Disentangled representational learning for anomaly detection in single-lead electrocardiogram signals using variational autoencoder

Wearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges regarding statistical approaches and explainability. This work investigates the feasibility of medically explainable anomaly detection through disenta...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Computers in biology and medicine Ročník 184; s. 109422
Hlavní autoři: Kapsecker, Maximilian, Möller, Matthias C., Jonas, Stephan M.
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States Elsevier Ltd 01.01.2025
Elsevier Limited
Témata:
ISSN:0010-4825, 1879-0534, 1879-0534
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Wearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges regarding statistical approaches and explainability. This work investigates the feasibility of medically explainable anomaly detection through disentangled representational learning of ECGs and personalization to mitigate inter-subject variations. Five open-source ECG datasets were converted into a set of denoised one-second traces of lead I signal, each covering individual features such as wave morphologies and pathologies. A beta total correlation variational autoencoder was optimized on four of these datasets for 68 systematic parameterization variants. The best-performing model revealed disentanglement in the 12-dimensional embedding space, specifically between atrial- and ventricular features. Within the embedding space, a k-nearest neighbor classifier was evaluated on a left-out test set tailored for anomaly detection. The result is a F1 score of 0.94 for the binary prediction of sinus rhythm and the pathological classes: Left bundle branch block, right bundle branch block, myocardial infarction, and AV block (1st degree). The 90.94% accuracy in anomaly detection falls within the range of established detectors (89.38%–99.77%) but offers the advantage of being explainable and largely unsupervised. Model fine-tuning for each of 100 randomly sampled individuals of the Icentia11k dataset mitigated inter-subject variations. The associated F1 score for predicting normal, premature atrial contraction, and premature ventricular contraction from the embedding space was 0.93. The distribution plots of pathologies along the explainable axis were reasonably consistent with medical expertise. The results suggest the presented disentangled variational autoencoder as a robust method for explainable ECG representation. •β total correlation variational autoencoder encode atomic electrocardiogram features.•Anomalies arise as linear combinations of outliers along interpretable axis encodings.•Fine-tuning ECG models per subject personalizes and mitigates data heterogeneity.•Performance remains competitive, despite the model being unsupervised and explainable.
AbstractList Wearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges regarding statistical approaches and explainability. This work investigates the feasibility of medically explainable anomaly detection through disentangled representational learning of ECGs and personalization to mitigate inter-subject variations. Five open-source ECG datasets were converted into a set of denoised one-second traces of lead I signal, each covering individual features such as wave morphologies and pathologies. A beta total correlation variational autoencoder was optimized on four of these datasets for 68 systematic parameterization variants. The best-performing model revealed disentanglement in the 12-dimensional embedding space, specifically between atrial- and ventricular features. Within the embedding space, a k-nearest neighbor classifier was evaluated on a left-out test set tailored for anomaly detection. The result is a F1 score of 0.94 for the binary prediction of sinus rhythm and the pathological classes: Left bundle branch block, right bundle branch block, myocardial infarction, and AV block (1st degree). The 90.94% accuracy in anomaly detection falls within the range of established detectors (89.38%–99.77%) but offers the advantage of being explainable and largely unsupervised. Model fine-tuning for each of 100 randomly sampled individuals of the Icentia11k dataset mitigated inter-subject variations. The associated F1 score for predicting normal, premature atrial contraction, and premature ventricular contraction from the embedding space was 0.93. The distribution plots of pathologies along the explainable axis were reasonably consistent with medical expertise. The results suggest the presented disentangled variational autoencoder as a robust method for explainable ECG representation. •β total correlation variational autoencoder encode atomic electrocardiogram features.•Anomalies arise as linear combinations of outliers along interpretable axis encodings.•Fine-tuning ECG models per subject personalizes and mitigates data heterogeneity.•Performance remains competitive, despite the model being unsupervised and explainable.
Wearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges regarding statistical approaches and explainability. This work investigates the feasibility of medically explainable anomaly detection through disentangled representational learning of ECGs and personalization to mitigate inter-subject variations. Five open-source ECG datasets were converted into a set of denoised one-second traces of lead I signal, each covering individual features such as wave morphologies and pathologies. A beta total correlation variational autoencoder was optimized on four of these datasets for 68 systematic parameterization variants. The best-performing model revealed disentanglement in the 12-dimensional embedding space, specifically between atrial- and ventricular features. Within the embedding space, a k-nearest neighbor classifier was evaluated on a left-out test set tailored for anomaly detection. The result is a F1 score of 0.94 for the binary prediction of sinus rhythm and the pathological classes: Left bundle branch block, right bundle branch block, myocardial infarction, and AV block (1st degree). The 90.94% accuracy in anomaly detection falls within the range of established detectors (89.38%-99.77%) but offers the advantage of being explainable and largely unsupervised. Model fine-tuning for each of 100 randomly sampled individuals of the Icentia11k dataset mitigated inter-subject variations. The associated F1 score for predicting normal, premature atrial contraction, and premature ventricular contraction from the embedding space was 0.93. The distribution plots of pathologies along the explainable axis were reasonably consistent with medical expertise. The results suggest the presented disentangled variational autoencoder as a robust method for explainable ECG representation.
AbstractWearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges regarding statistical approaches and explainability. This work investigates the feasibility of medically explainable anomaly detection through disentangled representational learning of ECGs and personalization to mitigate inter-subject variations. Five open-source ECG datasets were converted into a set of denoised one-second traces of lead I signal, each covering individual features such as wave morphologies and pathologies. A beta total correlation variational autoencoder was optimized on four of these datasets for 68 systematic parameterization variants. The best-performing model revealed disentanglement in the 12-dimensional embedding space, specifically between atrial- and ventricular features. Within the embedding space, a k-nearest neighbor classifier was evaluated on a left-out test set tailored for anomaly detection. The result is a F1 score of 0.94 for the binary prediction of sinus rhythm and the pathological classes: Left bundle branch block, right bundle branch block, myocardial infarction, and AV block (1st degree). The 90.94% accuracy in anomaly detection falls within the range of established detectors (89.38%–99.77%) but offers the advantage of being explainable and largely unsupervised. Model fine-tuning for each of 100 randomly sampled individuals of the Icentia11k dataset mitigated inter-subject variations. The associated F1 score for predicting normal, premature atrial contraction, and premature ventricular contraction from the embedding space was 0.93. The distribution plots of pathologies along the explainable axis were reasonably consistent with medical expertise. The results suggest the presented disentangled variational autoencoder as a robust method for explainable ECG representation.
Wearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges regarding statistical approaches and explainability. This work investigates the feasibility of medically explainable anomaly detection through disentangled representational learning of ECGs and personalization to mitigate inter-subject variations. Five open-source ECG datasets were converted into a set of denoised one-second traces of lead I signal, each covering individual features such as wave morphologies and pathologies. A beta total correlation variational autoencoder was optimized on four of these datasets for 68 systematic parameterization variants. The best-performing model revealed disentanglement in the 12-dimensional embedding space, specifically between atrial- and ventricular features. Within the embedding space, a k-nearest neighbor classifier was evaluated on a left-out test set tailored for anomaly detection. The result is a F1 score of 0.94 for the binary prediction of sinus rhythm and the pathological classes: Left bundle branch block, right bundle branch block, myocardial infarction, and AV block (1st degree). The 90.94% accuracy in anomaly detection falls within the range of established detectors (89.38%-99.77%) but offers the advantage of being explainable and largely unsupervised. Model fine-tuning for each of 100 randomly sampled individuals of the Icentia11k dataset mitigated inter-subject variations. The associated F1 score for predicting normal, premature atrial contraction, and premature ventricular contraction from the embedding space was 0.93. The distribution plots of pathologies along the explainable axis were reasonably consistent with medical expertise. The results suggest the presented disentangled variational autoencoder as a robust method for explainable ECG representation.Wearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges regarding statistical approaches and explainability. This work investigates the feasibility of medically explainable anomaly detection through disentangled representational learning of ECGs and personalization to mitigate inter-subject variations. Five open-source ECG datasets were converted into a set of denoised one-second traces of lead I signal, each covering individual features such as wave morphologies and pathologies. A beta total correlation variational autoencoder was optimized on four of these datasets for 68 systematic parameterization variants. The best-performing model revealed disentanglement in the 12-dimensional embedding space, specifically between atrial- and ventricular features. Within the embedding space, a k-nearest neighbor classifier was evaluated on a left-out test set tailored for anomaly detection. The result is a F1 score of 0.94 for the binary prediction of sinus rhythm and the pathological classes: Left bundle branch block, right bundle branch block, myocardial infarction, and AV block (1st degree). The 90.94% accuracy in anomaly detection falls within the range of established detectors (89.38%-99.77%) but offers the advantage of being explainable and largely unsupervised. Model fine-tuning for each of 100 randomly sampled individuals of the Icentia11k dataset mitigated inter-subject variations. The associated F1 score for predicting normal, premature atrial contraction, and premature ventricular contraction from the embedding space was 0.93. The distribution plots of pathologies along the explainable axis were reasonably consistent with medical expertise. The results suggest the presented disentangled variational autoencoder as a robust method for explainable ECG representation.
ArticleNumber 109422
Author Jonas, Stephan M.
Möller, Matthias C.
Kapsecker, Maximilian
Author_xml – sequence: 1
  givenname: Maximilian
  orcidid: 0000-0002-3907-0749
  surname: Kapsecker
  fullname: Kapsecker, Maximilian
  email: max.kapsecker@tum.de
  organization: TUM School of Computation, Information and Technology, Technical University of Munich, Boltzmannstraße 3, Garching bei München, 85748, Bavaria, Germany
– sequence: 2
  givenname: Matthias C.
  surname: Möller
  fullname: Möller, Matthias C.
  organization: Department of Paediatric Cardiology and Paediatric Cardiac Surgery, University Hospital Bonn, Venusberg-Campus 1, Bonn, 53127, North Rhine-Westphalia, Germany
– sequence: 3
  givenname: Stephan M.
  orcidid: 0000-0002-3687-6165
  surname: Jonas
  fullname: Jonas, Stephan M.
  organization: Institute for Digital Medicine, University Hospital Bonn, Venusberg-Campus 1, Bonn, 53127, North Rhine-Westphalia, Germany
BackLink https://www.ncbi.nlm.nih.gov/pubmed/39581125$$D View this record in MEDLINE/PubMed
BookMark eNqNkkuLFDEUhYOMOD2jf0ECbtxUm1dVpTaDOj5hwIW6DunkVpM2lbRJ1UCv_eOmrGmFBqFXITffPUnOuVfoIsQACGFK1pTQ5tVubeKw37g4gF0zwkQpd4KxR2hFZdtVpObiAq0IoaQSktWX6CrnHSFEEE6eoEve1ZJSVq_Qr3cuQxh12HqwOME-wZ_96GLQHnvQKbiwxX1MWIc4aH_AFkYwM4BdwNnNrVUBLQZf6ikanayL26SHcrotMhlPM4bvdXJHZT2NEYKJFtJT9LgvEDx7WK_R9w_vv91-qu6-fPx8--auMqLtWGVI32kqKPRcNpIb2wqjhaUNM12nDTMWusa2hvANF8LWLdMWaq1123cWZMev0ctFd5_izwnyqAaXDXivA8QpK045a4iULS_oixN0F6c0f6VQohF1LVlTqOcP1LQpSah9coNOB3W0twA3C2BSzDlBr4xbvB2Tdl5RouY81U79y1PNeaolzyIgTwSOd5zR-nZphWLpvYOksnHFcbAulZiUje4ckZsTEeNdcEb7H3CA_NcUqjJTRH2dJ24eOCYIrUk7O_D6_wLnveE3XfzupQ
CitedBy_id crossref_primary_10_1038_s41598_025_06680_9
crossref_primary_10_3390_app15147975
crossref_primary_10_1371_journal_pdig_0000793
crossref_primary_10_1038_s41598_025_07781_1
crossref_primary_10_1007_s13534_025_00473_9
Cites_doi 10.1109/TBME.2021.3108164
10.1093/ehjdh/ztac038
10.1161/01.CIR.101.23.e215
10.1016/j.jelectrocard.2021.02.011
10.1093/ehjdh/ztad007
10.1186/s12872-022-02746-y
10.1038/s41598-024-51258-6
10.3389/fgene.2021.638191
10.1016/j.gheart.2019.07.004
10.1109/TBME.2003.808805
10.1007/s44196-023-00186-w
10.3758/s13428-020-01516-y
10.1038/s41597-020-0386-x
10.1161/01.CIR.92.10.2929
10.1186/s40537-020-00320-x
10.1016/j.bspc.2020.102054
10.1109/RBME.2022.3154893
10.1016/j.asoc.2023.110176
10.1038/s41597-023-02416-4
10.1109/51.932724
10.3390/s20051461
10.3390/electronics9010135
10.1145/3534678.3539140
10.1371/journal.pone.0260612
10.3390/app13084964
10.1038/s41598-019-56927-5
10.1038/s41597-020-0495-6
10.1007/s00521-022-07366-3
10.1007/BF02764938
ContentType Journal Article
Copyright 2024 The Authors
The Authors
Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.
2024. The Authors
Copyright_xml – notice: 2024 The Authors
– notice: The Authors
– notice: Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.
– notice: 2024. The Authors
DBID 6I.
AAFTH
AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
8FD
FR3
JQ2
K9.
M7Z
NAPCQ
P64
7X8
DOI 10.1016/j.compbiomed.2024.109422
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
Technology Research Database
Engineering Research Database
ProQuest Computer Science Collection
ProQuest Health & Medical Complete (Alumni)
Biochemistry Abstracts 1
Nursing & Allied Health Premium
Biotechnology and BioEngineering Abstracts
MEDLINE - Academic
DatabaseTitle CrossRef
MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
Nursing & Allied Health Premium
Technology Research Database
ProQuest Computer Science Collection
Biochemistry Abstracts 1
ProQuest Health & Medical Complete (Alumni)
Engineering Research Database
Biotechnology and BioEngineering Abstracts
MEDLINE - Academic
DatabaseTitleList
MEDLINE

Nursing & Allied Health Premium

MEDLINE - Academic
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Medicine
EISSN 1879-0534
EndPage 109422
ExternalDocumentID 39581125
10_1016_j_compbiomed_2024_109422
S0010482524015075
1_s2_0_S0010482524015075
Genre Journal Article
GroupedDBID ---
--K
--M
--Z
-~X
.1-
.55
.DC
.FO
.GJ
.~1
0R~
1B1
1P~
1RT
1~.
