A deep learning algorithm for detecting acute myocardial infarction

Delayed diagnosis or misdiagnosis of acute myocardial infarction (AMI) is not unusual in daily practice. Since a 12-lead electrocardiogram (ECG) is crucial for the detection of AMI, a systematic algorithm to strengthen ECG interpretation may have important implications for improving diagnosis. We ai...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology Jg. 17; H. 9; S. 765
Hauptverfasser: Liu, Wen-Cheng, Lin, Chin-Sheng, Tsai, Chien-Sung, Tsao, Tien-Ping, Cheng, Cheng-Chung, Liou, Jun-Ting, Lin, Wei-Shiang, Cheng, Shu-Meng, Lou, Yu-Sheng, Lee, Chia-Cheng, Lin, Chin
Format: Journal Article
Sprache:Englisch
Veröffentlicht: France 01.10.2021
Schlagworte:
ISSN:1969-6213, 1969-6213
Online-Zugang:Weitere Angaben
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Delayed diagnosis or misdiagnosis of acute myocardial infarction (AMI) is not unusual in daily practice. Since a 12-lead electrocardiogram (ECG) is crucial for the detection of AMI, a systematic algorithm to strengthen ECG interpretation may have important implications for improving diagnosis. We aimed to develop a deep learning model (DLM) as a diagnostic support tool based on a 12-lead electrocardiogram. This retrospective cohort study included 1,051/697 ECGs from 737/287 coronary angiogram (CAG)-validated STEMI/NSTEMI patients and 140,336 ECGs from 76,775 non-AMI patients at the emergency department. The DLM was trained and validated in 80% and 20% of these ECGs. A human-machine competition was conducted. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to evaluate the performance of the DLM. The AUC of the DLM for STEMI detection was 0.976 in the human-machine competition, which was significantly better than that of the best physicians. Furthermore, the DLM independently demonstrated sufficient diagnostic capacity for STEMI detection (AUC=0.997; sensitivity, 98.4%; specificity, 96.9%). Regarding NSTEMI detection, the AUC of the combined DLM and conventional cardiac troponin I (cTnI) increased to 0.978, which was better than that of either the DLM (0.877) or cTnI (0.950). The DLM may serve as a timely, objective and precise diagnostic decision support tool to assist emergency medical system-based networks and frontline physicians in detecting AMI and subsequently initiating reperfusion therapy.
AbstractList Delayed diagnosis or misdiagnosis of acute myocardial infarction (AMI) is not unusual in daily practice. Since a 12-lead electrocardiogram (ECG) is crucial for the detection of AMI, a systematic algorithm to strengthen ECG interpretation may have important implications for improving diagnosis.BACKGROUNDDelayed diagnosis or misdiagnosis of acute myocardial infarction (AMI) is not unusual in daily practice. Since a 12-lead electrocardiogram (ECG) is crucial for the detection of AMI, a systematic algorithm to strengthen ECG interpretation may have important implications for improving diagnosis.We aimed to develop a deep learning model (DLM) as a diagnostic support tool based on a 12-lead electrocardiogram.AIMSWe aimed to develop a deep learning model (DLM) as a diagnostic support tool based on a 12-lead electrocardiogram.This retrospective cohort study included 1,051/697 ECGs from 737/287 coronary angiogram (CAG)-validated STEMI/NSTEMI patients and 140,336 ECGs from 76,775 non-AMI patients at the emergency department. The DLM was trained and validated in 80% and 20% of these ECGs. A human-machine competition was conducted. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to evaluate the performance of the DLM.METHODSThis retrospective cohort study included 1,051/697 ECGs from 737/287 coronary angiogram (CAG)-validated STEMI/NSTEMI patients and 140,336 ECGs from 76,775 non-AMI patients at the emergency department. The DLM was trained and validated in 80% and 20% of these ECGs. A human-machine competition was conducted. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to evaluate the performance of the DLM.The AUC of the DLM for STEMI detection was 0.976 in the human-machine competition, which was significantly better than that of the best physicians. Furthermore, the DLM independently demonstrated sufficient diagnostic capacity for STEMI detection (AUC=0.997; sensitivity, 98.4%; specificity, 96.9%). Regarding NSTEMI detection, the AUC of the combined DLM and conventional cardiac troponin I (cTnI) increased to 0.978, which was better than that of either the DLM (0.877) or cTnI (0.950).RESULTSThe AUC of the DLM for STEMI detection was 0.976 in the human-machine competition, which was significantly better than that of the best physicians. Furthermore, the DLM independently demonstrated sufficient diagnostic capacity for STEMI detection (AUC=0.997; sensitivity, 98.4%; specificity, 96.9%). Regarding NSTEMI detection, the AUC of the combined DLM and conventional cardiac troponin I (cTnI) increased to 0.978, which was better than that of either the DLM (0.877) or cTnI (0.950).The DLM may serve as a timely, objective and precise diagnostic decision support tool to assist emergency medical system-based networks and frontline physicians in detecting AMI and subsequently initiating reperfusion therapy.CONCLUSIONSThe DLM may serve as a timely, objective and precise diagnostic decision support tool to assist emergency medical system-based networks and frontline physicians in detecting AMI and subsequently initiating reperfusion therapy.
