EUVIMEDEUVIMED entdecken
Crossref

Non-invasive coronary fractional flow reserve prediction using a neural network with hemodynamic and geometric embeddings: A proof-of-concept study

Yang Yang, Bao Li, Luyao Fan, Chuanqi Wen, Na Liu, Hao Sun, Leo Hwa Liang, Yili Feng, Tongna Wang, Jincheng Liu, Liyuan Zhang, Jian Liu und 2 weitere

Computer Methods and Programs in Biomedicine · 2026-10 · 285 · 109529

DOI: 10.1016/j.cmpb.2026.109529 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Methods and Programs in Biomedicine, 2026-10. Verfasst von Yang Yang und 13 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Engineering, College of Chemistry and Life Science, Beijing University of Technology, Beijing, China. Hauptthemen laut MEDLINE: Neural Networks, Computer, Fractional Flow Reserve, Myocardial, Hemodynamics. Die Arbeit nennt 46 Literaturangaben. Gefördert von: Key Technologies Research and Development Program, Beijing Natural Science Foundation, National Key Research and Development Program of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND OBJECTIVES Invasive fractional flow reserve (FFR) is the clinical gold standard for assessing coronary artery stenosis, but its application is limited by its invasive nature. While CFD-based FFRCT offers a non-invasive alternative, its high computational cost restricts real-time use. Deep learning has emerged as a promising solution. However, many existing methods rely primarily on geometric data and neglect personalized physiological boundary conditions. These limitations hinder the efficient characterization of complex coronary hemodynamics. To address these challenges, this study proposes a method for the rapid, patient-specific prediction of coronary hemodynamics by integrating personalized boundary conditions with geometric information. METHODS In this study, we constructed a patient-specific steady-state coronary flow field dataset (288 patients). A dual-embedding neural network was proposed, which integrates personalized hemodynamic constraints with coronary geometry to enhance predictive accuracy. The network employs dual encoders to separately extract features from personalized hemodynamic boundary conditions and reduced-dimensional coronary geometry. By integrating these encoded representations into a Bi-LSTM architecture, the model learns the relationship between vascular topology and hemodynamic parameters. The predicted FFRBGE values were validated against both CFD-based FFRCT and invasive FFR to assess numerical agreement and diagnostic performance. RESULTS On the independent test set, the model achieved an RMSE of 7.42% compared to CFD-based FFRCT. When validated against invasive FFR, the FFRBGE showed a correlation of 0.87 and an AUC of 0.901. The inference time per case was significantly reduced compared to traditional CFD. CONCLUSIONS The dual-embedding neural network achieves high consistency with both CFD-based FFRCT and invasive FFR. By encoding patient-specific coronary structures and boundary conditions, this approach provides a proof of concept that FFR distributions can be computed in real time, even with a relatively small CFD training dataset. These findings demonstrate a robust, high-efficiency solution with significant potential for broader application in diverse cardiovascular scenarios.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerFractional Flow Reserve, MyocardialHemodynamicsHumansProof of Concept StudyCoronary StenosisDeep LearningCoronary VesselsAlgorithms

Stichwörter der Autor:innen

Computational fluid dynamicsFractional flow reserveHemodynamicsMyocardial ischemiaNeural network

Autor:innen und Institutionen

  • Department of Biomedical Engineering, College of Chemistry and Life Science, Beijing University of Technology, Beijing, China.
  • Intelligent Manufacturing and Equipment Technology Professional Group, Beijing Polytechnic University, Beijing, China.
  • BioFluid Mechanics Lab, Department of Biomedical Engineering, National University of Singapore, Singapore.
  • Department of Medical Engineering, Beijing Friendship Hospital, Capital Medical University, Beijing, China.
  • Center for Medical Metrology, National institute of metrology, Beijing, China.
  • Department of Cardiology, Peking University People's Hospital, Beijing, China.

Förderung

  • Key Technologies Research and Development Program
  • Beijing Natural Science Foundation
  • National Key Research and Development Program of China
  • Zhejiang Province Basic Public Welfare Research Program
  • National Natural Science Foundation of China

Zitieren

APA

Yang, Y., Li, B., Fan, L., Wen, C., Liu, N., Sun, H., Liang, L. H., Feng, Y., Wang, T., Liu, J., Zhang, L., Liu, J., Wang, L., & Liu, Y. (2026). Non-invasive coronary fractional flow reserve prediction using a neural network with hemodynamic and geometric embeddings: A proof-of-concept study. Computer Methods and Programs in Biomedicine, 285, 109529. https://doi.org/10.1016/j.cmpb.2026.109529

Vancouver

Yang Y, Li B, Fan L, Wen C, Liu N, Sun H, et al. Non-invasive coronary fractional flow reserve prediction using a neural network with hemodynamic and geometric embeddings: A proof-of-concept study. Computer Methods and Programs in Biomedicine. 2026;285:109529. doi:10.1016/j.cmpb.2026.109529

BibTeX
@article{Yang2026,
  author = {Yang, Yang and Li, Bao and Fan, Luyao and Wen, Chuanqi and Liu, Na and Sun, Hao and Liang, Leo Hwa and Feng, Yili and Wang, Tongna and Liu, Jincheng and Zhang, Liyuan and Liu, Jian and Wang, Lihua and Liu, Youjun},
  title = {Non-invasive coronary fractional flow reserve prediction using a neural network with hemodynamic and geometric embeddings: A proof-of-concept study},
  journal = {Computer Methods and Programs in Biomedicine},
  year = {2026},
  volume = {285},
  pages = {109529},
  doi = {10.1016/j.cmpb.2026.109529}
}
Kostenlos starten →