1~5
29F
4.4
457
4G.
53G
5GY
5VS
7-5
71M
77I
7RV
7X7
88E
8AO
8FE
8FG
8FH
8FI
8FJ
8G5
8P~
9JN
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AATTM
AAXKI
AAXUO
AAYFN
AAYWO
ABBOA
ABFNM
ABJNI
ABMAC
ABMZM
ABOCM
ABUWG
ABWVN
ABXDB
ACDAQ
ACGFS
ACIEU
ACIUM
ACIWK
ACLOT
ACNNM
ACPRK
ACRLP
ACRPL
ACVFH
ACZNC
ADBBV
ADCNI
ADEZE
ADJOM
ADMUD
ADNMO
AEBSH
AEIPS
AEKER
AENEX
AEUPX
AEVXI
AFJKZ
AFKRA
AFPUW
AFRAH
AFRHN
AFTJW
AFXIZ
AGHFR
AGQPQ
AGUBO
AGYEJ
AHHHB
AHMBA
AHZHX
AIALX
AIEXJ
AIGII
AIIUN
AIKHN
AITUG
AJRQY
AJUYK
AKBMS
AKRWK
AKYEP
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
ANZVX
AOUOD
APXCP
ARAPS
ASPBG
AVWKF
AXJTR
AZFZN
AZQEC
BBNVY
BENPR
BGLVJ
BHPHI
BKEYQ
BKOJK
BLXMC
BNPGV
BPHCQ
BVXVI
CCPQU
CS3
DU5
DWQXO
EBS
EFJIC
EFKBS
EFLBG
EJD
EMOBN
EO8
EO9
EP2
EP3
EX3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
FYUFA
G-2
G-Q
GBLVA
GBOLZ
GNUQQ
GUQSH
HCIFZ
HLZ
HMCUK
HMK
HMO
HVGLF
HZ~
IHE
J1W
K6V
K7-
KOM
LK8
LX9
M1P
M29
M2O
M41
M7P
MO0
N9A
NAPCQ
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
P62
PC.
PHGZM
PHGZT
PJZUB
PPXIY
PQGLB
PQQKQ
PROAC
PSQYO
Q38
R2-
ROL
RPZ
RXW
SAE
SBC
SCC
SDF
SDG
SDP
SEL
SES
SEW
SPC
SPCBC
SSH
SSV
SSZ
SV3
T5K
TAE
UAP
UKHRP
WOW
WUQ
X7M
XPP
Z5R
ZGI
~G-
~HD
AACTN
AFCTW
ALIPV
RIG
6I.
AAFTH
9DU
AAYXX
AFFHD
CITATION
3V.
AFKWA
AJOXV
AMFUW
CGR
CUY
CVF
ECM
EIF
M0N
NPM
8FD
FR3
JQ2
K9.
M7Z
P64
7X8
ID FETCH-LOGICAL-c4792-c0f9a141ef38683cd74ca4d162c99ac2cde96d7c03b344d572ade5aaa7f9de893
ISSN 0010-4825
1879-0534
IngestDate Wed Oct 01 09:58:09 EDT 2025
Tue Oct 07 06:18:25 EDT 2025
Wed Feb 19 01:58:15 EST 2025
Sat Nov 29 08:18:33 EST 2025
Tue Nov 18 22:23:51 EST 2025
Sat Mar 08 15:48:41 EST 2025
Fri Mar 14 01:54:07 EDT 2025
Tue Oct 14 19:38:01 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords Representational learning
Electrocardiography
Personalization
Explainable anomaly detection
Language English
License This is an open access article under the CC BY license.
Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c4792-c0f9a141ef38683cd74ca4d162c99ac2cde96d7c03b344d572ade5aaa7f9de893
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ORCID 0000-0002-3687-6165
0000-0002-3907-0749
OpenAccessLink https://dx.doi.org/10.1016/j.compbiomed.2024.109422
PMID 39581125
PQID 3146455826
PQPubID 1226355
PageCount 1
ParticipantIDs proquest_miscellaneous_3132608873
proquest_journals_3146455826
pubmed_primary_39581125
crossref_citationtrail_10_1016_j_compbiomed_2024_109422
crossref_primary_10_1016_j_compbiomed_2024_109422
elsevier_sciencedirect_doi_10_1016_j_compbiomed_2024_109422
elsevier_clinicalkeyesjournals_1_s2_0_S0010482524015075
elsevier_clinicalkey_doi_10_1016_j_compbiomed_2024_109422
PublicationCentury 2000
PublicationDate 2025-01-01
PublicationDateYYYYMMDD 2025-01-01
PublicationDate_xml – month: 01
  year: 2025
  text: 2025-01-01
  day: 01
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
– name: Oxford
PublicationTitle Computers in biology and medicine
PublicationTitleAlternate Comput Biol Med
PublicationYear 2025
Publisher Elsevier Ltd
Elsevier Limited
Publisher_xml – name: Elsevier Ltd
– name: Elsevier Limited
References Li, Boulanger (b56) 2020; 20
Gayathiri (b16) 2023
Jiang, Zheng, Lai (b14) 2023
Soydaner (b58) 2022; 34
Gyawali, Li, Knight, Ghimire, Horacek, Sapp, Wang (b20) 2019
Tan, Androz, Ortiz-Gagné, Chamseddine, Fecteau, Courville, Bengio, Cohen (b32) 2022
Johnson, Lindenstrauss, Schechtman (b42) 1986; 54
Y. Li, Z. Chen, D. Zha, M. Du, J. Ni, D. Zhang, H. Chen, X. Hu, Towards Learning Disentangled Representations for Time Series, in: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2022, pp. 3270–3278.
Li, Pei, Li (b6) 2023; 138
Esmaeili, Wu, Jain, Bozkurt, Siddharth, Paige, Brooks, Dy, Meent (b46) 2019
Ebrahimi, Loni, Daneshtalab, Gharehbaghi (b47) 2020; 7
Van Steenkiste, Deschrijver, Dhaene (b22) 2019
van de Leur, Bos, Taha, Sammani, Yeung, van Duijvenboden, Lambiase, Hassink, van der Harst, Doevendans (b23) 2022; 3
He, Spokoyny, Neubig, Berg-Kirkpatrick (b52) 2019
Kuznetsov, Moskalenko, Gribanov, Zolotykh (b8) 2021; 12
Locatello, Bauer, Lucic, Raetsch, Gelly, Schölkopf, Bachem (b57) 2019
Wagner, Strodthoff, Bousseljot, Kreiseler, Lunze, Samek, Schaeffter (b35) 2020; 7
M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, M. Devin, S. Ghemawat, G. Irving, M. Isard, et al., TensorFlow: A System for Large-Scale Machine Learning, in: 12th USENIX Symposium on Operating Systems Design and Implementation, OSDI 16, 2016, pp. 265–283.
Thudumu, Branch, Jin, Singh (b3) 2020; 7
van der Valk, Atsma, Scherptong, Staring (b26) 2023
Zhao, Zhang (b48) 2024; 14
Lucas, Tucker, Grosse, Norouzi (b51) 2019
Witvliet, Karregat, Himmelreich, de Jong, Lucassen, Harskamp (b54) 2021; 66
.
