EUVIMEDEUVIMED entdecken
Crossref

Robust prediction of parameterized cardiovascular hemodynamics using deep operator networks with time normalization

Junki Hong, Bomi Lee, Adelle Ria Persad, Hyunwoo Song, Changhee Min, Jae-Hak Jeong, Alireza Doostan, Yong-Hwa Park

Computer Methods and Programs in Biomedicine · 2026-06 · 280 · 109311

DOI: 10.1016/j.cmpb.2026.109311 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Methods and Programs in Biomedicine, 2026-06. Verfasst von Junki Hong und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon, 34141, Republic…. Hauptthemen laut MEDLINE: Hemodynamics, Models, Cardiovascular, Neural Networks, Computer. Die Arbeit nennt 52 Literaturangaben. Gefördert von: National Research Foundation of Korea, Korea Ministry of Science and ICT.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND OBJECTIVES Computational modeling of cardiovascular hemodynamics is essential for understanding disease mechanisms. While one-dimensional (1D) numerical solvers are widely used, their cumulative computational cost becomes prohibitive in many-query scenarios - such as global sensitivity analysis and Bayesian parameter estimation - that require thousands of iterative evaluations. This study presents a parameterized Deep Operator Network (DeepONet) framework integrated with time normalization to enable rapid and robust surrogate modeling. METHODS The proposed framework employs a two-step time normalization strategy comprising cycle and period normalization. A parameterized trunk network explicitly encodes heart rate, arterial stiffness, and relative arterial length to compensate for temporal scaling and facilitate explicit parameterization. The method was evaluated on simulated 1D cardiovascular cases using a testing protocol that included out-of-distribution scenarios with parameters extending significantly beyond the training range. RESULTS The model achieved high accuracy within the training distribution, with a root mean squared error (RMSE) below 0.5 mmHg. In out-of-distribution regimes where standard models showed performance degradation (RMSE >20 mmHg), the proposed framework maintained relatively robust performance (mean RMSE

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

HemodynamicsModels, CardiovascularNeural Networks, ComputerComputer SimulationHumansAlgorithmsBayes TheoremHeart Rate

Stichwörter der Autor:innen

Cardiovascular hemodynamicsDeep operator networksParameterized modelingPeriodic signalsSurrogate modelingTime normalization

Autor:innen und Institutionen

  • Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon, 34141, Republic of Korea; Ann and H.J. Smead Aerospace Engineering Sciences, Universi…
  • Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon, 34141, Republic of Korea.
  • SimPlatform Co. Ltd., 226, Gasan digital 1-ro, Geumcheon-gu, Seoul, 08502, Republic of Korea.
  • Department of Electronic Engineering, Korea National University of Transportation, Chungju-si, 27469, Chungcheongbuk-do, Republic of Korea.
  • Ann and H.J. Smead Aerospace Engineering Sciences, University of Colorado Boulder, Boulder, 80303, CO, United States of America.
  • Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon, 34141, Republic of Korea. Electronic address: yhpark@kaist.ac.kr.

Förderung

  • National Research Foundation of Korea
  • Korea Ministry of Science and ICT

Zitieren

APA

Hong, J., Lee, B., Persad, A. R., Song, H., Min, C., Jeong, J. H., Doostan, A., & Park, Y. H. (2026). Robust prediction of parameterized cardiovascular hemodynamics using deep operator networks with time normalization. Computer Methods and Programs in Biomedicine, 280, 109311. https://doi.org/10.1016/j.cmpb.2026.109311

Vancouver

Hong J, Lee B, Persad AR, Song H, Min C, Jeong JH, et al. Robust prediction of parameterized cardiovascular hemodynamics using deep operator networks with time normalization. Computer Methods and Programs in Biomedicine. 2026;280:109311. doi:10.1016/j.cmpb.2026.109311

BibTeX
@article{Hong2026,
  author = {Hong, Junki and Lee, Bomi and Persad, Adelle Ria and Song, Hyunwoo and Min, Changhee and Jeong, Jae-Hak and Doostan, Alireza and Park, Yong-Hwa},
  title = {Robust prediction of parameterized cardiovascular hemodynamics using deep operator networks with time normalization},
  journal = {Computer Methods and Programs in Biomedicine},
  year = {2026},
  volume = {280},
  pages = {109311},
  doi = {10.1016/j.cmpb.2026.109311}
}
Kostenlos starten →