ECG-Based Deep Learning and Clinical Risk Factors to Predict Atrial Fibrillation

Artificial intelligence (AI)-enabled analysis of 12-lead ECGs may facilitate efficient estimation of incident atrial fibrillation (AF) risk. However, it remains unclear whether AI provides meaningful and generalizable improvement in predictive accuracy beyond clinical risk factors for AF. We trained...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Circulation (New York, N.Y.) Ročník 145; číslo 2; s. 122
Hlavní autoři:	Khurshid, Shaan, Friedman, Samuel, Reeder, Christopher, Di Achille, Paolo, Diamant, Nathaniel, Singh, Pulkit, Harrington, Lia X, Wang, Xin, Al-Alusi, Mostafa A, Sarma, Gopal, Foulkes, Andrea S, Ellinor, Patrick T, Anderson, Christopher D, Ho, Jennifer E, Philippakis, Anthony A, Batra, Puneet, Lubitz, Steven A
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	United States 11.01.2022
Témata:	Atrial Fibrillation - diagnosis Atrial Fibrillation - pathology Deep Learning - standards Electrocardiography - methods Female Humans Male Middle Aged Risk Factors deep learning atrial fibrillation electronic health records
ISSN:	1524-4539, 1524-4539
On-line přístup:	Zjistit podrobnosti o přístupu
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	Artificial intelligence (AI)-enabled analysis of 12-lead ECGs may facilitate efficient estimation of incident atrial fibrillation (AF) risk. However, it remains unclear whether AI provides meaningful and generalizable improvement in predictive accuracy beyond clinical risk factors for AF. We trained a convolutional neural network (ECG-AI) to infer 5-year incident AF risk using 12-lead ECGs in patients receiving longitudinal primary care at Massachusetts General Hospital (MGH). We then fit 3 Cox proportional hazards models, composed of ECG-AI 5-year AF probability, CHARGE-AF clinical risk score (Cohorts for Heart and Aging in Genomic Epidemiology-Atrial Fibrillation), and terms for both ECG-AI and CHARGE-AF (CH-AI), respectively. We assessed model performance by calculating discrimination (area under the receiver operating characteristic curve) and calibration in an internal test set and 2 external test sets (Brigham and Women's Hospital [BWH] and UK Biobank). Models were recalibrated to estimate 2-year AF risk in the UK Biobank given limited available follow-up. We used saliency mapping to identify ECG features most influential on ECG-AI risk predictions and assessed correlation between ECG-AI and CHARGE-AF linear predictors. The training set comprised 45 770 individuals (age 55±17 years, 53% women, 2171 AF events) and the test sets comprised 83 162 individuals (age 59±13 years, 56% women, 2424 AF events). Area under the receiver operating characteristic curve was comparable using CHARGE-AF (MGH, 0.802 [95% CI, 0.767-0.836]; BWH, 0.752 [95% CI, 0.741-0.763]; UK Biobank, 0.732 [95% CI, 0.704-0.759]) and ECG-AI (MGH, 0.823 [95% CI, 0.790-0.856]; BWH, 0.747 [95% CI, 0.736-0.759]; UK Biobank, 0.705 [95% CI, 0.673-0.737]). Area under the receiver operating characteristic curve was highest using CH-AI (MGH, 0.838 [95% CI, 0.807 to 0.869]; BWH, 0.777 [95% CI, 0.766 to 0.788]; UK Biobank, 0.746 [95% CI, 0.716 to 0.776]). Calibration error was low using ECG-AI (MGH, 0.0212; BWH, 0.0129; UK Biobank, 0.0035) and CH-AI (MGH, 0.012; BWH, 0.0108; UK Biobank, 0.0001). In saliency analyses, the ECG P-wave had the greatest influence on AI model predictions. ECG-AI and CHARGE-AF linear predictors were correlated (Pearson : MGH, 0.61; BWH, 0.66; UK Biobank, 0.41). AI-based analysis of 12-lead ECGs has similar predictive usefulness to a clinical risk factor model for incident AF and the approaches are complementary. ECG-AI may enable efficient quantification of future AF risk.