Delayed diagnosis or misdiagnosis of acute myocardial infarction (AMI) is not unusual in daily practice. Since a 12-lead electrocardiogram (ECG) is crucial for the detection of AMI, a systematic algorithm to strengthen ECG interpretation may have important implications for improving diagnosis. We aimed to develop a deep learning model (DLM) as a diagnostic support tool based on a 12-lead electrocardiogram. This retrospective cohort study included 1,051/697 ECGs from 737/287 coronary angiogram (CAG)-validated STEMI/NSTEMI patients and 140,336 ECGs from 76,775 non-AMI patients at the emergency department. The DLM was trained and validated in 80% and 20% of these ECGs. A human-machine competition was conducted. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to evaluate the performance of the DLM. The AUC of the DLM for STEMI detection was 0.976 in the human-machine competition, which was significantly better than that of the best physicians. Furthermore, the DLM independently demonstrated sufficient diagnostic capacity for STEMI detection (AUC=0.997; sensitivity, 98.4%; specificity, 96.9%). Regarding NSTEMI detection, the AUC of the combined DLM and conventional cardiac troponin I (cTnI) increased to 0.978, which was better than that of either the DLM (0.877) or cTnI (0.950). The DLM may serve as a timely, objective and precise diagnostic decision support tool to assist emergency medical system-based networks and frontline physicians in detecting AMI and subsequently initiating reperfusion therapy.
Author Cheng, Cheng-Chung
Lin, Wei-Shiang
Liu, Wen-Cheng
Tsao, Tien-Ping
Lou, Yu-Sheng
Lee, Chia-Cheng
Tsai, Chien-Sung
Cheng, Shu-Meng
Liou, Jun-Ting
Lin, Chin-Sheng
Lin, Chin
Author_xml – sequence: 1
  givenname: Wen-Cheng
  surname: Liu
  fullname: Liu, Wen-Cheng
  organization: Division of Cardiology, Department of Internal Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan, R.O.C
– sequence: 2
  givenname: Chin-Sheng
  surname: Lin
  fullname: Lin, Chin-Sheng
– sequence: 3
  givenname: Chien-Sung
  surname: Tsai
  fullname: Tsai, Chien-Sung
– sequence: 4
  givenname: Tien-Ping
  surname: Tsao
  fullname: Tsao, Tien-Ping
– sequence: 5
  givenname: Cheng-Chung
  surname: Cheng
  fullname: Cheng, Cheng-Chung
– sequence: 6
  givenname: Jun-Ting
  surname: Liou
  fullname: Liou, Jun-Ting
– sequence: 7
  givenname: Wei-Shiang
  surname: Lin
  fullname: Lin, Wei-Shiang
– sequence: 8
  givenname: Shu-Meng
  surname: Cheng
  fullname: Cheng, Shu-Meng
– sequence: 9
  givenname: Yu-Sheng
  surname: Lou
  fullname: Lou, Yu-Sheng
– sequence: 10
  givenname: Chia-Cheng
  surname: Lee
  fullname: Lee, Chia-Cheng
– sequence: 11
  givenname: Chin
  surname: Lin
  fullname: Lin, Chin
BackLink https://www.ncbi.nlm.nih.gov/pubmed/33840640$$D View this record in MEDLINE/PubMed