Jolliffe, Cadima (b43) 2016; 374
Tan, Androz, Ortiz-Gagné, Chamseddine, Fecteau, Courville, Bengio, Cohen (b38) 2021
Porumb, Stranges, Pescapè, Pecchia (b12) 2020; 10
Ferreira, Kraus, Mitchell, Perel, Piñeiro, Chioncel, Colque, De Boer, Gomez-Mesa, Grancelli (b2) 2019; 14
Gyawali, Murkute, Toloubidokhti, Jiang, Horacek, Sapp, Wang (b19) 2021; 69
Jang, Kim, Lim, Yoon (b9) 2021; 16
Rubel, Pani, Schloegl, Fayn, Badilini, Macfarlane, Varri (b25) 2016
Moss, Zareba, Benhorin, Locati, Hall, Robinson, Schwartz, Towbin, Vincent, Lehmann (b18) 1995; 92
Maas, Hannun, Ng (b50) 2013
McSharry, Clifford, Tarassenko, Smith (b34) 2003; 50
Goldberger, Amaral, Glass, Hausdorff, Ivanov, Mark, Mietus, Moody, Peng, Stanley (b36) 2000; 101
Nezamabadi, Sardaripour, Haghi, Forouzanfar (b7) 2022; 16
Todo, Selmani, Laurent, Loubes (b27) 2023
I. Higgins, L. Matthey, A. Pal, C. Burgess, X. Glorot, M. Botvinick, S. Mohamed, A. Lerchner
Nithya, Rani (b10) 2022
TensorFlow Datasets, A collection of ready-to-use datasets
VAE: Learning Basic Visual Concepts with a Constrained Variational Framework, in: International Conference on Learning Representations, 2016.
Porumb, Griffen, Hattersley, Pecchia (b13) 2020; 62
J.-Y. Franceschi, A. Dieuleveut, M. Jaggi, Unsupervised Scalable Representation Learning for Multivariate Time Series, in: Advances in Neural Information Processing Systems, 32, 2019.
Wagner, Strodthoff, Bousseljot, Samek, Schaeffter (b30) 2022
World Heart Federation (b1) 2023
Kingma, Welling (b4) 2013
Saki, Karandish, Cheraghian, Heybar, Hashemi, Azhdari (b37) 2022; 22
Xiao, Lee, Mokhtar, Ismail, Pauzi, Zhang, Lim (b28) 2023; 13
Orini, van Duijvenboden, Young, Ramírez, Jones, Tinker, Munroe, Lambiase (b55) 2023; 4
P.K. Gyawali, B.M. Horacek, J.L. Sapp, L. Wang, Learning Disentangled Representation from 12-lead Electrograms: Application in Localizing the Origin of Ventricular Tachycardia, in: Workshops At the Thirty-Second AAAI Conference on Artificial Intelligence, 2018.
Zheng, Zhang, Danioko, Yao, Guo, Rakovski (b31) 2020; 7
Bank, Koenigstein, Giryes (b5) 2023
Gillette, Gsell, Nagel, Bender, Winkler, Williams, Bär, Schäffter, Dössel, Plank (b33) 2023; 10
Yang, Ngoc, Shin, Hwang (b49) 2023; 16
Nurmaini, Darmawahyuni, Sakti Mukti, Rachmatullah, Firdaus, Tutuko (b15) 2020; 9
Moody, Mark (b11) 2001; 20
Van der Maaten, Hinton (b44) 2008; 9
R.T. Chen, X. Li, R.B. Grosse, D.K. Duvenaud, Isolating Sources of Disentanglement in Variational Autoencoders, in: Advances in Neural Information Processing Systems, 31, 2018.
Hinton (b53) 2012
Makowski, Pham, Lau, Brammer, Lespinasse, Pham, Schölzel, Chen (b39) 2021; 53
Tan (10.1016/j.compbiomed.2024.109422_b38) 2021
Nithya (10.1016/j.compbiomed.2024.109422_b10) 2022
Van Steenkiste (10.1016/j.compbiomed.2024.109422_b22) 2019
Esmaeili (10.1016/j.compbiomed.2024.109422_b46) 2019
van der Valk (10.1016/j.compbiomed.2024.109422_b26) 2023
World Heart Federation (10.1016/j.compbiomed.2024.109422_b1) 2023
van de Leur (10.1016/j.compbiomed.2024.109422_b23) 2022; 3
Li (10.1016/j.compbiomed.2024.109422_b6) 2023; 138
Jiang (10.1016/j.compbiomed.2024.109422_b14) 2023
Jang (10.1016/j.compbiomed.2024.109422_b9) 2021; 16
Jolliffe (10.1016/j.compbiomed.2024.109422_b43) 2016; 374
Nezamabadi (10.1016/j.compbiomed.2024.109422_b7) 2022; 16
10.1016/j.compbiomed.2024.109422_b17
Zhao (10.1016/j.compbiomed.2024.109422_b48) 2024; 14
Gayathiri (10.1016/j.compbiomed.2024.109422_b16) 2023
Tan (10.1016/j.compbiomed.2024.109422_b32) 2022
Porumb (10.1016/j.compbiomed.2024.109422_b13) 2020; 62
Gyawali (10.1016/j.compbiomed.2024.109422_b20) 2019
Nurmaini (10.1016/j.compbiomed.2024.109422_b15) 2020; 9
10.1016/j.compbiomed.2024.109422_b40
10.1016/j.compbiomed.2024.109422_b41
Xiao (10.1016/j.compbiomed.2024.109422_b28) 2023; 13
Gyawali (10.1016/j.compbiomed.2024.109422_b19) 2021; 69
Yang (10.1016/j.compbiomed.2024.109422_b49) 2023; 16
Wagner (10.1016/j.compbiomed.2024.109422_b30) 2022
Johnson (10.1016/j.compbiomed.2024.109422_b42) 1986; 54
Gillette (10.1016/j.compbiomed.2024.109422_b33) 2023; 10
Lucas (10.1016/j.compbiomed.2024.109422_b51) 2019
Ebrahimi (10.1016/j.compbiomed.2024.109422_b47) 2020; 7
10.1016/j.compbiomed.2024.109422_b45
Goldberger (10.1016/j.compbiomed.2024.109422_b36) 2000; 101
Li (10.1016/j.compbiomed.2024.109422_b56) 2020; 20
Wagner (10.1016/j.compbiomed.2024.109422_b35) 2020; 7
He (10.1016/j.compbiomed.2024.109422_b52) 2019
Orini (10.1016/j.compbiomed.2024.109422_b55) 2023; 4
McSharry (10.1016/j.compbiomed.2024.109422_b34) 2003; 50
Moss (10.1016/j.compbiomed.2024.109422_b18) 1995; 92
Witvliet (10.1016/j.compbiomed.2024.109422_b54) 2021; 66
Rubel (10.1016/j.compbiomed.2024.109422_b25) 2016
Porumb (10.1016/j.compbiomed.2024.109422_b12) 2020; 10
10.1016/j.compbiomed.2024.109422_b21
Saki (10.1016/j.compbiomed.2024.109422_b37) 2022; 22
Todo (10.1016/j.compbiomed.2024.109422_b27) 2023
Locatello (10.1016/j.compbiomed.2024.109422_b57) 2019
Thudumu (10.1016/j.compbiomed.2024.109422_b3) 2020; 7
Van der Maaten (10.1016/j.compbiomed.2024.109422_b44) 2008; 9
Moody (10.1016/j.compbiomed.2024.109422_b11) 2001; 20
Hinton (10.1016/j.compbiomed.2024.109422_b53) 2012
Kingma (10.1016/j.compbiomed.2024.109422_b4) 2013
Kuznetsov (10.1016/j.compbiomed.2024.109422_b8) 2021; 12
10.1016/j.compbiomed.2024.109422_b24
Zheng (10.1016/j.compbiomed.2024.109422_b31) 2020; 7
Soydaner (10.1016/j.compbiomed.2024.109422_b58) 2022; 34
Makowski (10.1016/j.compbiomed.2024.109422_b39) 2021; 53
Bank (10.1016/j.compbiomed.2024.109422_b5) 2023
10.1016/j.compbiomed.2024.109422_b29
Maas (10.1016/j.compbiomed.2024.109422_b50) 2013
Ferreira (10.1016/j.compbiomed.2024.109422_b2) 2019; 14
References_xml – volume: 53
  start-page: 1689
  year: 2021
  end-page: 1696
  ident: b39
  article-title: NeuroKit2: A Python toolbox for neurophysiological signal processing
  publication-title: Behav. Res. Methods
– volume: 101
  start-page: e215
  year: 2000
  end-page: e220
  ident: b36
  article-title: PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals
  publication-title: Circulation
– volume: 92
  start-page: 2929
  year: 1995
  end-page: 2934
  ident: b18
  article-title: ECG T-wave Patterns in Genetically Distinct Forms of the Hereditary Long QT Syndrome
  publication-title: Circulation
– volume: 50
  start-page: 289
  year: 2003
  end-page: 294
  ident: b34
  article-title: A dynamical model for generating synthetic electrocardiogram signals
  publication-title: IEEE Trans. Biomed. Eng.
