Evaluation of Quantitative Decision‐Making for Rhythm Management of Atrial Fibrillation Using Tabular Q‐Learning

Background Rhythm management is a complex decision for patients with atrial fibrillation (AF). Although clinical trials have identified subsets of patients who might benefit from a given rhythm-management strategy, for individual patients it is not always clear which strategy is expected to have the...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Journal of the American Heart Association Ročník 12; číslo 9; s. e028483
Hlavní autoři:	Barrett, Christopher D., Suzuki, Yuto, Hussein, Sundos, Garg, Lohit, Tumolo, Alexis, Sandhu, Amneet, West, John J., Zipse, Matthew, Aleong, Ryan, Varosy, Paul, Tzou, Wendy S., Banaei‐Kashani, Farnoush, Rosenberg, Michael A.
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	England John Wiley and Sons Inc 02.05.2023 Wiley
Témata:	Anti-Arrhythmia Agents - therapeutic use Artificial Intelligence atrial fibrillation Atrial Fibrillation - drug therapy Atrial Fibrillation - therapy Electric Countershock Humans Original Research Q‐learning rate control reinforcement learning Retrospective Studies rhythm control atrial fibrillation Q‐learning rate control artificial intelligence reinforcement learning rhythm control unsupervised learning
ISSN:	2047-9980, 2047-9980
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	Background Rhythm management is a complex decision for patients with atrial fibrillation (AF). Although clinical trials have identified subsets of patients who might benefit from a given rhythm-management strategy, for individual patients it is not always clear which strategy is expected to have the greatest mortality benefit or durability. Methods and Results In this investigation 52 547 patients with a new atrial fibrillation diagnosis between 2010 and 2020 were retrospectively identified. We applied a type of artificial intelligence called tabular Q-learning to identify the optimal initial rhythm-management strategy, based on a composite outcome of mortality, change in treatment, and sustainability of the given treatment, termed the reward function. We first applied an unsupervised learning algorithm using a variational autoencoder with K-means clustering to cluster atrial fibrillation patients into 8 distinct phenotypes. We then fit a Q-learning algorithm to predict the best outcome for each cluster. Although rate-control strategy was most frequently selected by treating providers, the outcome was superior for rhythm-control strategies across all clusters. Subjects in whom provider-selected treatment matched the Q-table recommendation had fewer total deaths (4 [8.5%] versus 473 [22.4%], odds ratio=0.32, =0.02) and a greater reward ( =4.8×10 ). We then demonstrated application of dynamic learning by updating the Q-table prospectively using batch gradient descent, in which the optimal strategy in some clusters changed from cardioversion to ablation. Conclusions Tabular Q-learning provides a dynamic and interpretable approach to apply artificial intelligence to clinical decision-making for atrial fibrillation. Further work is needed to examine application of Q-learning prospectively in clinical patients.
AbstractList	Background Rhythm management is a complex decision for patients with atrial fibrillation (AF). Although clinical trials have identified subsets of patients who might benefit from a given rhythm-management strategy, for individual patients it is not always clear which strategy is expected to have the greatest mortality benefit or durability. Methods and Results In this investigation 52 547 patients with a new atrial fibrillation diagnosis between 2010 and 2020 were retrospectively identified. We applied a type of artificial intelligence called tabular Q-learning to identify the optimal initial rhythm-management strategy, based on a composite outcome of mortality, change in treatment, and sustainability of the given treatment, termed the reward function. We first applied an unsupervised learning algorithm using a variational autoencoder with K-means clustering to cluster atrial fibrillation patients into 8 distinct phenotypes. We then fit a Q-learning algorithm to predict the best outcome for each cluster. Although rate-control strategy was most frequently selected by treating providers, the outcome was superior for rhythm-control strategies across all clusters. Subjects in whom provider-selected treatment matched the Q-table recommendation had fewer total deaths (4 [8.5%] versus 473 [22.4%], odds ratio=0.32, =0.02) and a greater reward ( =4.8×10 ). We then demonstrated application of dynamic learning by updating the Q-table prospectively using batch gradient descent, in which the optimal strategy in some clusters changed from cardioversion to ablation. Conclusions Tabular Q-learning provides a dynamic and interpretable approach to apply artificial intelligence to clinical decision-making for atrial fibrillation. Further work is needed to examine application of Q-learning prospectively in clinical patients. Background Rhythm management is a complex decision for patients with atrial fibrillation (AF). Although clinical trials have identified subsets of patients who might benefit from a given rhythm-management strategy, for