EUVIMEDEUVIMED entdecken
Multizentrische StudieOpen Access · CC BY 4.0

Automated Deep Learning Pipeline for Characterizing Left Ventricular Diastolic Function

Victoria Yuan, Yuki Sahashi, Hirotaka Ieki, Teodora Szasz, Miloš Vukadinovic, Christina Binder, Karima Addetia, Martin Gruca, Konrad Pieszko, Andrew P Ambrosy, Jeremy A Slivnick, Paul P Cheng und 2 weitere

Journal of Cardiac Failure · 2026-11 · 32 (11) · 1878-1888

DOI: 10.1016/j.cardfail.2026.08.031 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift Journal of Cardiac Failure, 2026-11. Verfasst von Victoria Yuan und 13 weiteren Autor:innen. Beteiligte Einrichtung: Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, CA; David Geffen School of Medicine, University of Califor…. Hauptthemen laut MEDLINE: Deep Learning, Ventricular Dysfunction, Left, Ventricular Function, Left, Echocardiography. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Institutes of Health, Sarnoff Cardiovascular Research Foundation Inc, American Heart Association Inc.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

INTRODUCTION: Left ventricular diastolic dysfunction (LVDD) is most commonly evaluated by echocardiography. However, without a sole identifying metric, LVDD is assessed by a diagnostic algorithm relying on secondary characteristics that is laborious and has the potential for interobserver variability. METHODS: To characterize concordance in clinical evaluations of LVDD, we evaluated historical echocardiogram studies at 2 academic medical centers for variability between clinical diastology reported by clinicians and the 2016 and 2025 American Society of Echocardiography (ASE) guidelines. We developed a workflow of 8 deep learning (DL) models trained on over 155,000 studies to automate LVDD assessment. Model performance was evaluated on held-out test sets from 3 academic centers. RESULTS: In 592 studies from Cedars-Sinai Medical Center (CSMC), DL demonstrated 80.4% agreement and a weighted Cohen kappa of 0.62 with the 2016 ASE guideline assessment. In contrast, the CSMC clinician evaluation had 52.2% agreement and a kappa of 0.46 with the 2016 guidelines. In the Stanford Healthcare (SHC) cohort of 1572 studies, DL had 66.7% agreement and a kappa of 0.33 with the 2016 guidelines vs clinical diastology, with 32.7% agreement and a kappa of 0.14. Our DL workflow showed strong concordance with the 2025 guidelines, with agreements of 81.8% in CSMC and 68.1% in SHC. In a third validation cohort of 204 studies from the University of Chicago, DL had 34.3% agreement with the 2025 guidelines. CONCLUSIONS: Evaluating LVDD has high heterogeneity. We developed DL measurement-driven diastolic dysfunction, a pipeline of models that automates the clinical workflow of grading LVDD and can contribute to improved diagnosis of heart failure.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningVentricular Dysfunction, LeftVentricular Function, LeftEchocardiographyHumansDiastole

Stichwörter der Autor:innen

Deep learning, echocardiography, diastolic function

Autor:innen und Institutionen

  • Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, CA; David Geffen School of Medicine, University of California, Los Angeles, CA.
  • Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, CA.
  • Division of Cardiology, Department of Medicine, Stanford University, Palo Alto, CA.
  • Philips Healthcare, Cambridge, MA.
  • Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, CA; Department of Bioengineering, University of California, Los Angeles, CA.
  • Division of Cardiology, University of Chicago Medicine, Chicago IL.

Förderung

  • National Institutes of Health
  • Sarnoff Cardiovascular Research Foundation Inc
  • American Heart Association Inc

Zitieren

APA

Yuan, V., Sahashi, Y., Ieki, H., Szasz, T., Vukadinovic, M., Binder, C., Addetia, K., Gruca, M., Pieszko, K., Ambrosy, A. P., Slivnick, J. A., Cheng, P. P., Cheng, S., & Ouyang, D. (2026). Automated Deep Learning Pipeline for Characterizing Left Ventricular Diastolic Function. Journal of Cardiac Failure, 32(11), 1878-1888. https://doi.org/10.1016/j.cardfail.2026.08.031

Vancouver

Yuan V, Sahashi Y, Ieki H, Szasz T, Vukadinovic M, Binder C, et al. Automated Deep Learning Pipeline for Characterizing Left Ventricular Diastolic Function. Journal of Cardiac Failure. 2026;32(11):1878-1888. doi:10.1016/j.cardfail.2026.08.031

BibTeX
@article{Yuan2026,
  author = {Yuan, Victoria and Sahashi, Yuki and Ieki, Hirotaka and Szasz, Teodora and Vukadinovic, Miloš and Binder, Christina and Addetia, Karima and Gruca, Martin and Pieszko, Konrad and Ambrosy, Andrew P and Slivnick, Jeremy A and Cheng, Paul P and Cheng, Susan and Ouyang, David},
  title = {Automated Deep Learning Pipeline for Characterizing Left Ventricular Diastolic Function},
  journal = {Journal of Cardiac Failure},
  year = {2026},
  volume = {32},
  number = {11},
  pages = {1878--1888},
  doi = {10.1016/j.cardfail.2026.08.031}
}
Kostenlos starten →