Automated stenosis estimation of coronary angiographies using end-to-end learning.

Journal: The international journal of cardiovascular imaging
PMID: 39789341

Abstract

The initial evaluation of stenosis during coronary angiography is typically performed by visual assessment. Visual assessment has limited accuracy compared to fractional flow reserve and quantitative coronary angiography, which are more time-consuming and costly. Applying deep learning might yield a faster and more accurate stenosis assessment. We developed a deep learning model to classify cine loops into left or right coronary artery (LCA/RCA) or "other". Data were obtained by manual annotation. Using these classifications, cine loops before revascularization were identified and curated automatically. Separate deep learning models for LCA and RCA were developed to estimate stenosis using these identified cine loops. From a cohort of 19,414 patients and 332,582 cine loops, we identified cine loops for 13,480 patients for model development and 5056 for internal testing. External testing was conducted using automated identified cine loops from 608 patients. For identification of significant stenosis (visual assessment of diameter stenosis > 70%), our model obtained a receiver operator characteristic (ROC) area under the curve (ROC-AUC) of 0.903 (95% CI: 0.900-0.906) on the internal test. The performance was evaluated on the external test set against visual assessment, 3D quantitative coronary angiography, and fractional flow reserve (≤ 0.80), obtaining ROC AUC values of 0.833 (95% CI: 0.814-0.852), 0.798 (95% CI: 0.741-0.842), and 0.780 (95% CI: 0.743-0.817), respectively. The deep-learning-based stenosis estimation models showed promising results for predicting stenosis. Compared to previous work, our approach demonstrates performance increase, includes all 16 segments, does not exclude revascularized patients, is externally tested, and is simpler using fewer steps.

Authors

  • Christian Kim Eschen
    Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.
  • Karina Banasik
    Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.
  • Anders Bjorholm Dahl
    Section for Visual Computing, Department of Applied Mathematics and Computer Science, Technical University of Denmark, Lyngby, Denmark.
  • Piotr Jaroslaw Chmura
    Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark.
  • Peter Bruun-Rasmussen
    Department of Clinical Immunology, Faculty of Health and Medical Sciences, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark.
  • Frants Pedersen
    Department of Cardiology, Faculty of Health and Medical Sciences, Rigshospitalet, The Heart Center, University of Copenhagen, Copenhagen, Denmark.
  • Lars Køber
    From the Danish Heart Foundation, Copenhagen, Denmark (P.W.H., T.S.G.S., E.L.F., G.H.G.); DTU Compute, Technical University of Denmark, Lyngby (L.C.); The Heart Centre, Rigshospitalet (E.L.F., L.K.), and Department of Clinical Medicine (G.H.G.), University of Copenhagen, Denmark; Institute of Health, Science and Technology, Aalborg University, Denmark (C.T.-P.); The National Institute of Public Health, University of Southern Denmark, Copenhagen (G.H.G.); University of Copenhagen, Denmark; and Department of Medicine, Section of Cardiology, Glostrup Hospital, University of Copenhagen, Denmark (C.A.).
  • Thomas Engstrøm
    Department of Cardiology, Faculty of Health and Medical Sciences, Rigshospitalet, The Heart Center, University of Copenhagen, Copenhagen, Denmark.
  • Morten Bøttcher
    Department of Cardiology, Gødstrup Hospital, Herning, Denmark.
  • Simon Winther
    Department of Cardiology, Gødstrup Hospital, Herning, Denmark.
  • Alex Hørby Christensen
    Department of Cardiology, Faculty of Health and Medical Sciences, Rigshospitalet, The Heart Center, University of Copenhagen, Copenhagen, Denmark.
  • Henning Bundgaard
    Department of Clinical Medicine, University of Copenhagen, Blegdamsvej 3B, 2200 Copenhagen N, Denmark.
  • Søren Brunak
    NNF Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark.

Keywords

External Resources

Popular Topics
Recent Journals