Evaluating convolutional neural network-enhanced electrocardiography for hypertrophic cardiomyopathy detection in a specialized cardiovascular setting
Heart and Vessels · 2024
DOI: 10.1007/s00380-024-02367-9 ↗Kurz gefasst
Fachartikel in der Zeitschrift Heart and Vessels, 2024. Verfasst von Naomi Hirota und 23 weiteren Autor:innen. Beteiligte Einrichtung: Department of Cardiovascular Medicine, The Cardiovascular Institute, 3-2-19 Nishiazabu, Minato-Ku, Tokyo, 106-0031, Japan. n-hirota@cvi.or.jp. Hauptthemen laut MEDLINE: Cardiomyopathy, Hypertrophic, Electrocardiography, Neural Networks, Computer. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 31 Literaturangaben.
Abstract (Originalsprache)
The efficacy of convolutional neural network (CNN)-enhanced electrocardiography (ECG) in detecting hypertrophic cardiomyopathy (HCM) and dilated HCM (dHCM) remains uncertain in real-world applications. This retrospective study analyzed data from 19,170 patients (including 140 HCM or dHCM) in the Shinken Database (2010-2017). We evaluated the sensitivity, positive predictive rate (PPR), and F1 score of CNN-enhanced ECG in a ''basic diagnosis'' model (total disease label) and a ''comprehensive diagnosis'' model (including disease subtypes). Using all-lead ECG in the "basic diagnosis" model, we observed a sensitivity of 76%, PPR of 2.9%, and F1 score of 0.056. These metrics improved in cases with a diagnostic probability of ≥ 0.9 and left ventricular hypertrophy (LVH) on ECG: 100% sensitivity, 8.6% PPR, and 0.158 F1 score. The ''comprehensive diagnosis'' model further enhanced these figures to 100%, 13.0%, and 0.230, respectively. Performance was broadly consistent across CNN models using different lead configurations, particularly when including leads viewing the lateral walls. While the precision of CNN models in detecting HCM or dHCM in real-world settings is initially low, it improves by targeting specific patient groups and integrating disease subtype models. The use of ECGs with fewer leads, especially those involving the lateral walls, appears comparably effective.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Cardiovascular Medicine, The Cardiovascular Institute, 3-2-19 Nishiazabu, Minato-Ku, Tokyo, 106-0031, Japan. n-hirota@cvi.or.jp.
- Department of Cardiovascular Medicine, The Cardiovascular Institute, 3-2-19 Nishiazabu, Minato-Ku, Tokyo, 106-0031, Japan.
- Nihon Kohden Corporation, Tokyo, Japan.
- Information System Division, The Cardiovascular Institute, Tokyo, Japan.
- Department of Cardiovascular Surgery, The Cardiovascular Institute, Tokyo, Japan.
Zitieren
APA
Hirota, N., Suzuki, S., Motogi, J., Umemoto, T., Nakai, H., Matsuzawa, W., Takayanagi, T., Hyodo, A., Satoh, K., Arita, T., Yagi, N., Kishi, M., Semba, H., Kano, H., Matsuno, S., Kato, Y., Otsuka, T., Uejima, T., Oikawa, Y., …, & Yamashita, T. (2024). Evaluating convolutional neural network-enhanced electrocardiography for hypertrophic cardiomyopathy detection in a specialized cardiovascular setting. Heart and Vessels. https://doi.org/10.1007/s00380-024-02367-9
Vancouver
Hirota N, Suzuki S, Motogi J, Umemoto T, Nakai H, Matsuzawa W, et al. Evaluating convolutional neural network-enhanced electrocardiography for hypertrophic cardiomyopathy detection in a specialized cardiovascular setting. Heart and Vessels. 2024. doi:10.1007/s00380-024-02367-9
BibTeX
@article{Hirota2024,
author = {Hirota, Naomi and Suzuki, Shinya and Motogi, Jun and Umemoto, Takuya and Nakai, Hiroshi and Matsuzawa, Wataru and Takayanagi, Tsuneo and Hyodo, Akira and Satoh, Keiichi and Arita, Takuto and Yagi, Naoharu and Kishi, Mikio and Semba, Hiroaki and Kano, Hiroto and Matsuno, Shunsuke and Kato, Yuko and Otsuka, Takayuki and Uejima, Tokuhisa and Oikawa, Yuji and Hori, Takayuki and Matsuhama, Minoru and Iida, Mitsuru and Yajima, Junji and Yamashita, Takeshi},
title = {Evaluating convolutional neural network-enhanced electrocardiography for hypertrophic cardiomyopathy detection in a specialized cardiovascular setting},
journal = {Heart and Vessels},
year = {2024},
doi = {10.1007/s00380-024-02367-9}
}
EUVIMED