AbstractList	Artificial intelligence (AI)-enabled analysis of 12-lead ECGs may facilitate efficient estimation of incident atrial fibrillation (AF) risk. However, it remains unclear whether AI provides meaningful and generalizable improvement in predictive accuracy beyond clinical risk factors for AF. We trained a convolutional neural network (ECG-AI) to infer 5-year incident AF risk using 12-lead ECGs in patients receiving longitudinal primary care at Massachusetts General Hospital (MGH). We then fit 3 Cox proportional hazards models, composed of ECG-AI 5-year AF probability, CHARGE-AF clinical risk score (Cohorts for Heart and Aging in Genomic Epidemiology-Atrial Fibrillation), and terms for both ECG-AI and CHARGE-AF (CH-AI), respectively. We assessed model performance by calculating discrimination (area under the receiver operating characteristic curve) and calibration in an internal test set and 2 external test sets (Brigham and Women's Hospital [BWH] and UK Biobank). Models were recalibrated to estimate 2-year AF risk in the UK Biobank given limited available follow-up. We used saliency mapping to identify ECG features most influential on ECG-AI risk predictions and assessed correlation between ECG-AI and CHARGE-AF linear predictors. The training set comprised 45 770 individuals (age 55±17 years, 53% women, 2171 AF events) and the test sets comprised 83 162 individuals (age 59±13 years, 56% women, 2424 AF events). Area under the receiver operating characteristic curve was comparable using CHARGE-AF (MGH, 0.802 [95% CI, 0.767-0.836]; BWH, 0.752 [95% CI, 0.741-0.763]; UK Biobank, 0.732 [95% CI, 0.704-0.759]) and ECG-AI (MGH, 0.823 [95% CI, 0.790-0.856]; BWH, 0.747 [95% CI, 0.736-0.759]; UK Biobank, 0.705 [95% CI, 0.673-0.737]). Area under the receiver operating characteristic curve was highest using CH-AI (MGH, 0.838 [95% CI, 0.807 to 0.869]; BWH, 0.777 [95% CI, 0.766 to 0.788]; UK Biobank, 0.746 [95% CI, 0.716 to 0.776]). Calibration error was low using ECG-AI (MGH, 0.0212; BWH, 0.0129; UK Biobank, 0.0035) and CH-AI (MGH, 0.012; BWH, 0.0108; UK Biobank, 0.0001). In saliency analyses, the ECG P-wave had the greatest influence on AI model predictions. ECG-AI and CHARGE-AF linear predictors were correlated (Pearson : MGH, 0.61; BWH, 0.66; UK Biobank, 0.41). AI-based analysis of 12-lead ECGs has similar predictive usefulness to a clinical risk factor model for incident AF and the approaches are complementary. ECG-AI may enable efficient quantification of future AF risk. Artificial intelligence (AI)-enabled analysis of 12-lead ECGs may facilitate efficient estimation of incident atrial fibrillation (AF) risk. However, it remains unclear whether AI provides meaningful and generalizable improvement in predictive accuracy beyond clinical risk factors for AF.BACKGROUNDArtificial intelligence (AI)-enabled analysis of 12-lead ECGs may facilitate efficient estimation of incident atrial fibrillation (AF) risk. However, it remains unclear whether AI provides meaningful and generalizable improvement in predictive accuracy beyond clinical risk factors for AF.We trained a convolutional neural network (ECG-AI) to infer 5-year incident AF risk using 12-lead ECGs in patients receiving longitudinal primary care at Massachusetts General Hospital (MGH). We then fit 3 Cox proportional hazards models, composed of ECG-AI 5-year AF probability, CHARGE-AF clinical risk score (Cohorts for Heart and Aging in Genomic Epidemiology-Atrial Fibrillation), and terms for both ECG-AI and CHARGE-AF (CH-AI), respectively. We assessed model performance by calculating discrimination (area under the receiver operating characteristic curve) and calibration in an internal test set and 2 external test sets (Brigham and Women's Hospital [BWH] and