individual patients it is not always clear which strategy is expected to have the greatest mortality benefit or durability. Methods and Results In this investigation 52 547 patients with a new atrial fibrillation diagnosis between 2010 and 2020 were retrospectively identified. We applied a type of artificial intelligence called tabular Q-learning to identify the optimal initial rhythm-management strategy, based on a composite outcome of mortality, change in treatment, and sustainability of the given treatment, termed the reward function. We first applied an unsupervised learning algorithm using a variational autoencoder with K-means clustering to cluster atrial fibrillation patients into 8 distinct phenotypes. We then fit a Q-learning algorithm to predict the best outcome for each cluster. Although rate-control strategy was most frequently selected by treating providers, the outcome was superior for rhythm-control strategies across all clusters. Subjects in whom provider-selected treatment matched the Q-table recommendation had fewer total deaths (4 [8.5%] versus 473 [22.4%], odds ratio=0.32, P=0.02) and a greater reward (P=4.8×10-6). We then demonstrated application of dynamic learning by updating the Q-table prospectively using batch gradient descent, in which the optimal strategy in some clusters changed from cardioversion to ablation. Conclusions Tabular Q-learning provides a dynamic and interpretable approach to apply artificial intelligence to clinical decision-making for atrial fibrillation. Further work is needed to examine application of Q-learning prospectively in clinical patients.Background Rhythm management is a complex decision for patients with atrial fibrillation (AF). Although clinical trials have identified subsets of patients who might benefit from a given rhythm-management strategy, for individual patients it is not always clear which strategy is expected to have the greatest mortality benefit or durability. Methods and Results In this investigation 52 547 patients with a new atrial fibrillation diagnosis between 2010 and 2020 were retrospectively identified. We applied a type of artificial intelligence called tabular Q-learning to identify the optimal initial rhythm-management strategy, based on a composite outcome of mortality, change in treatment, and sustainability of the given treatment, termed the reward function. We first applied an unsupervised learning algorithm using a variational autoencoder with K-means clustering to cluster atrial fibrillation patients into 8 distinct phenotypes. We then fit a Q-learning algorithm to predict the best outcome for each cluster. Although rate-control strategy was most frequently selected by treating providers, the outcome was superior for rhythm-control strategies across all clusters. Subjects in whom provider-selected treatment matched the Q-table recommendation had fewer total deaths (4 [8.5%] versus 473 [22.4%], odds ratio=0.32, P=0.02) and a greater reward (P=4.8×10-6). We then demonstrated application of dynamic learning by updating the Q-table prospectively using batch gradient descent, in which the optimal strategy in some clusters changed from cardioversion to ablation. Conclusions Tabular Q-learning provides a dynamic and interpretable approach to apply artificial intelligence to clinical decision-making for atrial fibrillation. Further work is needed to examine application of Q-learning prospectively in clinical patients. Background Rhythm management is a complex decision for patients with atrial fibrillation (AF). Although clinical trials have identified subsets of patients who might benefit from a given rhythm‐management strategy, for individual patients it is not always clear which strategy is expected to have the greatest mortality benefit or durability. Methods and Results In this investigation 52 547 patients with a new atrial fibrillation diagnosis between 2010 and 2020 were retrospectively identified. We applied a type of artificial intelligence called tabular Q‐learning to identify the optimal initial rhythm‐management strategy, based on a composite outcome of mortality, change in treatment, and sustainability of the given treatment, termed the reward function. We first applied an unsupervised learning algorithm using a variational autoencoder with K‐means clustering to cluster atrial fibrillation patients into 8 distinct phenotypes. We then fit a Q‐learning algorithm to predict the best outcome for each cluster. Although rate‐control strategy was most frequently selected by treating providers, the outcome was superior for rhythm‐control strategies across all clusters. Subjects in whom provider‐selected treatment matched the Q‐table recommendation had fewer total deaths (4 [8.5%] versus 473 [22.4%], odds ratio=0.32, P=0.02) and a greater reward (P=4.8×10−6). We then demonstrated application of dynamic learning by updating the Q‐table prospectively using batch gradient descent, in which the optimal strategy in some clusters changed from cardioversion to ablation. Conclusions Tabular Q‐learning provides a dynamic and interpretable approach to apply artificial intelligence to clinical decision‐making for atrial fibrillation. Further work is needed to examine application of Q‐learning prospectively in clinical patients.