– start-page: 353
  year: 2023
  end-page: 374
  ident: b5
  article-title: Autoencoders
  publication-title: Machine Learning for Data Science Handbook: Data Mining and Knowledge Discovery Handbook
– year: 2022
  ident: b30
  article-title: PTB-XL, a large publicly available electrocardiography dataset (version 1.0.3)
– volume: 374
  year: 2016
  ident: b43
  article-title: Principal component analysis: a review and recent developments
  publication-title: Philos. Trans. Royal Soc.a: Math. Phys. Eng. Sci.
– volume: 4
  start-page: 112
  year: 2023
  end-page: 118
  ident: b55
  article-title: Premature atrial and ventricular contractions detected on wearable-format electrocardiograms and prediction of cardiovascular events
  publication-title: Eur. Heart J.-Dig. Health
– year: 2019
  ident: b51
  article-title: Understanding Posterior Collapse in Generative Latent Variable Models
– start-page: 1
  year: 2023
  end-page: 5
  ident: b27
  article-title: Counterfactual Explanation for Multivariate Times Series Using A Contrastive Variational Autoencoder
  publication-title: ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing
– year: 2012
  ident: b53
  article-title: Mini-batch Gradient Descent
– volume: 138
  year: 2023
  ident: b6
  article-title: A comprehensive survey on design and application of autoencoder in deep learning
  publication-title: Appl. Soft Comput.
– volume: 7
  start-page: 1
  year: 2020
  end-page: 15
  ident: b35
  article-title: PTB-XL, a large publicly available electrocardiography dataset
  publication-title: Sci. Data
– volume: 20
  start-page: 45
  year: 2001
  end-page: 50
  ident: b11
  article-title: The impact of the MIT-BIH Arrhythmia Database
  publication-title: IEEE Eng. Med. Biol. Mag.
– volume: 22
  start-page: 1
  year: 2022
  end-page: 10
  ident: b37
  article-title: Prevalence of cardiovascular diseases and associated factors among adults from southwest Iran: Baseline data from hoveyzeh cohort study
  publication-title: BMC Cardiovasc. Disorders
– volume: 12
  year: 2021
  ident: b8
  article-title: Interpretable Feature Generation in ECG Using a Variational Autoencoder
  publication-title: Front. Genet.
– reference: J.-Y. Franceschi, A. Dieuleveut, M. Jaggi, Unsupervised Scalable Representation Learning for Multivariate Time Series, in: Advances in Neural Information Processing Systems, 32, 2019.
– volume: 20
  start-page: 1461
  year: 2020
  ident: b56
  article-title: A Survey of Heart Anomaly Detection Using Ambulatory Electrocardiogram (ECG)
  publication-title: Sensors
– volume: 7
  start-page: 1
  year: 2020
  end-page: 30
  ident: b3
  article-title: A comprehensive survey of anomaly detection techniques for high dimensional big data
  publication-title: J. Big Data
– start-page: 1
  year: 2023
  end-page: 4
  ident: b14
  article-title: A Semi-supervised Algorithm for Atrial Fibrillation Attack Prediction Using Convolution Auto-encoder of Time Series Signal
  publication-title: 2023 45th Annual International Conference of the IEEE Engineering in Medicine & Biology Society
– start-page: 222
  year: 2022
  end-page: 226
  ident: b10
  article-title: Stacked Variational Autoencoder in the Classification of Cardiac Arrhythmia using ECG Signals with 2D-ECG Images
  publication-title: 2022 International Conference on Intelligent Innovations in Engineering and Technology
– reference: TensorFlow Datasets, A collection of ready-to-use datasets,
– volume: 7
  year: 2020
  ident: b47
  article-title: A review on deep learning methods for ECG arrhythmia classification
  publication-title: Expert Syst. Appl. X
– start-page: 2525
  year: 2019
  end-page: 2534
  ident: b46
  article-title: Structured Disentangled Representations
  publication-title: The 22nd International Conference on Artificial Intelligence and Statistics
– volume: 16
  start-page: 11
  year: 2023
  ident: b49
  article-title: DPReLU: Dynamic Parametric Rectified Linear Unit and Its Proper Weight Initialization Method
  publication-title: Int. J. Comput. Intell. Syst.
– volume: 10
  start-page: 531
  year: 2023
  ident: b33
  article-title: MedalCare-XL: 16,900 healthy and pathological synthetic 12 lead ECGs from electrophysiological simulations
  publication-title: Sci. Data
– year: 2023
  ident: b1
  article-title: World Heart Report 2023
– start-page: 263
  year: 2023
  end-page: 268
  ident: b16
  article-title: Unsupervised Anomaly Detection to handle Imbalanced Datasets using Auto encoders for ECG signal Classification
  publication-title: 2023 Third International Conference on Secure Cyber Computing and Communication
– reference: R.T. Chen, X. Li, R.B. Grosse, D.K. Duvenaud, Isolating Sources of Disentanglement in Variational Autoencoders, in: Advances in Neural Information Processing Systems, 31, 2018.
– year: 2022
  ident: b32
  article-title: Icentia11k Single Lead Continuous Raw Electrocardiogram Dataset (version 1.0)
– start-page: 3
  year: 2013
  ident: b50
  article-title: Rectifier Nonlinearities Improve Neural Network Acoustic Models
  publication-title: Proc. Icml, Vol. 30, No. 1
– volume: 14
  start-page: 1589
  year: 2024
  ident: b48
  article-title: A improved pooling method for convolutional neural networks
  publication-title: Sci. Rep.