UK Biobank). Models were recalibrated to estimate 2-year AF risk in the UK Biobank given limited available follow-up. We used saliency mapping to identify ECG features most influential on ECG-AI risk predictions and assessed correlation between ECG-AI and CHARGE-AF linear predictors.METHODSWe trained a convolutional neural network (ECG-AI) to infer 5-year incident AF risk using 12-lead ECGs in patients receiving longitudinal primary care at Massachusetts General Hospital (MGH). We then fit 3 Cox proportional hazards models, composed of ECG-AI 5-year AF probability, CHARGE-AF clinical risk score (Cohorts for Heart and Aging in Genomic Epidemiology-Atrial Fibrillation), and terms for both ECG-AI and CHARGE-AF (CH-AI), respectively. We assessed model performance by calculating discrimination (area under the receiver operating characteristic curve) and calibration in an internal test set and 2 external test sets (Brigham and Women's Hospital [BWH] and UK Biobank). Models were recalibrated to estimate 2-year AF risk in the UK Biobank given limited available follow-up. We used saliency mapping to identify ECG features most influential on ECG-AI risk predictions and assessed correlation between ECG-AI and CHARGE-AF linear predictors.The training set comprised 45 770 individuals (age 55±17 years, 53% women, 2171 AF events) and the test sets comprised 83 162 individuals (age 59±13 years, 56% women, 2424 AF events). Area under the receiver operating characteristic curve was comparable using CHARGE-AF (MGH, 0.802 [95% CI, 0.767-0.836]; BWH, 0.752 [95% CI, 0.741-0.763]; UK Biobank, 0.732 [95% CI, 0.704-0.759]) and ECG-AI (MGH, 0.823 [95% CI, 0.790-0.856]; BWH, 0.747 [95% CI, 0.736-0.759]; UK Biobank, 0.705 [95% CI, 0.673-0.737]). Area under the receiver operating characteristic curve was highest using CH-AI (MGH, 0.838 [95% CI, 0.807 to 0.869]; BWH, 0.777 [95% CI, 0.766 to 0.788]; UK Biobank, 0.746 [95% CI, 0.716 to 0.776]). Calibration error was low using ECG-AI (MGH, 0.0212; BWH, 0.0129; UK Biobank, 0.0035) and CH-AI (MGH, 0.012; BWH, 0.0108; UK Biobank, 0.0001). In saliency analyses, the ECG P-wave had the greatest influence on AI model predictions. ECG-AI and CHARGE-AF linear predictors were correlated (Pearson r: MGH, 0.61; BWH, 0.66; UK Biobank, 0.41).RESULTSThe training set comprised 45 770 individuals (age 55±17 years, 53% women, 2171 AF events) and the test sets comprised 83 162 individuals (age 59±13 years, 56% women, 2424 AF events). Area under the receiver operating characteristic curve was comparable using CHARGE-AF (MGH, 0.802 [95% CI, 0.767-0.836]; BWH, 0.752 [95% CI, 0.741-0.763]; UK Biobank, 0.732 [95% CI, 0.704-0.759]) and ECG-AI (MGH, 0.823 [95% CI, 0.790-0.856]; BWH, 0.747 [95% CI, 0.736-0.759]; UK Biobank, 0.705 [95% CI, 0.673-0.737]). Area under the receiver operating characteristic curve was highest using CH-AI (MGH, 0.838 [95% CI, 0.807 to 0.869]; BWH, 0.777 [95% CI, 0.766 to 0.788]; UK Biobank, 0.746 [95% CI, 0.716 to 0.776]). Calibration error was low using ECG-AI (MGH, 0.0212; BWH, 0.0129; UK Biobank, 0.0035) and CH-AI (MGH, 0.012; BWH, 0.0108; UK Biobank, 0.0001). In saliency analyses, the ECG P-wave had the greatest influence on AI model predictions. ECG-AI and CHARGE-AF linear predictors were correlated (Pearson r: MGH, 0.61; BWH, 0.66; UK Biobank, 0.41).AI-based analysis of 12-lead ECGs has similar predictive usefulness to a clinical risk factor model for incident AF and the approaches are complementary. ECG-AI may enable efficient quantification of future AF risk.CONCLUSIONSAI-based analysis of 12-lead ECGs has similar predictive usefulness to a clinical risk factor model for incident AF and the approaches are complementary. ECG-AI may enable efficient quantification of future AF risk.