Author	Banaei‐Kashani, Farnoush Tumolo, Alexis Varosy, Paul Aleong, Ryan Garg, Lohit Hussein, Sundos Zipse, Matthew Suzuki, Yuto Sandhu, Amneet West, John J. Barrett, Christopher D. Tzou, Wendy S. Rosenberg, Michael A.
AuthorAffiliation	4 Department of Cardiac Electrophysiology Denver Health Medical Center Denver CO USA 3 Department of Cardiac Electrophysiology Rocky Mountain Regional VA Healthcare System Aurora CO USA 2 Department of Computer Science University of Colorado Denver CO USA 1 Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA
AuthorAffiliation_xml	– name: 4 Department of Cardiac Electrophysiology Denver Health Medical Center Denver CO USA – name: 1 Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA – name: 2 Department of Computer Science University of Colorado Denver CO USA – name: 3 Department of Cardiac Electrophysiology Rocky Mountain Regional VA Healthcare System Aurora CO USA
Author_xml	– sequence: 1 givenname: Christopher D. orcidid: 0000-0002-8073-5773 surname: Barrett fullname: Barrett, Christopher D. organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA – sequence: 2 givenname: Yuto orcidid: 0000-0001-5018-7596 surname: Suzuki fullname: Suzuki, Yuto organization: Department of Computer Science University of Colorado Denver CO USA – sequence: 3 givenname: Sundos orcidid: 0000-0002-5994-7480 surname: Hussein fullname: Hussein, Sundos organization: Department of Computer Science University of Colorado Denver CO USA – sequence: 4 givenname: Lohit surname: Garg fullname: Garg, Lohit organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA – sequence: 5 givenname: Alexis surname: Tumolo fullname: Tumolo, Alexis organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA – sequence: 6 givenname: Amneet orcidid: 0000-0002-6435-5746 surname: Sandhu fullname: Sandhu, Amneet organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA, Department of Cardiac Electrophysiology Rocky Mountain Regional VA Healthcare System Aurora CO USA – sequence: 7 givenname: John J. orcidid: 0000-0003-1937-4582 surname: West fullname: West, John J. organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA, Department of Cardiac Electrophysiology Denver Health Medical Center Denver CO USA – sequence: 8 givenname: Matthew orcidid: 0000-0002-5662-8505 surname: Zipse fullname: Zipse, Matthew organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA – sequence: 9 givenname: Ryan surname: Aleong fullname: Aleong, Ryan organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA – sequence: 10 givenname: Paul surname: Varosy fullname: Varosy, Paul organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA, Department of Cardiac Electrophysiology Denver Health Medical Center Denver CO USA – sequence: 11 givenname: Wendy S. orcidid: 0000-0002-8514-1165 surname: Tzou fullname: Tzou, Wendy S. organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA – sequence: 12 givenname: Farnoush orcidid: 0000-0003-4102-9873 surname: Banaei‐Kashani fullname: Banaei‐Kashani, Farnoush organization: Department of Computer Science University of Colorado Denver CO USA – sequence: 13 givenname: Michael A. orcidid: 0000-0002-6708-1648 surname: Rosenberg fullname: Rosenberg, Michael A. organization: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/37119087$$D View this record in MEDLINE/PubMed