BookMark eNpNjz1PwzAYhC1UBKWwMqKMLAa_jhPssQoFiiqxwBz543WJlNjBSYb-eyIoEtOd7jmddBdkEWJAQq6B3QkuxP1m-0ofKWeUARTFCVmCKhUtOeSLf_6MnOe5FKwUbEmqdeYQ-6xFnUIT9plu9zE142eX-ZhmNqIdf3I7jZh1h2h1co1usyZ4nWYWwyU59bod8OqoK_LxtHmvXuju7XlbrXfU5nkx0tJp65i3YJxkJQPvpPJaFgaF0AakVuCUNF6KArh5YE5YzSxyQGWkcsBX5PZ3t0_xa8JhrLtmsNi2OmCchpoXAFJJrvhcvTlWJ9Ohq_vUdDod6r_j_Btg0VmQ
CitedBy_id crossref_primary_10_3390_diagnostics13172723
crossref_primary_10_3390_jpm11080725
crossref_primary_10_1136_bmjdhai_2025_000115
crossref_primary_10_2459_JCM_0000000000001716
crossref_primary_10_1093_eurheartj_ehaf254
crossref_primary_10_3390_jpm12071150
crossref_primary_10_1016_j_cca_2022_09_021
crossref_primary_10_1177_20552076221143249
crossref_primary_10_1515_cclm_2024_1090
crossref_primary_10_1136_openhrt_2022_002237
crossref_primary_10_1016_j_cmpb_2023_107359
crossref_primary_10_2196_44791
crossref_primary_10_1007_s10916_025_02177_0
crossref_primary_10_3390_life15040515
crossref_primary_10_1016_j_ijmedinf_2024_105683
crossref_primary_10_1111_eci_70002
crossref_primary_10_1007_s43678_023_00598_9
crossref_primary_10_3390_jpm12020315
crossref_primary_10_1097_CRD_0000000000001016
crossref_primary_10_1166_jmihi_2021_3913
crossref_primary_10_3390_jpm12050700
crossref_primary_10_1097_SCS_0000000000010111
crossref_primary_10_3390_jpm12030336
crossref_primary_10_1016_j_annemergmed_2024_06_004
crossref_primary_10_3389_fcvm_2022_754909
crossref_primary_10_3390_jcdd9020056
crossref_primary_10_1186_s12872_023_03326_4
crossref_primary_10_3389_frai_2025_1618492
crossref_primary_10_1093_ehjdh_ztaf096
crossref_primary_10_1093_ehjdh_ztaf049
crossref_primary_10_1109_ACCESS_2024_3447096
crossref_primary_10_1210_jendso_bvab120
crossref_primary_10_1016_j_jacadv_2025_102011
crossref_primary_10_3390_jpm11111149
crossref_primary_10_4037_ajcc2025597
crossref_primary_10_3389_fcvm_2025_1525966
crossref_primary_10_3349_ymj_2024_0148
crossref_primary_10_1007_s11886_024_02062_1
crossref_primary_10_1007_s43678_023_00572_5
crossref_primary_10_1186_s12938_025_01349_w
crossref_primary_10_1016_j_ajem_2024_06_026
crossref_primary_10_3389_fcvm_2022_895201
crossref_primary_10_1007_s12325_024_03060_z
crossref_primary_10_3389_fcvm_2022_876543
crossref_primary_10_3390_jpm12030455
crossref_primary_10_1007_s10916_024_02088_6
crossref_primary_10_3390_diagnostics14111103
crossref_primary_10_1016_j_ijcard_2022_06_012
crossref_primary_10_3390_diagnostics14171839
ContentType Journal Article
DBID CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.4244/EIJ-D-20-01155
DatabaseName Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic
MEDLINE
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 no_fulltext_linktorsrc
EISSN 1969-6213
ExternalDocumentID 33840640
Genre Journal Article
GroupedDBID CGR
CUY
CVF
ECM
EIF
NPM
7X8
ID FETCH-LOGICAL-c335t-6dacd0fc1bd80601fd89fa85be44ab18a91d98bf84512b70d4ca0ce21e9b89d12
IEDL.DBID 7X8
ISICitedReferencesCount 62