Author	Reeder, Christopher Diamant, Nathaniel Harrington, Lia X Singh, Pulkit Friedman, Samuel Wang, Xin Philippakis, Anthony A Ellinor, Patrick T Di Achille, Paolo Khurshid, Shaan Sarma, Gopal Foulkes, Andrea S Ho, Jennifer E Lubitz, Steven A Anderson, Christopher D Al-Alusi, Mostafa A Batra, Puneet
Author_xml	– sequence: 1 givenname: Shaan orcidid: 0000-0002-2840-4539 surname: Khurshid fullname: Khurshid, Shaan organization: Cardiovascular Disease Initiative (S.K., L.X.H., X.W., M.A.A., P.T.E., J.E.H., S.A.L.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 2 givenname: Samuel surname: Friedman fullname: Friedman, Samuel organization: Data Sciences Platform (S.F., C.R., P.D.A., N.D., P.S., G.S., A.A.P., P.B.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 3 givenname: Christopher orcidid: 0000-0002-3893-2423 surname: Reeder fullname: Reeder, Christopher organization: Data Sciences Platform (S.F., C.R., P.D.A., N.D., P.S., G.S., A.A.P., P.B.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 4 givenname: Paolo orcidid: 0000-0001-9256-0678 surname: Di Achille fullname: Di Achille, Paolo organization: Data Sciences Platform (S.F., C.R., P.D.A., N.D., P.S., G.S., A.A.P., P.B.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 5 givenname: Nathaniel surname: Diamant fullname: Diamant, Nathaniel organization: Data Sciences Platform (S.F., C.R., P.D.A., N.D., P.S., G.S., A.A.P., P.B.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 6 givenname: Pulkit surname: Singh fullname: Singh, Pulkit organization: Data Sciences Platform (S.F., C.R., P.D.A., N.D., P.S., G.S., A.A.P., P.B.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 7 givenname: Lia X surname: Harrington fullname: Harrington, Lia X organization: Cardiovascular Disease Initiative (S.K., L.X.H., X.W., M.A.A., P.T.E., J.E.H., S.A.L.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 8 givenname: Xin surname: Wang fullname: Wang, Xin organization: Cardiovascular Disease Initiative (S.K., L.X.H., X.W., M.A.A., P.T.E., J.E.H., S.A.L.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 9 givenname: Mostafa A surname: Al-Alusi fullname: Al-Alusi, Mostafa A organization: Data Sciences Platform (S.F., C.R., P.D.A., N.D., P.S., G.S., A.A.P., P.B.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 10 givenname: Gopal surname: Sarma fullname: Sarma, Gopal organization: Data Sciences Platform (S.F., C.R., P.D.A., N.D., P.S., G.S., A.A.P., P.B.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 11 givenname: Andrea S orcidid: 0000-0002-9520-0501 surname: Foulkes fullname: Foulkes, Andrea S organization: Biostatistics Center (A.S.F.), Massachusetts General Hospital, Boston – sequence: 12 givenname: Patrick T orcidid: 0000-0002-2067-0533 surname: Ellinor fullname: Ellinor, Patrick T organization: Cardiovascular Disease Initiative (S.K., L.X.H., X.W., M.A.A., P.T.E., J.E.H., S.A.L.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 13 givenname: Christopher D orcidid: 0000-0002-0053-2002 surname: Anderson fullname: Anderson, Christopher D organization: Department of Neurology, Brigham and Women's Hospital, Boston, MA (C.D.A.) – sequence: 14 givenname: Jennifer E orcidid: 0000-0002-7987-4768 surname: Ho fullname: Ho, Jennifer E organization: Harvard Medical School, Boston, MA (A.S.F., P.T.E., C.D.A., J.E.H., S.A.L.) – sequence: 15 givenname: Anthony A surname: Philippakis fullname: Philippakis, Anthony A organization: Eric and Wendy Schmidt Center (A.A.P.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 16 givenname: Puneet orcidid: 0000-0001-6822-0593 surname: Batra fullname: Batra, Puneet organization: Data Sciences Platform (S.F., C.R., P.D.A., N.D., P.S., G.S., A.A.P., P.B.), Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge – sequence: 17 givenname: Steven A orcidid: 0000-0002-9599-4866 surname: Lubitz fullname: Lubitz, Steven A organization: Harvard Medical School, Boston, MA (A.S.F., P.T.E., C.D.A., J.E.H., S.A.L.)