BookMark	eNp1kstu1DAUhiNURC90zQ5lyWamvsSOvUKj0tKiqVDRsLZOHGfGJWMX2xmpOx6BZ-yT4DQtapHwwse3_zu2z39Y7DnvTFG8w2iOMccnXxYXizkmZI6IqAR9VRwQVNUzKQXaezbeL45jvEG5cVJTJt8U-7TGWCJRHxTpbAf9AMl6V_quvB7AJZvyfGfKT0bbmDfuf_2-gh_WrcvOh_Lb5i5ttuUVOFibrXFp1C1SsNCX57YJtu8n3Pc4SlbQDD2E8jpTlgaCy4tvi9cd9NEcP8ajYnV-tjq9mC2_fr48XSxnuuIozUwrOiEbke8sGaNNC7LWjNCOcSIrknvDqxpagnVLKlN3uBI5Ms4BU4PoUXE5YVsPN-o22C2EO-XBqocFH9YKQrK6Nwo3-Qc1pVQwVAHnsgUmDWu55G0jkc6sjxPrdmi2ptX53QH6F9CXO85u1NrvFEaE1ITgTPjwSAj-52BiUlsbtcm_5YwfoiIC1RKLWo4Xf_882d8sT2XLB9h0QAcfYzCd0g9F82Nu2-ekanSIGh2iskPU5JCsO_lH94T-n-IPhgm-7A
CitedBy_id	crossref_primary_10_1038_s41746_025_01509_1 crossref_primary_10_3390_healthcare13141752 crossref_primary_10_1002_widm_1530 crossref_primary_10_1007_s11831_024_10100_y crossref_primary_10_1093_eurheartj_ehaf474
Cites_doi	10.1161/01.cir.84.2.527 10.2196/medinform.3445 10.1016/j.jacc.2018.04.058 10.1017/S1047951119000416 10.1161/CIRCULATIONAHA.121.056323 10.1093/eurheartj/ehz782 10.1253/circj.72.705 10.1161/CIRCEP.108.837294 10.1056/NEJMoa021375 10.1056/NEJM199211123272002 10.1161/CIRCULATIONAHA.112.092494 10.1161/CIRCOUTCOMES.111.962688 10.1016/j.neunet.2021.11.026 10.48550/arXiv.1606.03490 10.1056/NEJMoa2019422 10.1016/S0140-6736(13)62343-0 10.1016/S0735-1097(03)00332-2 10.1056/NEJM199103213241201 10.1016/0735-1097(91)90585-W 10.1161/CIRCOUTCOMES.110.958165 10.1161/JAHA.118.011560 10.1161/CIRCULATIONAHA.121.053733 10.2196/29225 10.1093/europace/euab200 10.1161/CIRCULATIONAHA.112.000491 10.1161/CIRCULATIONAHA.121.057095 10.1016/j.ahj.2004.11.030 10.1038/s41586-021-03819-2 10.2196/42163 10.1161/CIRCEP.109.878355 10.1016/j.ahj.2020.01.011 10.3389/fdgth.2021.645232 10.1016/S0140-6736(89)91200-2 10.2196/17162 10.1001/jama.288.19.2441 10.1056/NEJMoa0708789 10.1016/j.ahj.2021.08.007 10.1016/j.ahj.2010.06.034 10.1038/nature16961 10.1093/eurheartj/ehz085 10.1093/eurheartj/ehp235 10.2174/157340312803760730 10.1056/NEJMoa0905561 10.1161/CIRCEP.110.958033 10.1056/NEJMoa1009638 10.1056/NEJMoa1007432 10.1161/CIRCOUTCOMES.118.005358 10.1161/01.STR.0000185927.63746.23 10.1111/j.1540-8167.2011.02035.x 10.1161/CIRCULATIONAHA.111.069450 10.1056/NEJMoa021328 10.1001/archinte.166.12.1269 10.1056/NEJMoa1707855 10.1056/NEJM199011293232201 10.1177/2050312118759444 10.1001/archinte.1994.00420130036007 10.1001/jama.2019.0693 10.4022/jafib.1422 10.1161/01.CIR.0000104567.72914.BF 10.1056/NEJMoa0803778 10.7326/0003-4819-146-12-200706190-00007
ContentType	Journal Article
Copyright	2023 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley.
