Deep learning-based stenosis quantification from coronary CT angiography

Youngtaek Hong, Frederic Commandeur, Sebastien Cadet, Markus Goeller, Mhairi K. Doris, Xi Chen, Jacek Kwiecinski, Daniel S. Berman, Piotr J. Slomka, Hyuk Jae Chang, Damini Dey

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Coronary computed tomography angiography (CTA) allows quantification of stenosis. However, such quantitative analysis is not part of clinical routine. We evaluated the feasibility of utilizing deep learning for quantifying coronary artery disease from CTA. Methods: A total of 716 diseased segments in 156 patients (66 ± 10 years) who underwent CTA were analyzed. Minimal luminal area (MLA), percent diameter stenosis (DS), and percent contrast density difference (CDD) were measured using semi-automated software (Autoplaque) by an expert reader. Using the expert annotations, deep learning was performed with convolutional neural networks using 10-fold cross-validation to segment CTA lumen and calcified plaque. MLA, DS and CDD computed using deep-learning-based approach was compared to expert reader measurements. Results: There was excellent correlation between the expert reader and deep learning for all quantitative measures (r=0.984 for MLA; r=0.957 for DS; and r=0.975 for CDD, p<0.001 for all). The expert reader and deep learning method was not significantly different for MLA (median 4.3 mm2 for both, p=0.68) and CDD (11.6 vs 11.1%, p=0.30), and was significantly different for DS (26.0 vs 26.6%, p<0.05); however, the ranges of all the quantitative measures were within inter-observer variability between 2 expert readers. Conclusions: Our deep learning-based method allows quantitative measurement of coronary artery disease segments accurately from CTA and may enhance clinical reporting.

Original languageEnglish
Title of host publicationMedical Imaging 2019
Subtitle of host publicationImage Processing
EditorsElsa D. Angelini, Elsa D. Angelini, Elsa D. Angelini, Bennett A. Landman
PublisherSPIE
ISBN (Electronic)9781510625457
DOIs
Publication statusPublished - 2019 Jan 1
EventMedical Imaging 2019: Image Processing - San Diego, United States
Duration: 2019 Feb 192019 Feb 21

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume10949
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2019: Image Processing
CountryUnited States
CitySan Diego
Period19/2/1919/2/21

Fingerprint

Angiography
angiography
Coronary Angiography
learning
readers
Pathologic Constriction
Tomography
Learning
tomography
coronary artery disease
Coronary Artery Disease
annotations
lumens
Observer Variation
quantitative analysis
Computed Tomography Angiography
Deep learning
Software
Neural networks
computer programs