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/34743566$$D View this record in MEDLINE/PubMed
BookMark	eNpNkEtLxDAAhIOsuA_9CxJvXromaZI2x1r3BcVdlvVc8qpEu-madA_-ewuu4GkG5mMYZgpGvvMWgAeM5hhz_FRu9uVbVRw229diXcwxwXPEMpqjKzDBjNCEslSM_vkxmMb4gRDiacZuwDilGU0Z5xOwW5Sr5FlGa-CLtSdYWRm88-9QegPL1nmnZQv3Ln7CpdR9FyLsO7gL1jjdw6IPboiXTgXXtrJ3nb8F141so7276AwclotDuU6q7WpTFlWiaSZ4oixOLTEME9Ew00hrtVY55pqoXGmis8YKJagWFBskDDU5ppQPDOUNUimZgcff2lPovs429vXRRW2HEd5251gTJhjGmDE6oPcX9KyO1tSn4I4yfNd_J5Afv19ifw
CitedBy_id	crossref_primary_10_1038_s41746_025_01491_8 crossref_primary_10_1016_j_jacadv_2025_102041 crossref_primary_10_1161_JAHA_122_026974 crossref_primary_10_3389_fcvm_2023_1212837 crossref_primary_10_1016_j_mayocp_2022_11_020 crossref_primary_10_1093_eurheartj_ehad431 crossref_primary_10_1136_fmch_2023_002625 crossref_primary_10_1515_cclm_2023_0743 crossref_primary_10_1016_j_bspc_2024_107028 crossref_primary_10_1093_eurheartj_ehad670 crossref_primary_10_1186_s40001_023_01065_y crossref_primary_10_1016_j_cmet_2024_02_002 crossref_primary_10_1002_widm_1530 crossref_primary_10_1007_s11042_025_21044_1 crossref_primary_10_1016_j_jelectrocard_2023_07_002 crossref_primary_10_1016_j_medj_2025_100668 crossref_primary_10_1055_a_2566_7244 crossref_primary_10_1038_s41598_022_25284_1 crossref_primary_10_3390_s24154787 crossref_primary_10_1007_s00059_025_05298_x crossref_primary_10_1016_j_compbiomed_2024_109088 crossref_primary_10_1007_s12170_024_00747_4 crossref_primary_10_1053_j_akdh_2022_11_009 crossref_primary_10_1097_HCO_0000000000001031 crossref_primary_10_1016_j_anclin_2025_05_005 crossref_primary_10_1161_CIRCULATIONAHA_121_058678 crossref_primary_10_14309_ctg_0000000000000771 crossref_primary_10_1038_s41746_023_00806_x crossref_primary_10_1093_eurheartj_ehae595 crossref_primary_10_1007_s11831_023_09935_8 crossref_primary_10_3390_s24092705 crossref_primary_10_1007_s10916_025_02177_0 crossref_primary_10_1055_a_2559_9994 crossref_primary_10_1016_j_cpcardiol_2022_101482 crossref_primary_10_1016_j_medj_2024_02_006 crossref_primary_10_2174_011573403X334095241205041550 crossref_primary_10_1161_CIRCGEN_122_003808 crossref_primary_10_1016_j_media_2024_103451 crossref_primary_10_3390_jpm13020347 crossref_primary_10_3390_cancers16010208 crossref_primary_10_3390_jcm13103003 crossref_primary_10_1038_s41467_025_58283_7 crossref_primary_10_1371_journal_pone_0299932 crossref_primary_10_3389_fcvm_2024_1473482 crossref_primary_10_1002_anse_202200062 crossref_primary_10_1088_1361_6579_ad55a1 crossref_primary_10_1093_europace_euae201 crossref_primary_10_1007_s11886_024_02136_0 crossref_primary_10_1007_s00399_022_00839_x crossref_primary_10_1016_j_ijcha_2025_101783 crossref_primary_10_1038_s44385_024_00001_x crossref_primary_10_1186_s12911_024_02620_1 crossref_primary_10_1038_s41746_023_00966_w crossref_primary_10_1097_CD9_0000000000000155 crossref_primary_10_3390_math13172872 crossref_primary_10_3389_fcvm_2023_1160091 crossref_primary_10_3390_jcm13051313 crossref_primary_10_1038_s41598_024_60219_y crossref_primary_10_1016_j_bspc_2024_107255 crossref_primary_10_1016_j_clnu_2024_07_046 crossref_primary_10_1177_17474930241302272 crossref_primary_10_1016_j_irbm_2023_100811 crossref_primary_10_1038_s41746_024_01234_1 crossref_primary_10_3748_wjg_v30_i10_1270 crossref_primary_10_1016_j_cvdhj_2022_06_001 crossref_primary_10_1371_journal_pone_0305339 crossref_primary_10_3390_jcm12134484 crossref_primary_10_1109_ACCESS_2022_3231743 