– year: 2013
  ident: b4
  article-title: Auto-Encoding Variational Bayes
– year: 2023
  ident: b26
  article-title: Joint Optimization of a
– volume: 14
  start-page: 197
  year: 2019
  end-page: 214
  ident: b2
  article-title: World Heart Federation Roadmap for Heart Failure
  publication-title: Global Heart
– volume: 62
  year: 2020
  ident: b13
  article-title: Nocturnal low glucose detection in healthy elderly from one-lead ECG using convolutional denoising autoencoders
  publication-title: Biomed. Signal Process. Control
– reference: M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, M. Devin, S. Ghemawat, G. Irving, M. Isard, et al., TensorFlow: A System for Large-Scale Machine Learning, in: 12th USENIX Symposium on Operating Systems Design and Implementation, OSDI 16, 2016, pp. 265–283.
– volume: 66
  start-page: 33
  year: 2021
  end-page: 37
  ident: b54
  article-title: Usefulness, pitfalls and interpretation of handheld single-lead electrocardiograms
  publication-title: J. Electrocardiol.
– volume: 3
  start-page: 390
  year: 2022
  end-page: 404
  ident: b23
  article-title: Improving explainability of deep neural network-based electrocardiogram interpretation using variational auto-encoders
  publication-title: Eur. Heart J.-Dig. Health
– volume: 9
  start-page: 135
  year: 2020
  ident: b15
  article-title: Deep Learning-Based Stacked Denoising and Autoencoder for ECG Heartbeat Classification
  publication-title: Electronics
– volume: 69
  start-page: 860
  year: 2021
  end-page: 870
  ident: b19
  article-title: Learning to Disentangle Inter-Subject Anatomical Variations in Electrocardiographic Data
  publication-title: IEEE Trans. Biomed. Eng.
– reference: I. Higgins, L. Matthey, A. Pal, C. Burgess, X. Glorot, M. Botvinick, S. Mohamed, A. Lerchner,
– volume: 9
  year: 2008
  ident: b44
  article-title: Visualizing Data using t-SNE
  publication-title: J. Mach. Learn. Res.
– start-page: 373
  year: 2019
  end-page: 378
  ident: b22
  article-title: Interpretable ECG Beat Embedding using Disentangled Variational Auto-Encoders
  publication-title: 2019 IEEE 32nd International Symposium on Computer-Based Medical Systems (CBMS)
– volume: 54
  start-page: 129
  year: 1986
  end-page: 138
  ident: b42
  article-title: Extensions of lipschitz maps into Banach spaces
  publication-title: Israel J. Math.
– reference: -VAE: Learning Basic Visual Concepts with a Constrained Variational Framework, in: International Conference on Learning Representations, 2016.
– volume: 34
  start-page: 13371
  year: 2022
  end-page: 13385
  ident: b58
  article-title: Attention mechanism in neural networks: where it comes and where it goes
  publication-title: Neural Comput. Appl.
– volume: 7
  start-page: 1
  year: 2020
  end-page: 8
  ident: b31
  article-title: A 12-lead electrocardiogram database for arrhythmia research covering more than 10,000 patients
  publication-title: Sci. Data
– reference: .
– start-page: 309
  year: 2016
  end-page: 312
  ident: b25
  article-title: SCP-ECG V3.0: An enhanced standard communication protocol for computer-assisted electrocardiography
  publication-title: 2016 Computing in Cardiology Conference
– reference: Y. Li, Z. Chen, D. Zha, M. Du, J. Ni, D. Zhang, H. Chen, X. Hu, Towards Learning Disentangled Representations for Time Series, in: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2022, pp. 3270–3278.
– start-page: 4114
  year: 2019
  end-page: 4124
  ident: b57
  article-title: Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
  publication-title: International Conference on Machine Learning
– volume: 13
  start-page: 4964
  year: 2023
  ident: b28
  article-title: Deep Learning-Based ECG Arrhythmia Classification: A Systematic Review
  publication-title: Appl. Sci.
– volume: 16
  year: 2021
  ident: b9
  article-title: Unsupervised feature learning for electrocardiogram data using the convolutional variational autoencoder
  publication-title: PLoS One
– start-page: 1078
  year: 2019
  end-page: 1083
  ident: b20
  article-title: Improving Disentangled Representation Learning with the Beta Bernoulli Process
  publication-title: 2019 IEEE International Conference on Data Mining (ICDM)
– year: 2021
  ident: b38
  article-title: Icentia11k: An Unsupervised Representation Learning Dataset for Arrhythmia Subtype Discovery
  publication-title: Computing in Cardiology Conference (CinC)
– reference: P.K. Gyawali, B.M. Horacek, J.L. Sapp, L. Wang, Learning Disentangled Representation from 12-lead Electrograms: Application in Localizing the Origin of Ventricular Tachycardia, in: Workshops At the Thirty-Second AAAI Conference on Artificial Intelligence, 2018.
– volume: 16
  start-page: 208
  year: 2022
  end-page: 224
  ident: b7
  article-title: Unsupervised ECG Analysis: A Review
  publication-title: IEEE Rev. Biomed. Eng.
– volume: 10
  start-page: 170
  year: 2020
  ident: b12
  article-title: Precision Medicine and Artificial Intelligence: A Pilot Study on Deep Learning for Hypoglycemic Events Detection based on ECG
  publication-title: Sci. Rep.
– year: 2019
  ident: b52
  article-title: Lagging Inference Networks and Posterior Collapse in Variational Autoencoders
– year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b26
– volume: 69
  start-page: 860
  issue: 2
  year: 2021
  ident: 10.1016/j.compbiomed.2024.109422_b19
  article-title: Learning to Disentangle Inter-Subject Anatomical Variations in Electrocardiographic Data
  publication-title: IEEE Trans. Biomed. Eng.
  doi: 10.1109/TBME.2021.3108164
– start-page: 3
  year: 2013
  ident: 10.1016/j.compbiomed.2024.109422_b50
  article-title: Rectifier Nonlinearities Improve Neural Network Acoustic Models
– volume: 3
  start-page: 390
  issue: 3
  year: 2022
  ident: 10.1016/j.compbiomed.2024.109422_b23
  article-title: Improving explainability of deep neural network-based electrocardiogram interpretation using variational auto-encoders
  publication-title: Eur. Heart J.-Dig. Health
  doi: 10.1093/ehjdh/ztac038
– ident: 10.1016/j.compbiomed.2024.109422_b40
– volume: 101
  start-page: e215
  issue: 23
  year: 2000
  ident: 10.1016/j.compbiomed.2024.109422_b36
  article-title: PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals
  publication-title: Circulation
  doi: 10.1161/01.CIR.101.23.e215
– start-page: 1
  year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b27
  article-title: Counterfactual Explanation for Multivariate Times Series Using A Contrastive Variational Autoencoder
– volume: 7
  year: 2020
  ident: 10.1016/j.compbiomed.2024.109422_b47
  article-title: A review on deep learning methods for ECG arrhythmia classification
  publication-title: Expert Syst. Appl. X
– volume: 66
  start-page: 33
  year: 2021
  ident: 10.1016/j.compbiomed.2024.109422_b54
  article-title: Usefulness, pitfalls and interpretation of handheld single-lead electrocardiograms
  publication-title: J. Electrocardiol.
