. Moreover, a probability calibration based on mc-dropout and isotonic regression is used to calibrate confidences, helping to select reliable low-confidence samples. Various ablation studies were performed to show the influence of autoencoder-based sampling, calibration, and transfer learning, which showed that the proposed method could achieve 92% accuracy using 399 training samples. In contrast, other methods required more training samples to reach the same level of accuracy without calibration or an autoencoder and clustering approach or training without active learning. The study also found that transfer learning significantly impacted faster convergence and that the proposed active learning approach was more effective than traditional methods.
publishDate 2023
dc.date.none.fl_str_mv 2023
2023-01-01
2024
2024-06-14
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
VoR
http://purl.org/coar/version/c_970fb48d4fbd8a85
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/2117/410000
https://dx.doi.org/10.1016/j.cmpbup.2023.100096
url https://hdl.handle.net/2117/410000
https://dx.doi.org/10.1016/j.cmpbup.2023.100096
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
Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0/
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
Attribution 4.0 International
http://creativecommons.org/licenses/by/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:UPCommons. Portal del coneixement obert de la UPC
instname:Universitat Politècnica de Catalunya (UPC)
instname_str Universitat Politècnica de Catalunya (UPC)
reponame_str UPCommons. Portal del coneixement obert de la UPC
collection UPCommons. Portal del coneixement obert de la UPC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869415476250017792
score 15,812455

Similares en LA Referencia