crossref_primary_10_1016_j_artmed_2023_102548 crossref_primary_10_1038_s41598_022_27254_z crossref_primary_10_4103_jpbs_jpbs_557_25 crossref_primary_10_1016_j_hrthm_2024_08_001 crossref_primary_10_1016_j_hrthm_2025_08_024 crossref_primary_10_1093_eurheartj_ehae651 crossref_primary_10_1016_j_ajpc_2025_100951 crossref_primary_10_1016_j_semnephrol_2024_151518 crossref_primary_10_3390_jpm12101608 crossref_primary_10_3389_fpubh_2021_818439 crossref_primary_10_1136_heartjnl_2024_324177 crossref_primary_10_3390_bioengineering12090961 crossref_primary_10_3390_s24154978 crossref_primary_10_1161_CIRCEP_124_012959 crossref_primary_10_3390_healthcare12141380 crossref_primary_10_1159_000539837 crossref_primary_10_1186_s13321_022_00590_y crossref_primary_10_1016_j_jelectrocard_2023_08_011 crossref_primary_10_1016_j_cvdhj_2022_09_001 crossref_primary_10_1016_j_hlc_2024_08_008 crossref_primary_10_1016_j_jacep_2024_01_022 crossref_primary_10_1016_j_cmpb_2024_108164 crossref_primary_10_1186_s12874_023_01989_3 crossref_primary_10_1007_s00395_024_01038_0 crossref_primary_10_1093_eurheartj_ehae691 crossref_primary_10_1016_j_jacc_2024_03_400 crossref_primary_10_1161_CIRCOUTCOMES_123_010602 crossref_primary_10_1161_CIRCGEN_124_004943 crossref_primary_10_3390_biomedicines13071685 crossref_primary_10_1016_j_jacep_2023_04_008 crossref_primary_10_3390_jpm12071150 crossref_primary_10_1016_j_acvd_2024_02_001 crossref_primary_10_3390_jcm12155066 crossref_primary_10_1016_j_cell_2025_05_018 crossref_primary_10_1161_CIR_0000000000001201 crossref_primary_10_1016_j_jacc_2025_07_031 crossref_primary_10_1016_j_sigpro_2025_110068 crossref_primary_10_1038_s41440_023_01469_7 crossref_primary_10_1136_heartjnl_2024_324954 crossref_primary_10_3390_jcm14082627 crossref_primary_10_1093_ehjdh_ztaf100 crossref_primary_10_1016_j_jacadv_2023_100686 crossref_primary_10_1016_j_rccl_2025_07_001 crossref_primary_10_1093_ehjdh_ztae095 crossref_primary_10_1213_ANE_0000000000006789 crossref_primary_10_1016_j_ahj_2025_08_019 crossref_primary_10_1007_s00500_023_08680_1 crossref_primary_10_3390_jcm11144004 crossref_primary_10_1016_j_cvdhj_2023_11_003 crossref_primary_10_1097_MD_0000000000038264 crossref_primary_10_1016_j_cpcardiol_2023_102181 crossref_primary_10_1016_j_jare_2025_08_036 crossref_primary_10_1016_j_jacep_2024_02_011 crossref_primary_10_1093_ehjdh_ztaf054 crossref_primary_10_1002_hup_2889 crossref_primary_10_1016_j_gerinurse_2023_02_007 crossref_primary_10_1161_CIRCOUTCOMES_122_009821 crossref_primary_10_1016_j_jacadv_2024_100998 crossref_primary_10_1109_TBME_2023_3321792 crossref_primary_10_1016_j_compbiomed_2024_108097 crossref_primary_10_3389_fdgth_2025_1547208 crossref_primary_10_1007_s00399_024_00997_0 crossref_primary_10_1038_s41569_025_01130_5 crossref_primary_10_1161_CIRCULATIONAHA_123_067750 crossref_primary_10_1007_s10462_024_10852_w crossref_primary_10_2196_51375 crossref_primary_10_1055_a_2566_7133 crossref_primary_10_3390_diagnostics14232675 crossref_primary_10_1080_17434440_2025_2514008 crossref_primary_10_3389_fcvm_2024_1401143 crossref_primary_10_1016_j_jelectrocard_2022_11_001 crossref_primary_10_1016_j_neunet_2025_107835 crossref_primary_10_1093_eurjpc_zwad321 crossref_primary_10_1038_s41598_025_14579_8 crossref_primary_10_1007_s11886_023_01859_w crossref_primary_10_1038_s41467_023_39472_8 crossref_primary_10_1016_j_artmed_2024_103065
ContentType	Journal Article
DBID	CGR CUY CVF ECM EIF NPM 7X8
DOI	10.1161/CIRCULATIONAHA.121.057480