  doi: 10.1016/j.jelectrocard.2021.02.011
– volume: 4
  start-page: 112
  issue: 2
  year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b55
  article-title: Premature atrial and ventricular contractions detected on wearable-format electrocardiograms and prediction of cardiovascular events
  publication-title: Eur. Heart J.-Dig. Health
  doi: 10.1093/ehjdh/ztad007
– volume: 22
  start-page: 1
  issue: 1
  year: 2022
  ident: 10.1016/j.compbiomed.2024.109422_b37
  article-title: Prevalence of cardiovascular diseases and associated factors among adults from southwest Iran: Baseline data from hoveyzeh cohort study
  publication-title: BMC Cardiovasc. Disorders
  doi: 10.1186/s12872-022-02746-y
– volume: 14
  start-page: 1589
  issue: 1
  year: 2024
  ident: 10.1016/j.compbiomed.2024.109422_b48
  article-title: A improved pooling method for convolutional neural networks
  publication-title: Sci. Rep.
  doi: 10.1038/s41598-024-51258-6
– ident: 10.1016/j.compbiomed.2024.109422_b24
– volume: 12
  year: 2021
  ident: 10.1016/j.compbiomed.2024.109422_b8
  article-title: Interpretable Feature Generation in ECG Using a Variational Autoencoder
  publication-title: Front. Genet.
  doi: 10.3389/fgene.2021.638191
– volume: 14
  start-page: 197
  issue: 3
  year: 2019
  ident: 10.1016/j.compbiomed.2024.109422_b2
  article-title: World Heart Federation Roadmap for Heart Failure
  publication-title: Global Heart
  doi: 10.1016/j.gheart.2019.07.004
– volume: 50
  start-page: 289
  issue: 3
  year: 2003
  ident: 10.1016/j.compbiomed.2024.109422_b34
  article-title: A dynamical model for generating synthetic electrocardiogram signals
  publication-title: IEEE Trans. Biomed. Eng.
  doi: 10.1109/TBME.2003.808805
– year: 2021
  ident: 10.1016/j.compbiomed.2024.109422_b38
  article-title: Icentia11k: An Unsupervised Representation Learning Dataset for Arrhythmia Subtype Discovery
– volume: 16
  start-page: 11
  issue: 1
  year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b49
  article-title: DPReLU: Dynamic Parametric Rectified Linear Unit and Its Proper Weight Initialization Method
  publication-title: Int. J. Comput. Intell. Syst.
  doi: 10.1007/s44196-023-00186-w
– volume: 9
  issue: 11
  year: 2008
  ident: 10.1016/j.compbiomed.2024.109422_b44
  article-title: Visualizing Data using t-SNE
  publication-title: J. Mach. Learn. Res.
– volume: 53
  start-page: 1689
  issue: 4
  year: 2021
  ident: 10.1016/j.compbiomed.2024.109422_b39
  article-title: NeuroKit2: A Python toolbox for neurophysiological signal processing
  publication-title: Behav. Res. Methods
  doi: 10.3758/s13428-020-01516-y
– volume: 7
  start-page: 1
  issue: 1
  year: 2020
  ident: 10.1016/j.compbiomed.2024.109422_b31
  article-title: A 12-lead electrocardiogram database for arrhythmia research covering more than 10,000 patients
  publication-title: Sci. Data
  doi: 10.1038/s41597-020-0386-x
– start-page: 1078
  year: 2019
  ident: 10.1016/j.compbiomed.2024.109422_b20
  article-title: Improving Disentangled Representation Learning with the Beta Bernoulli Process
– ident: 10.1016/j.compbiomed.2024.109422_b41
– start-page: 353
  year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b5
  article-title: Autoencoders
– start-page: 1
  year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b14
  article-title: A Semi-supervised Algorithm for Atrial Fibrillation Attack Prediction Using Convolution Auto-encoder of Time Series Signal
– volume: 92
  start-page: 2929
  issue: 10
  year: 1995
  ident: 10.1016/j.compbiomed.2024.109422_b18
  article-title: ECG T-wave Patterns in Genetically Distinct Forms of the Hereditary Long QT Syndrome
  publication-title: Circulation
  doi: 10.1161/01.CIR.92.10.2929
– volume: 7
  start-page: 1
  year: 2020
  ident: 10.1016/j.compbiomed.2024.109422_b3
  article-title: A comprehensive survey of anomaly detection techniques for high dimensional big data
  publication-title: J. Big Data
  doi: 10.1186/s40537-020-00320-x
– volume: 62
  year: 2020
  ident: 10.1016/j.compbiomed.2024.109422_b13
  article-title: Nocturnal low glucose detection in healthy elderly from one-lead ECG using convolutional denoising autoencoders
  publication-title: Biomed. Signal Process. Control
  doi: 10.1016/j.bspc.2020.102054
– volume: 16
  start-page: 208
  year: 2022
  ident: 10.1016/j.compbiomed.2024.109422_b7
  article-title: Unsupervised ECG Analysis: A Review
  publication-title: IEEE Rev. Biomed. Eng.
  doi: 10.1109/RBME.2022.3154893
– volume: 138
  year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b6
  article-title: A comprehensive survey on design and application of autoencoder in deep learning
  publication-title: Appl. Soft Comput.
  doi: 10.1016/j.asoc.2023.110176
– volume: 10
  start-page: 531
  issue: 1
  year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b33
  article-title: MedalCare-XL: 16,900 healthy and pathological synthetic 12 lead ECGs from electrophysiological simulations
  publication-title: Sci. Data
  doi: 10.1038/s41597-023-02416-4
– year: 2019
  ident: 10.1016/j.compbiomed.2024.109422_b52
– volume: 20
  start-page: 45
  issue: 3
  year: 2001
  ident: 10.1016/j.compbiomed.2024.109422_b11
  article-title: The impact of the MIT-BIH Arrhythmia Database
  publication-title: IEEE Eng. Med. Biol. Mag.
  doi: 10.1109/51.932724
– start-page: 373
  year: 2019
  ident: 10.1016/j.compbiomed.2024.109422_b22
  article-title: Interpretable ECG Beat Embedding using Disentangled Variational Auto-Encoders
– year: 2012
  ident: 10.1016/j.compbiomed.2024.109422_b53
– year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b1
– start-page: 222
  year: 2022
  ident: 10.1016/j.compbiomed.2024.109422_b10
  article-title: Stacked Variational Autoencoder in the Classification of Cardiac Arrhythmia using ECG Signals with 2D-ECG Images
– ident: 10.1016/j.compbiomed.2024.109422_b45
– year: 2022
  ident: 10.1016/j.compbiomed.2024.109422_b32
– year: 2019
  ident: 10.1016/j.compbiomed.2024.109422_b51
– volume: 374
  issue: 2065
  year: 2016
  ident: 10.1016/j.compbiomed.2024.109422_b43
  article-title: Principal component analysis: a review and recent developments
  publication-title: Philos. Trans. Royal Soc.a: Math. Phys. Eng. Sci.