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000722529400012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1969-6213
IngestDate Fri Jul 11 02:10:11 EDT 2025
Thu Jan 02 22:53:51 EST 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 9
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c335t-6dacd0fc1bd80601fd89fa85be44ab18a91d98bf84512b70d4ca0ce21e9b89d12
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
OpenAccessLink https://www.ncbi.nlm.nih.gov/pmc/articles/9724911
PMID 33840640
PQID 2511898292
PQPubID 23479
ParticipantIDs proquest_miscellaneous_2511898292
pubmed_primary_33840640
PublicationCentury 2000
PublicationDate 2021-10-01
PublicationDateYYYYMMDD 2021-10-01
PublicationDate_xml – month: 10
  year: 2021
  text: 2021-10-01
  day: 01
PublicationDecade 2020
PublicationPlace France
PublicationPlace_xml – name: France
PublicationTitle EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology
PublicationTitleAlternate EuroIntervention
PublicationYear 2021
Score 2.5285797
Snippet Delayed diagnosis or misdiagnosis of acute myocardial infarction (AMI) is not unusual in daily practice. Since a 12-lead electrocardiogram (ECG) is crucial for...
SourceID proquest
pubmed
SourceType Aggregation Database
Index Database
StartPage 765
SubjectTerms Algorithms
Deep Learning
Electrocardiography
Humans
Myocardial Infarction - diagnosis
Retrospective Studies
Sensitivity and Specificity
Troponin I
Title A deep learning algorithm for detecting acute myocardial infarction
URI https://www.ncbi.nlm.nih.gov/pubmed/33840640
https://www.proquest.com/docview/2511898292
Volume 17
WOSCitedRecordID wos000722529400012&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LSwMxEA5qPXjxga_6IoLX0N1stpucpNQWFSw9KPS25DGpQl_areC_d7K7xZMgeMlhQ2CZ-TLzJTOZIeQm5VJCFgNLIusZWr-U6cwDi53iNgsVzMGXzSaywUCORmpYX7gt67TKtU0sDbWb23BH3iqpsJJc8dvFOwtdo0J0tW6hsUkaCVKZgPNsVL5-U23F2jxOqjqN4TVXq_fwyO4QFSywoPR3Rll6lv7ef_9pn-zWnJJ2KhAckA2YHZJuhzqABa3bQoypnoxxbfE6pchTcS5ED8rvdlUAnX6hUwtgmVDEHKI_KOyIvPR7z917VndMYDZJ0oK1nbYu8jY2ToZCK95J5bVMDQihTSy1Qh1I46VAP2-yyAmrIws8BmWkcjE_Jluz-QxOCRVJZpwTqfQG97nBcwd6cgAPSWy4MLxJrtcyyRGRIcygZzBfLfMfqTTJSSXYfFGVzsjxQCxC7PDsD6vPyQ4PCSRl5twFaXjcj3BJtu1n8bb8uCpVjeNg-PQNCDW1Wg
linkProvider ProQuest
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=A+deep+learning+algorithm+for+detecting+acute+myocardial+infarction&rft.jtitle=EuroIntervention+%3A+journal+of+EuroPCR+in+collaboration+with+the+Working+Group+on+Interventional+Cardiology+of+the+European+Society+of+Cardiology&rft.au=Liu%2C+Wen-Cheng&rft.au=Lin%2C+Chin-Sheng&rft.au=Tsai%2C+Chien-Sung&rft.au=Tsao%2C+Tien-Ping&rft.date=2021-10-01&rft.issn=1969-6213&rft.eissn=1969-6213&rft.volume=17&rft.issue=9&rft.spage=765&rft_id=info:doi/10.4244%2FEIJ-D-20-01155&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1969-6213&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1969-6213&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1969-6213&client=summon