Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Deep learning prediction of quantitative coronary angiography values using myocardial perfusion images with a CZT camera

Arvidsson, Ida LU ; Davidsson, Anette ; Overgaard, Niels Christian LU ; Pagonis, Christos ; Åström, Kalle LU orcid ; Good, Elin ; Frias-Rose, Jeronimo ; Heyden, Anders LU orcid and Ochoa-Figueroa, Miguel (2023) In Journal of Nuclear Cardiology 30(1). p.116-126
Abstract

Purpose: Evaluate the prediction of quantitative coronary angiography (QCA) values from MPI, by means of deep learning. Methods: 546 patients (67% men) undergoing stress 99mTc-tetrofosmin MPI in a CZT camera in the upright and supine position were included (1092 MPIs). Patients were divided into two groups: ICA group included 271 patients who performed an ICA within 6 months of MPI and a control group with 275 patients with low pre-test probability for CAD and a normal MPI. QCA analyses were performed using radiologic software and verified by an expert reader. Left ventricular myocardium was segmented using clinical nuclear cardiology software and verified by an expert reader. A deep learning model was trained using a double... (More)

Purpose: Evaluate the prediction of quantitative coronary angiography (QCA) values from MPI, by means of deep learning. Methods: 546 patients (67% men) undergoing stress 99mTc-tetrofosmin MPI in a CZT camera in the upright and supine position were included (1092 MPIs). Patients were divided into two groups: ICA group included 271 patients who performed an ICA within 6 months of MPI and a control group with 275 patients with low pre-test probability for CAD and a normal MPI. QCA analyses were performed using radiologic software and verified by an expert reader. Left ventricular myocardium was segmented using clinical nuclear cardiology software and verified by an expert reader. A deep learning model was trained using a double cross-validation scheme such that all data could be used as test data as well. Results: Area under the receiver-operating characteristic curve for the prediction of QCA, with > 50% narrowing of the artery, by deep learning for the external test cohort: per patient 85% [95% confidence interval (CI) 84%-87%] and per vessel; LAD 74% (CI 72%-76%), RCA 85% (CI 83%-86%), LCx 81% (CI 78%-84%), and average 80% (CI 77%-83%). Conclusion: Deep learning can predict the presence of different QCA percentages of coronary artery stenosis from MPIs.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Artificial intelligence, cadmium-zinc-telluride, coronary angiography, deep learning, myocardial scintigraphy
in
Journal of Nuclear Cardiology
volume
30
issue
1
pages
116 - 126
publisher
Springer
external identifiers
  • pmid:35610536
  • scopus:85130744085
ISSN
1071-3581
DOI
10.1007/s12350-022-02995-6
project
Deep learning based evaluation of coronary artery disease and estimation of quantitative coronary angiography using myocardial perfusion imaging
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2022, The Author(s) under exclusive licence to American Society of Nuclear Cardiology.
id
e2236d6a-ef8c-4075-b0bd-864c99f68cec
date added to LUP
2022-08-19 13:44:24
date last changed
2024-04-18 05:08:49
@article{e2236d6a-ef8c-4075-b0bd-864c99f68cec,
  abstract     = {{<p>Purpose: Evaluate the prediction of quantitative coronary angiography (QCA) values from MPI, by means of deep learning. Methods: 546 patients (67% men) undergoing stress 99mTc-tetrofosmin MPI in a CZT camera in the upright and supine position were included (1092 MPIs). Patients were divided into two groups: ICA group included 271 patients who performed an ICA within 6 months of MPI and a control group with 275 patients with low pre-test probability for CAD and a normal MPI. QCA analyses were performed using radiologic software and verified by an expert reader. Left ventricular myocardium was segmented using clinical nuclear cardiology software and verified by an expert reader. A deep learning model was trained using a double cross-validation scheme such that all data could be used as test data as well. Results: Area under the receiver-operating characteristic curve for the prediction of QCA, with &gt; 50% narrowing of the artery, by deep learning for the external test cohort: per patient 85% [95% confidence interval (CI) 84%-87%] and per vessel; LAD 74% (CI 72%-76%), RCA 85% (CI 83%-86%), LCx 81% (CI 78%-84%), and average 80% (CI 77%-83%). Conclusion: Deep learning can predict the presence of different QCA percentages of coronary artery stenosis from MPIs.</p>}},
  author       = {{Arvidsson, Ida and Davidsson, Anette and Overgaard, Niels Christian and Pagonis, Christos and Åström, Kalle and Good, Elin and Frias-Rose, Jeronimo and Heyden, Anders and Ochoa-Figueroa, Miguel}},
  issn         = {{1071-3581}},
  keywords     = {{Artificial intelligence; cadmium-zinc-telluride; coronary angiography; deep learning; myocardial scintigraphy}},
  language     = {{eng}},
  number       = {{1}},
  pages        = {{116--126}},
  publisher    = {{Springer}},
  series       = {{Journal of Nuclear Cardiology}},
  title        = {{Deep learning prediction of quantitative coronary angiography values using myocardial perfusion images with a CZT camera}},
  url          = {{http://dx.doi.org/10.1007/s12350-022-02995-6}},
  doi          = {{10.1007/s12350-022-02995-6}},
  volume       = {{30}},
  year         = {{2023}},
}