– year: 2022
  ident: 10.1016/j.compbiomed.2024.109422_b30
– volume: 20
  start-page: 1461
  issue: 5
  year: 2020
  ident: 10.1016/j.compbiomed.2024.109422_b56
  article-title: A Survey of Heart Anomaly Detection Using Ambulatory Electrocardiogram (ECG)
  publication-title: Sensors
  doi: 10.3390/s20051461
– year: 2013
  ident: 10.1016/j.compbiomed.2024.109422_b4
– volume: 9
  start-page: 135
  issue: 1
  year: 2020
  ident: 10.1016/j.compbiomed.2024.109422_b15
  article-title: Deep Learning-Based Stacked Denoising and Autoencoder for ECG Heartbeat Classification
  publication-title: Electronics
  doi: 10.3390/electronics9010135
– ident: 10.1016/j.compbiomed.2024.109422_b29
– ident: 10.1016/j.compbiomed.2024.109422_b17
  doi: 10.1145/3534678.3539140
– volume: 16
  issue: 12
  year: 2021
  ident: 10.1016/j.compbiomed.2024.109422_b9
  article-title: Unsupervised feature learning for electrocardiogram data using the convolutional variational autoencoder
  publication-title: PLoS One
  doi: 10.1371/journal.pone.0260612
– ident: 10.1016/j.compbiomed.2024.109422_b21
– start-page: 2525
  year: 2019
  ident: 10.1016/j.compbiomed.2024.109422_b46
  article-title: Structured Disentangled Representations
– volume: 13
  start-page: 4964
  issue: 8
  year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b28
  article-title: Deep Learning-Based ECG Arrhythmia Classification: A Systematic Review
  publication-title: Appl. Sci.
  doi: 10.3390/app13084964
– start-page: 309
  year: 2016
  ident: 10.1016/j.compbiomed.2024.109422_b25
  article-title: SCP-ECG V3.0: An enhanced standard communication protocol for computer-assisted electrocardiography
– volume: 10
  start-page: 170
  issue: 1
  year: 2020
  ident: 10.1016/j.compbiomed.2024.109422_b12
  article-title: Precision Medicine and Artificial Intelligence: A Pilot Study on Deep Learning for Hypoglycemic Events Detection based on ECG
  publication-title: Sci. Rep.
  doi: 10.1038/s41598-019-56927-5
– volume: 7
  start-page: 1
  issue: 1
  year: 2020
  ident: 10.1016/j.compbiomed.2024.109422_b35
  article-title: PTB-XL, a large publicly available electrocardiography dataset
  publication-title: Sci. Data
  doi: 10.1038/s41597-020-0495-6
– start-page: 4114
  year: 2019
  ident: 10.1016/j.compbiomed.2024.109422_b57
  article-title: Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
– volume: 34
  start-page: 13371
  issue: 16
  year: 2022
  ident: 10.1016/j.compbiomed.2024.109422_b58
  article-title: Attention mechanism in neural networks: where it comes and where it goes
  publication-title: Neural Comput. Appl.
  doi: 10.1007/s00521-022-07366-3
– start-page: 263
  year: 2023
  ident: 10.1016/j.compbiomed.2024.109422_b16
  article-title: Unsupervised Anomaly Detection to handle Imbalanced Datasets using Auto encoders for ECG signal Classification
– volume: 54
  start-page: 129
  issue: 2
  year: 1986
  ident: 10.1016/j.compbiomed.2024.109422_b42
  article-title: Extensions of lipschitz maps into Banach spaces
  publication-title: Israel J. Math.
  doi: 10.1007/BF02764938
SSID ssj0004030
Score 2.4445198
Snippet Wearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges regarding...
AbstractWearable technology enables the unsupervised recording of electrocardiogram (ECG) signals. Analyzing these high-dimensional ECG data poses challenges...
SourceID proquest
pubmed
crossref
elsevier
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 109422
SubjectTerms Anomalies
Datasets
Dimensional analysis
EKG
Electrocardiography
Electrocardiography - methods
Embedding
Explainable anomaly detection
Heart
Humans
Internal Medicine
Learning
Myocardial infarction
Other
Parameterization
Personalization
Representational learning
Signal Processing, Computer-Assisted
Ventricle
Wearable technology
Title Disentangled representational learning for anomaly detection in single-lead electrocardiogram signals using variational autoencoder
URI https://www.clinicalkey.com/#!/content/1-s2.0-S0010482524015075
https://www.clinicalkey.es/playcontent/1-s2.0-S0010482524015075
https://dx.doi.org/10.1016/j.compbiomed.2024.109422
https://www.ncbi.nlm.nih.gov/pubmed/39581125
https://www.proquest.com/docview/3146455826
https://www.proquest.com/docview/3132608873
Volume 184
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1879-0534
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0004030
  issn: 0010-4825
  databaseCode: AIEXJ
  dateStart: 19950101
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3da9swEBdpO0pfxr6XrSsa7C24xPKHJPbUbR37IGWwDvJmFEkuLo4T6iRkz_tP9pfuZEu21zSQPezFGMuSwPfz6e70uxNCb2CR0AwWQi9WUoODIobehMJdzERKSRpLHteHTdCLCzYe82-93m-XC7PKaVGw9ZrP_6uo4RkI26TO_oO4m0HhAdyD0OEKYofrToL_kFX5RMVVrk1myrzNLzKGZ-5CITV7cjYV-c-B0gstHevRBA9y7cGLamAPyZEVadXwuAaG72EqLi-rGMMKPG03slguZqYqprJ8X1f-wB4bUfFuuyWfbu_qfxXzUjuSx0iss2mWd7A7Mlv672KXuVidU56Jso3ymn2AsiGumT2obkSDRJ2Ihq61MKPcA-0Q_q2mw46i9cEtrROaN9aAOhxxbUQ4r2sYnMIs4elmFxDcfFrBIOARA8szalfFhqvomvbQAaERB815cPb5fPylzb4FPWk5YjVz8O6Jj9ChG2qbDbTNx6lsncsH6L51UvBZDa6HqKeLR-hwZAX2GP3qYgzfxhh2GMOAMWwxhhuM4azAHYzhDYxhizFcYQx3MIY7GHuCfnw8v3z_ybOneXgypJx4cphy4Ye-TgMWs0AqGkoRKj8mknMhiVSax4rKYTAJwlBFlAilIyEETbnSYFY_RfvFrNDPEWYsMjnfwtcTEiqwrlRMZUDSVJniRTTsI-q-byJtqXtz4kqeOE7jddIKKTFCSmoh9ZHf9JzX5V526MOdCBOXzgwLcAJ43KEvvauvLq1CKRM_KUkyTL5XhbQYiUC5Gh8u6qO3TU9rLNdG8I7zHjusJc1UgW9oDhEjcR-9bpphuTF7iKLQs6V5B_w9Y5kEffSsxmjzoRy8X2xteYmO2h_-GO0vbpb6FbonV4usvDlBe3TMTuzv9QfGUQQP
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Disentangled+representational+learning+for+anomaly+detection+in+single-lead+electrocardiogram+signals+using+variational+autoencoder&rft.jtitle=Computers+in+biology+and+medicine&rft.au=Kapsecker%2C+Maximilian&rft.au=M%C3%B6ller%2C+Matthias+C&rft.au=Jonas%2C+Stephan+M&rft.date=2025-01-01&rft.eissn=1879-0534&rft.volume=184&rft.spage=109422&rft_id=info:doi/10.1016%2Fj.compbiomed.2024.109422&rft_id=info%3Apmid%2F39581125&rft.externalDocID=39581125
thumbnail_m http://cvtisr.summon.serialssolutions.com/2.0.0/image/custom?url=https%3A%2F%2Fcdn.clinicalkey.com%2Fck-thumbnails%2F00104825%2FS0010482524X00165%2Fcov150h.gif