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 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	no_fulltext_linktorsrc
Discipline	Medicine Anatomy & Physiology
EISSN	1524-4539
ExternalDocumentID	34743566
Genre	Research Support, Non-U.S. Gov't Journal Article Research Support, N.I.H., Extramural
GrantInformation_xml	– fundername: NHLBI NIH HHS grantid: R01 HL134893 – fundername: British Heart Foundation grantid: CH/1996001/9454 – fundername: NHLBI NIH HHS grantid: K24 HL153669 – fundername: NHLBI NIH HHS grantid: R01 HL092577 – fundername: American Heart Association-American Stroke Association grantid: 18SFRN34250007 – fundername: Medical Research Council grantid: MC_PC_17228 – fundername: NHLBI NIH HHS grantid: T32 HL007208 – fundername: NHLBI NIH HHS grantid: R01 HL139731 – fundername: NHLBI NIH HHS grantid: R01 HL157635 – fundername: NHLBI NIH HHS grantid: R01 HL140224
GroupedDBID	--- .-D .3C .XZ .Z2 01R 0R~ 0ZK 18M 1J1 29B 2FS 2WC 354 40H 4Q1 4Q2 4Q3 53G 5GY 5RE 5VS 6PF 71W 77Y 7O~ AAAAV AAAXR AAFWJ AAGIX AAHPQ AAIQE AAJCS AAMOA AAMTA AAQKA AARTV AASCR AASOK AASXQ AAUEB AAWTL AAXQO ABASU ABBUW ABDIG ABJNI ABOCM ABPMR ABPXF ABQRW ABVCZ ABXVJ ABXYN ABZAD ABZZY ACDDN ACDOF ACEWG ACGFO ACGFS ACILI ACLDA ACOAL ACRKK ACWDW ACWRI ACXJB ACXNZ ACZKN ADBBV ADCYY ADGGA ADHPY AE3 AE6 AEBDS AENEX AFBFQ AFCHL AFDTB AFEXH AFMBP AFNMH AFSOK AFUWQ AGINI AHMBA AHOMT AHQNM AHQVU AHRYX AHVBC AIJEX AINUH AJCLO AJIOK AJNWD AJZMW AKCTQ AKULP ALKUP ALMA_UNASSIGNED_HOLDINGS ALMTX AMJPA AMKUR AMNEI AOHHW AOQMC ASPBG AVWKF AYCSE AZFZN BAWUL BOYCO BQLVK BYPQX C45 CGR CS3 CUY CVF DIK DIWNM DU5 E3Z EBS ECM EEVPB EIF ERAAH EX3 F2K F2L F2M F2N F5P FCALG GNXGY GQDEL GX1 H0~ HLJTE HZ~ IKREB IKYAY IN~ IPNFZ JF9 JG8 JK3 K-A K-F K8S KD2 KMI KQ8 L-C L7B N9A NPM N~7 N~B O9- OAG OAH OBH OCB ODMTH OGEVE OHH OHYEH OK1 OL1 OLB OLG OLH OLU OLV OLY OLZ OPUJH OVD OVDNE OVIDH OVLEI OVOZU OWBYB OWU OWV OWW OWX OWY OWZ OXXIT P2P PQQKQ RAH RIG RLZ S4R S4S T8P TEORI TR2 TSPGW UPT V2I VVN W2D W3M W8F WH7 WOQ WOW X3V X3W XXN XYM YFH YOC YSK YYM YZZ ZFV ZY1 ~H1 7X8 ADKSD ADSXY
ID	FETCH-LOGICAL-c4796-be13e2d5129f5dfaeeccb816c2b8bc2c7fe9b94c941d09d4d81446ccb46f0b32
IEDL.DBID	7X8
ISICitedReferencesCount	187
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=00003017-202201110-00006&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1524-4539
IngestDate	Sat Sep 27 19:04:53 EDT 2025 Mon Jul 21 06:03:18 EDT 2025
IsDoiOpenAccess	false
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	2
Keywords	deep learning atrial fibrillation electronic health records
Language	English
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c4796-be13e2d5129f5dfaeeccb816c2b8bc2c7fe9b94c941d09d4d81446ccb46f0b32
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ORCID	0000-0001-6822-0593 0000-0002-2067-0533 0000-0002-7987-4768 0000-0002-3893-2423 0000-0002-2840-4539 0000-0002-0053-2002 0000-0002-9520-0501 0000-0001-9256-0678 0000-0002-9599-4866
OpenAccessLink	https://www.ahajournals.org/doi/pdf/10.1161/CIRCULATIONAHA.121.057480
PMID	34743566
PQID	2595111554
PQPubID	23479
ParticipantIDs	proquest_miscellaneous_2595111554 pubmed_primary_34743566
PublicationCentury	2000
PublicationDate	2022-01-11 20220111
PublicationDateYYYYMMDD	2022-01-11
PublicationDate_xml	– month: 01 year: 2022 text: 2022-01-11 day: 11
PublicationDecade	2020
PublicationPlace	United States
PublicationPlace_xml	– name: United States
PublicationTitle	Circulation (New York, N.Y.)
