EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access · CC BY 4.0

Large Language Model‐Based Localization of Premature Ventricular Contraction Origins: A Retrospective Diagnostic Accuracy Study

Kazuhisa Matsumoto iD, Yuya Fujisaki, Syunta Higuchi iD, Masataka Narita iD, Wataru Sasaki iD, Tsukasa Naganuma iD, Naomichi Tanaka, Daisuke Kawano iD, Kei Matsumoto, Wataru Yoshino, Yoshifumi Ikeda iD, Ritsushi Kato iD und 1 weitere

Journal of Cardiovascular Electrophysiology · 2026

DOI: 10.1111/jce.70464 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift Journal of Cardiovascular Electrophysiology, 2026. Verfasst von Kazuhisa Matsumoto und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Cardiology, Saitama Medical University, International Medical Center, Hidaka, Saitama, Japan. Hauptthemen laut MEDLINE: Ventricular Premature Complexes, Large Language Models, Electrocardiography, Heart Rate. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hidaka Research Project: 07-D-1-03.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Accurate localization of premature ventricular contraction (PVC) origin from 12‐lead electrocardiography (ECG) is important for procedural planning in catheter ablation. Although convolutional neural network (CNN)‐based models have shown promising diagnostic performance, they require task‐specific training and remain limited in interpretability. We evaluated whether large language model (LLM)‐based ECG image interpretation could perform binary left‐versus‐right PVC origin localization from 12‐lead ECG images while providing a traceable diagnostic process.

Methods

We retrospectively studied 157 patients who underwent successful catheter ablation of PVCs or idiopathic ventricular tachycardia. ECG images were classified as RIGHT‐origin ( n = 103) or LEFT‐origin ( n = 54) according to the final successful ablation site. Three approaches were compared: a CNN‐based baseline model, an LLM‐based one‐shot approach, and an LLM‐based staged extraction framework with deterministic rule‐based integration. Performance was evaluated across five independent seeds using positive predictive value (PPV), negative predictive value (NPV), recall, and PPV + NPV.

Results

The CNN‐based baseline model achieved a PPV of 0.546 ± 0.058 and an NPV of 0.793 ± 0.047, with area under the curve values ranging from 0.595 to 0.730. The staged extraction framework generated a continuous rule‐based score with discrimination numerically comparable to the CNN‐based model (AUC 0.720 ± 0.045 vs. 0.712 ± 0.054). A stricter threshold, determined from training data, increased PPV at the expense of recall and warrants prospective external validation of this operating point.

Conclusions

LLM‐based staged extraction demonstrated the potential to achieve binary left‐versus‐right PVC origin localization from 12‐lead ECG images while providing a traceable stepwise diagnostic process. The continuous rule‐based score showed discrimination numerically comparable to the CNN‐based model, and the strict threshold—determined using training data—identified a potential high‐PPV operating point that requires prospective external validation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Ventricular Premature ComplexesLarge Language ModelsElectrocardiographyHeart RateSignal Processing, Computer-AssistedTachycardia, VentricularHumansRetrospective StudiesPredictive Value of TestsReproducibility of ResultsConvolutional Neural NetworksFemaleMaleAction PotentialsCatheter AblationMiddle AgedTreatment Outcome

Stichwörter der Autor:innen

artificial intelligencedeep learningelectrocardiogramlarge language modelpremature ventricular contraction

Autor:innen und Institutionen

  • Department of Cardiology, Saitama Medical University, International Medical Center, Hidaka, Saitama, Japan.
  • Sera, Inc, Meguro, Tokyo, Japan.
  • Department of Clinical Engineer, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan.

ORCID-Profile: Kazuhisa Matsumoto, Syunta Higuchi, Masataka Narita, Wataru Sasaki, Tsukasa Naganuma, Daisuke Kawano, Yoshifumi Ikeda, Ritsushi Kato, Hitoshi Mori

Förderung

  • Hidaka Research Project: 07-D-1-03

Zitieren

APA

Matsumoto, K., Fujisaki, Y., Higuchi, S., Narita, M., Sasaki, W., Naganuma, T., Tanaka, N., Kawano, D., Matsumoto, K., Yoshino, W., Ikeda, Y., Kato, R., & Mori, H. (2026). Large Language Model‐Based Localization of Premature Ventricular Contraction Origins: A Retrospective Diagnostic Accuracy Study. Journal of Cardiovascular Electrophysiology. https://doi.org/10.1111/jce.70464

Vancouver

Matsumoto K, Fujisaki Y, Higuchi S, Narita M, Sasaki W, Naganuma T, et al. Large Language Model‐Based Localization of Premature Ventricular Contraction Origins: A Retrospective Diagnostic Accuracy Study. Journal of Cardiovascular Electrophysiology. 2026. doi:10.1111/jce.70464

BibTeX
@article{Matsumoto2026,
  author = {Matsumoto, Kazuhisa and Fujisaki, Yuya and Higuchi, Syunta and Narita, Masataka and Sasaki, Wataru and Naganuma, Tsukasa and Tanaka, Naomichi and Kawano, Daisuke and Matsumoto, Kei and Yoshino, Wataru and Ikeda, Yoshifumi and Kato, Ritsushi and Mori, Hitoshi},
  title = {Large Language Model‐Based Localization of Premature Ventricular Contraction Origins: A Retrospective Diagnostic Accuracy Study},
  journal = {Journal of Cardiovascular Electrophysiology},
  year = {2026},
  doi = {10.1111/jce.70464}
}
Kostenlos starten →