EUVIMEDEUVIMED entdecken
Open Access

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

Christopher D Barrett iD, Yuto Suzuki iD, Sundos Hussein iD, Lohit Garg, Alexis Tumolo, Amneet Sandhu iD, John J West iD, Matthew Zipse iD, Ryan Aleong, Paul Varosy, Wendy S Tzou iD, Farnoush Banaei-Kashani iD und 1 weitere

Journal of the American Heart Association · 2023

DOI: 10.1161/jaha.122.028483 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the American Heart Association, 2023. Verfasst von Christopher D Barrett und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA. Hauptthemen laut MEDLINE: Atrial Fibrillation. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 64 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NHLBI NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Atrial FibrillationHumansAnti-Arrhythmia AgentsRetrospective StudiesArtificial IntelligenceElectric Countershock

Stichwörter der Autor:innen

Q‐learningartificial intelligenceatrial fibrillationrate controlreinforcement learningrhythm controlunsupervised learning

Substanzen

Anti-Arrhythmia Agents

Autor:innen und Institutionen

  • Department of Cardiac Electrophysiology University of Colorado Anschutz Medical Campus Aurora CO USA.
  • Department of Computer Science University of Colorado Denver CO USA.
  • Department of Cardiac Electrophysiology Rocky Mountain Regional VA Healthcare System Aurora CO USA.
  • Department of Cardiac Electrophysiology Denver Health Medical Center Denver CO USA.

ORCID-Profile: Christopher D Barrett, Yuto Suzuki, Sundos Hussein, Amneet Sandhu, John J West, Matthew Zipse, Wendy S Tzou, Farnoush Banaei-Kashani, Michael A Rosenberg

Förderung

  • NHLBI NIH HHS

Zitieren

APA

Barrett, C. D., Suzuki, Y., Hussein, S., Garg, L., Tumolo, A., Sandhu, A., West, J. J., Zipse, M., Aleong, R., Varosy, P., Tzou, W. S., Banaei-Kashani, F., & Rosenberg, M. A. (2023). Evaluation of Quantitative Decision‐Making for Rhythm Management of Atrial Fibrillation Using Tabular Q‐Learning. Journal of the American Heart Association. https://doi.org/10.1161/jaha.122.028483

Vancouver

Barrett CD, Suzuki Y, Hussein S, Garg L, Tumolo A, Sandhu A, et al. Evaluation of Quantitative Decision‐Making for Rhythm Management of Atrial Fibrillation Using Tabular Q‐Learning. Journal of the American Heart Association. 2023. doi:10.1161/jaha.122.028483

BibTeX
@article{Barrett2023,
  author = {Barrett, Christopher D and Suzuki, Yuto and Hussein, Sundos and Garg, Lohit and Tumolo, Alexis and Sandhu, Amneet and West, John J and Zipse, Matthew and Aleong, Ryan and Varosy, Paul and Tzou, Wendy S and Banaei-Kashani, Farnoush and Rosenberg, Michael A},
  title = {Evaluation of Quantitative Decision‐Making for Rhythm Management of Atrial Fibrillation Using Tabular Q‐Learning},
  journal = {Journal of the American Heart Association},
  year = {2023},
  doi = {10.1161/jaha.122.028483}
}
Kostenlos starten →