Copyright_xml	– notice: 2023 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley.
DBID	AAYXX CITATION CGR CUY CVF ECM EIF NPM 7X8 5PM DOA
DOI	10.1161/JAHA.122.028483
DatabaseName	CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed MEDLINE - Academic PubMed Central (Full Participant titles) DOAJ Directory of Open Access Journals
DatabaseTitle	CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic
DatabaseTitleList	MEDLINE MEDLINE - Academic
Database_xml	– sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 3 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Medicine
DocumentTitleAlternate	Artificial Intelligence for Atrial Fibrillation
EISSN	2047-9980
ExternalDocumentID	oai_doaj_org_article_1b848c3338504a669da59e5d696db90c PMC10227221 37119087 10_1161_JAHA_122_028483
Genre	Journal Article
GrantInformation_xml	– fundername: NHLBI NIH HHS grantid: R01 HL146824
GroupedDBID	0R~ 1OC 53G 5VS 8-1 AAMMB AAYXX AAZKR ACGFO ACXQS ADBBV ADKYN ADZMN AEFGJ AEGXH AENEX AGXDD AIAGR AIDQK AIDYY ALAGY ALMA_UNASSIGNED_HOLDINGS ALUQN AOIJS AVUZU BAWUL BCNDV CITATION DIK EBS EMOBN GODZA GROUPED_DOAJ GX1 HYE KQ8 M48 M~E OK1 RAH RNS RPM WIN 24P ACCMX CGR CUY CVF ECM EIF NPM 7X8 5PM
ID	FETCH-LOGICAL-c460t-ed8f89b83599553bda97c523f562942562e647ad21cd24e7f14824e566a13e03
IEDL.DBID	DOA
ISICitedReferencesCount	5
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000992550200005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	2047-9980
IngestDate	Mon Nov 10 04:30:37 EST 2025 Tue Nov 04 02:06:42 EST 2025 Fri Sep 05 14:38:58 EDT 2025 Mon Jul 21 06:00:57 EDT 2025 Tue Nov 18 21:43:52 EST 2025 Sat Nov 29 05:05:07 EST 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	9
Keywords	atrial fibrillation Q‐learning rate control artificial intelligence reinforcement learning rhythm control unsupervised learning
Language	English
License	This is an open access article under the terms of the http://creativecommons.org/licenses/by-nc-nd/4.0/ License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made.
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c460t-ed8f89b83599553bda97c523f562942562e647ad21cd24e7f14824e566a13e03
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 This article was sent to Wei‐Qi Wei, MD, PhD, Guest Editor, for review by expert referees, editorial decision, and final disposition. For Sources of Funding and Disclosures, see page 13. Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.122.028483
ORCID	0000-0002-5662-8505 0000-0002-5994-7480 0000-0003-1937-4582 0000-0002-6708-1648 0000-0002-6435-5746 0000-0002-8514-1165 0000-0003-4102-9873 0000-0002-8073-5773 0000-0001-5018-7596
OpenAccessLink	https://doaj.org/article/1b848c3338504a669da59e5d696db90c