PublicationTitleAlternate	Circulation
PublicationYear	2022
SSID	ssj0006375
Score	2.6986713
Snippet	Artificial intelligence (AI)-enabled analysis of 12-lead ECGs may facilitate efficient estimation of incident atrial fibrillation (AF) risk. However, it...
SourceID	proquest pubmed
SourceType	Aggregation Database Index Database
StartPage	122
SubjectTerms	Atrial Fibrillation - diagnosis Atrial Fibrillation - pathology Deep Learning - standards Electrocardiography - methods Female Humans Male Middle Aged Risk Factors
Title	ECG-Based Deep Learning and Clinical Risk Factors to Predict Atrial Fibrillation
URI	https://www.ncbi.nlm.nih.gov/pubmed/34743566 https://www.proquest.com/docview/2595111554
Volume	145
WOSCitedRecordID	wos00003017-202201110-00006&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/eLvHCXMwpV1LS8NAEB58Ubz4aH0_2IJ4W-0mm032JLVaK9hSikJvJfuIiJpWWwX_vbObSE-C4CXkkA1hZzLzzWO_ATjRkVGBiA2NdMxc6iaiqcEoJUgiIUWcCa1SP2wi7vWS4VD2y4TbtGyr_LGJ3lCbsXY58nOE6YgNnPe7mLxRNzXKVVfLERqLsBwilHEtXfFwzhYuQk-0iy6KUx6FsgJ1byQEO2_dDloPdwXhbMdlBNkZIheeNH5Hmt7jtNf_-60bsFZiTdIslGMTFmxehVozxzj79YucEt_96dPqVah0yyJ7DfrXrRt6id7NkCtrJ6SkYH0kaW5IySP6QgZP02fSLqb1kNmY9N_dC2ak6eeAkLY7SvBSNNptwX37-r7VoeXgBapRWoIqy0IbGIcFsshkqUU5q4QJHahE6UDHmZVKci05Mw1puElcVInPcJE1VBhsw1I-zu0uEASUgqeRSnE5l8xVGTPjaocWb2Rm9qD-s4Mj1GtXrEhzO_6YjuZ7uAc7hRhGk4KAYxRyxD2IQ_f_sPoAVgN3YqHBKGOHsJzhX22PYEV_zp6m78deYfDa63e_AcJryKM
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=ECG-Based+Deep+Learning+and+Clinical+Risk+Factors+to+Predict+Atrial+Fibrillation&rft.jtitle=Circulation+%28New+York%2C+N.Y.%29&rft.au=Khurshid%2C+Shaan&rft.au=Friedman%2C+Samuel&rft.au=Reeder%2C+Christopher&rft.au=Di+Achille%2C+Paolo&rft.date=2022-01-11&rft.issn=1524-4539&rft.eissn=1524-4539&rft.volume=145&rft.issue=2&rft.spage=122&rft_id=info:doi/10.1161%2FCIRCULATIONAHA.121.057480&rft.externalDBID=NO_FULL_TEXT
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1524-4539&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1524-4539&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1524-4539&client=summon