PMID	37119087
PQID	2807918790
PQPubID	23479
ParticipantIDs	doaj_primary_oai_doaj_org_article_1b848c3338504a669da59e5d696db90c pubmedcentral_primary_oai_pubmedcentral_nih_gov_10227221 proquest_miscellaneous_2807918790 pubmed_primary_37119087 crossref_citationtrail_10_1161_JAHA_122_028483 crossref_primary_10_1161_JAHA_122_028483
PublicationCentury	2000
PublicationDate	2023-05-02
PublicationDateYYYYMMDD	2023-05-02
PublicationDate_xml	– month: 05 year: 2023 text: 2023-05-02 day: 02
PublicationDecade	2020
PublicationPlace	England
PublicationPlace_xml	– name: England – name: Hoboken
PublicationTitle	Journal of the American Heart Association
PublicationTitleAlternate	J Am Heart Assoc
PublicationYear	2023
Publisher	John Wiley and Sons Inc Wiley
Publisher_xml	– name: John Wiley and Sons Inc – name: Wiley
References	e_1_3_1_60_2 e_1_3_1_43_2 e_1_3_1_22_2 e_1_3_1_45_2 e_1_3_1_24_2 e_1_3_1_8_2 e_1_3_1_62_2 e_1_3_1_41_2 e_1_3_1_64_2 e_1_3_1_20_2 e_1_3_1_4_2 e_1_3_1_6_2 e_1_3_1_26_2 e_1_3_1_47_2 e_1_3_1_2_2 e_1_3_1_28_2 e_1_3_1_49_2 e_1_3_1_32_2 e_1_3_1_55_2 e_1_3_1_34_2 e_1_3_1_57_2 e_1_3_1_13_2 e_1_3_1_51_2 e_1_3_1_11_2 e_1_3_1_30_2 e_1_3_1_53_2 e_1_3_1_17_2 e_1_3_1_15_2 e_1_3_1_36_2 e_1_3_1_59_2 e_1_3_1_19_2 e_1_3_1_38_2 (e_1_3_1_9_2) 1994; 343 e_1_3_1_21_2 e_1_3_1_44_2 e_1_3_1_65_2 e_1_3_1_23_2 e_1_3_1_46_2 e_1_3_1_7_2 e_1_3_1_61_2 e_1_3_1_42_2 e_1_3_1_63_2 e_1_3_1_29_2 e_1_3_1_3_2 e_1_3_1_5_2 e_1_3_1_25_2 e_1_3_1_48_2 e_1_3_1_27_2 e_1_3_1_33_2 e_1_3_1_54_2 e_1_3_1_35_2 e_1_3_1_56_2 e_1_3_1_12_2 e_1_3_1_50_2 e_1_3_1_10_2 e_1_3_1_31_2 e_1_3_1_52_2 e_1_3_1_16_2 e_1_3_1_14_2 e_1_3_1_37_2 e_1_3_1_58_2 Lundberg SM (e_1_3_1_40_2) 2017 e_1_3_1_18_2 e_1_3_1_39_2
References_xml	– ident: e_1_3_1_8_2 doi: 10.1161/01.cir.84.2.527 – ident: e_1_3_1_38_2 doi: 10.2196/medinform.3445 – ident: e_1_3_1_22_2 doi: 10.1016/j.jacc.2018.04.058 – ident: e_1_3_1_59_2 doi: 10.1017/S1047951119000416 – ident: e_1_3_1_43_2 doi: 10.1161/CIRCULATIONAHA.121.056323 – ident: e_1_3_1_21_2 doi: 10.1093/eurheartj/ehz782 – ident: e_1_3_1_42_2 doi: 10.1253/circj.72.705 – ident: e_1_3_1_58_2 doi: 10.1161/CIRCEP.108.837294 – ident: e_1_3_1_25_2 doi: 10.1056/NEJMoa021375 – ident: e_1_3_1_11_2 doi: 10.1056/NEJM199211123272002 – volume: 343 start-page: 687 year: 1994 ident: e_1_3_1_9_2 article-title: Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study publication-title: Lancet – ident: e_1_3_1_30_2 doi: 10.1161/CIRCULATIONAHA.112.092494 – volume-title: Advances in Neural Information Processing Systems year: 2017 ident: e_1_3_1_40_2 – ident: e_1_3_1_3_2 doi: 10.1161/CIRCOUTCOMES.111.962688 – ident: e_1_3_1_44_2 doi: 10.1016/j.neunet.2021.11.026 – ident: e_1_3_1_46_2 doi: 10.48550/arXiv.1606.03490 – ident: e_1_3_1_34_2 doi: 10.1056/NEJMoa2019422 – ident: e_1_3_1_23_2 doi: 10.1016/S0140-6736(13)62343-0 – ident: e_1_3_1_27_2 doi: 10.1016/S0735-1097(03)00332-2 – ident: e_1_3_1_61_2 doi: 10.1056/NEJM199103213241201 – ident: e_1_3_1_12_2 doi: 10.1016/0735-1097(91)90585-W – ident: e_1_3_1_4_2 doi: 10.1161/CIRCOUTCOMES.110.958165 – ident: e_1_3_1_41_2 doi: 10.1161/JAHA.118.011560 – ident: e_1_3_1_47_2 doi: 10.1161/CIRCULATIONAHA.121.053733 – ident: e_1_3_1_36_2 doi: 10.2196/29225 – ident: e_1_3_1_35_2 doi: 10.1093/europace/euab200 – ident: e_1_3_1_20_2 doi: 10.1161/CIRCULATIONAHA.112.000491 – ident: e_1_3_1_55_2 doi: 10.1161/CIRCULATIONAHA.121.057095 – ident: e_1_3_1_62_2 doi: 10.1016/j.ahj.2004.11.030 – ident: e_1_3_1_49_2 doi: 10.1038/s41586-021-03819-2 – ident: e_1_3_1_45_2 doi: 10.2196/42163 – ident: e_1_3_1_64_2 doi: 10.1161/CIRCEP.109.878355 – ident: e_1_3_1_53_2 doi: 10.1016/j.ahj.2020.01.011 – ident: e_1_3_1_37_2 doi: 10.3389/fdgth.2021.645232 – ident: e_1_3_1_10_2 doi: 10.1016/S0140-6736(89)91200-2 – ident: e_1_3_1_52_2 doi: 10.2196/17162 – ident: e_1_3_1_15_2 doi: 10.1001/jama.288.19.2441 – ident: e_1_3_1_26_2 doi: 10.1056/NEJMoa0708789 – ident: e_1_3_1_51_2 doi: 10.1016/j.ahj.2021.08.007 – ident: e_1_3_1_57_2 doi: 10.1016/j.ahj.2010.06.034 – ident: e_1_3_1_48_2 doi: 10.1038/nature16961 – ident: e_1_3_1_32_2 doi: 10.1093/eurheartj/ehz085 – ident: e_1_3_1_33_2 doi: 10.1093/eurheartj/ehp235 – ident: e_1_3_1_56_2 doi: 10.2174/157340312803760730 – ident: e_1_3_1_17_2 doi: 10.1056/NEJMoa0905561 – ident: e_1_3_1_50_2 doi: 10.1161/CIRCEP.110.958033 – ident: e_1_3_1_18_2 doi: 10.1056/NEJMoa1009638 – ident: e_1_3_1_19_2 doi: 10.1056/NEJMoa1007432 – ident: e_1_3_1_54_2 doi: 10.1161/CIRCOUTCOMES.118.005358 – ident: e_1_3_1_2_2 doi: 10.1161/01.STR.0000185927.63746.23 – ident: e_1_3_1_31_2 doi: 10.1111/j.1540-8167.2011.02035.x – ident: e_1_3_1_63_2 doi: 10.1161/CIRCULATIONAHA.111.069450 – ident: e_1_3_1_28_2 doi: 10.1056/NEJMoa021328 – ident: e_1_3_1_16_2 doi: 10.1001/archinte.166.12.1269 – ident: e_1_3_1_29_2 doi: 10.1056/NEJMoa1707855 – ident: e_1_3_1_7_2 doi: 10.1056/NEJM199011293232201 – ident: e_1_3_1_5_2 doi: 10.1177/2050312118759444 – ident: e_1_3_1_13_2 doi: 10.1001/archinte.1994.00420130036007 – ident: e_1_3_1_6_2 doi: 10.1001/jama.2019.0693 – ident: e_1_3_1_65_2 doi: 10.4022/jafib.1422 – ident: e_1_3_1_60_2 doi: 10.1161/01.CIR.0000104567.72914.BF – ident: e_1_3_1_24_2 doi: 10.1056/NEJMoa0803778 – ident: e_1_3_1_39_2 – ident: e_1_3_1_14_2 doi: 10.7326/0003-4819-146-12-200706190-00007
SSID	ssj0000627359
Score	2.362755
Snippet	Background Rhythm management is a complex decision for patients with atrial fibrillation (AF). Although clinical trials have identified subsets of patients who...
SourceID	doaj pubmedcentral proquest pubmed crossref
SourceType	Open Website Open Access Repository Aggregation Database Index Database Enrichment Source
StartPage	e028483
SubjectTerms	Anti-Arrhythmia Agents - therapeutic use Artificial Intelligence atrial fibrillation Atrial Fibrillation - drug therapy Atrial Fibrillation - therapy Electric Countershock Humans Original Research Q‐learning rate control reinforcement learning Retrospective Studies rhythm control
Title	Evaluation of Quantitative Decision‐Making for Rhythm Management of Atrial Fibrillation Using Tabular Q‐Learning
URI	https://www.ncbi.nlm.nih.gov/pubmed/37119087 https://www.proquest.com/docview/2807918790 https://pubmed.ncbi.nlm.nih.gov/PMC10227221 https://doaj.org/article/1b848c3338504a669da59e5d696db90c
Volume	12
WOSCitedRecordID	wos000992550200005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 2047-9980 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000627359 issn: 2047-9980 databaseCode: DOA dateStart: 20120101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources (ISSN International Center) customDbUrl: eissn: 2047-9980 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000627359 issn: 2047-9980 databaseCode: M~E dateStart: 20120101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre – providerCode: PRVWIB databaseName: Wiley Online Library Free Content customDbUrl: eissn: 2047-9980 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000627359 issn: 2047-9980 databaseCode: WIN dateStart: 20120101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1Nb9QwEB1BhRAXVL5DoTISBy5pE3_7uMCuKqStKNrD3izHcdhKbRa1WSQuVX8Cv5FfwthJl10E4sIlh8ROLM_YfhOP3wN47QI1gQWWG9WwnEt0Y6eCz4taG6EbyoNxSWxCHR_r-dx83JD6ijlhPT1w33GHZaW59gwjKVFwJ6WpnTBB1NLIujKFj7NvocxGMNXPwbgsCzNw-SCqOfwwOoo__ugBLqhcs61lKLH1_wli_p4pubH0THbh_oAZyahv6wO4FdqHcHc67Io_gm68puwmy4acrFybjo7hREbeDxo6P66_T5PwFEGUSj4tvnWLc_Ir9yXWGyUFDzKJhwDO-hQ5kjIKyMxVMVuVnOBbBkLWz49hNhnP3h3lg5pC7rksujzUutGmQsRljBCsqp1RHsPQBhGQwZEraZBcuZqWvkYTqSYyhPKAcM-VLBTsCey0yzY8A8K0YrTwlOva8yZUFXeh4VpIp0UQvMzg4KZvrR-YxqPgxZlNEYcsbTSGRWPY3hgZvFlX-NKTbPy96NtorHWxyI6dbqDP2MFn7L98JoNXN6a2OJriFolrw3J1aSM3kIkC7EUGT3vTrz_FVInoSasM9JZTbLVl-0l7ukiM3TGsVpSWz_9H6_fgXtS8T1mX9AXsdBer8BLu-K_d6eXFPtxWc72fRgNep1fjn_m-EA0
linkProvider	Directory of Open Access Journals
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=Evaluation+of+Quantitative+Decision%E2%80%90Making+for+Rhythm+Management+of+Atrial+Fibrillation+Using+Tabular+Q%E2%80%90Learning&rft.jtitle=Journal+of+the+American+Heart+Association&rft.au=Barrett%2C+Christopher+D.&rft.au=Suzuki%2C+Yuto&rft.au=Hussein%2C+Sundos&rft.au=Garg%2C+Lohit&rft.date=2023-05-02&rft.pub=John+Wiley+and+Sons+Inc&rft.eissn=2047-9980&rft.volume=12&rft.issue=9&rft_id=info:doi/10.1161%2FJAHA.122.028483&rft_id=info%3Apmid%2F37119087&rft.externalDocID=PMC10227221
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2047-9980&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2047-9980&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2047-9980&client=summon