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging

Cite this

Hong, Y., Commandeur, F., Cadet, S., Goeller, M., Doris, M. K., Chen, X., ... Dey, D. (2019). Deep learning-based stenosis quantification from coronary CT angiography. In E. D. Angelini, E. D. Angelini, E. D. Angelini, & B. A. Landman (Eds.), Medical Imaging 2019: Image Processing [109492I] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10949). SPIE. https://doi.org/10.1117/12.2512168
Hong, Youngtaek ; Commandeur, Frederic ; Cadet, Sebastien ; Goeller, Markus ; Doris, Mhairi K. ; Chen, Xi ; Kwiecinski, Jacek ; Berman, Daniel S. ; Slomka, Piotr J. ; Chang, Hyuk Jae ; Dey, Damini. / Deep learning-based stenosis quantification from coronary CT angiography. Medical Imaging 2019: Image Processing. editor / Elsa D. Angelini ; Elsa D. Angelini ; Elsa D. Angelini ; Bennett A. Landman. SPIE, 2019. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).
@inproceedings{bab9f281e9f34ed0a3fb8c79d12ef682,
title = "Deep learning-based stenosis quantification from coronary CT angiography",
abstract = "Coronary computed tomography angiography (CTA) allows quantification of stenosis. However, such quantitative analysis is not part of clinical routine. We evaluated the feasibility of utilizing deep learning for quantifying coronary artery disease from CTA. Methods: A total of 716 diseased segments in 156 patients (66 ± 10 years) who underwent CTA were analyzed. Minimal luminal area (MLA), percent diameter stenosis (DS), and percent contrast density difference (CDD) were measured using semi-automated software (Autoplaque) by an expert reader. Using the expert annotations, deep learning was performed with convolutional neural networks using 10-fold cross-validation to segment CTA lumen and calcified plaque. MLA, DS and CDD computed using deep-learning-based approach was compared to expert reader measurements. Results: There was excellent correlation between the expert reader and deep learning for all quantitative measures (r=0.984 for MLA; r=0.957 for DS; and r=0.975 for CDD, p<0.001 for all). The expert reader and deep learning method was not significantly different for MLA (median 4.3 mm2 for both, p=0.68) and CDD (11.6 vs 11.1{\%}, p=0.30), and was significantly different for DS (26.0 vs 26.6{\%}, p<0.05); however, the ranges of all the quantitative measures were within inter-observer variability between 2 expert readers. Conclusions: Our deep learning-based method allows quantitative measurement of coronary artery disease segments accurately from CTA and may enhance clinical reporting.",
author = "Youngtaek Hong and Frederic Commandeur and Sebastien Cadet and Markus Goeller and Doris, {Mhairi K.} and Xi Chen and Jacek Kwiecinski and Berman, {Daniel S.} and Slomka, {Piotr J.} and Chang, {Hyuk Jae} and Damini Dey",
year = "2019",
month = "1",
day = "1",
doi = "10.1117/12.2512168",
language = "English",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
publisher = "SPIE",
editor = "Angelini, {Elsa D.} and Angelini, {Elsa D.} and Angelini, {Elsa D.} and Landman, {Bennett A.}",
booktitle = "Medical Imaging 2019",
address = "United States",

}

Hong, Y, Commandeur, F, Cadet, S, Goeller, M, Doris, MK, Chen, X, Kwiecinski, J, Berman, DS, Slomka, PJ, Chang, HJ & Dey, D 2019, Deep learning-based stenosis quantification from coronary CT angiography. in ED Angelini, ED Angelini, ED Angelini & BA Landman (eds), Medical Imaging 2019: Image Processing., 109492I, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 10949, SPIE, Medical Imaging 2019: Image Processing, San Diego, United States, 19/2/19. https://doi.org/10.1117/12.2512168

Deep learning-based stenosis quantification from coronary CT angiography. / Hong, Youngtaek; Commandeur, Frederic; Cadet, Sebastien; Goeller, Markus; Doris, Mhairi K.; Chen, Xi; Kwiecinski, Jacek; Berman, Daniel S.; Slomka, Piotr J.; Chang, Hyuk Jae; Dey, Damini.

Medical Imaging 2019: Image Processing. ed. / Elsa D. Angelini; Elsa D. Angelini; Elsa D. Angelini; Bennett A. Landman. SPIE, 2019. 109492I (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10949).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Deep learning-based stenosis quantification from coronary CT angiography

AU - Hong, Youngtaek

AU - Commandeur, Frederic

AU - Cadet, Sebastien

AU - Goeller, Markus

AU - Doris, Mhairi K.

AU - Chen, Xi

AU - Kwiecinski, Jacek

AU - Berman, Daniel S.

AU - Slomka, Piotr J.

AU - Chang, Hyuk Jae

AU - Dey, Damini

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Coronary computed tomography angiography (CTA) allows quantification of stenosis. However, such quantitative analysis is not part of clinical routine. We evaluated the feasibility of utilizing deep learning for quantifying coronary artery disease from CTA. Methods: A total of 716 diseased segments in 156 patients (66 ± 10 years) who underwent CTA were analyzed. Minimal luminal area (MLA), percent diameter stenosis (DS), and percent contrast density difference (CDD) were measured using semi-automated software (Autoplaque) by an expert reader. Using the expert annotations, deep learning was performed with convolutional neural networks using 10-fold cross-validation to segment CTA lumen and calcified plaque. MLA, DS and CDD computed using deep-learning-based approach was compared to expert reader measurements. Results: There was excellent correlation between the expert reader and deep learning for all quantitative measures (r=0.984 for MLA; r=0.957 for DS; and r=0.975 for CDD, p<0.001 for all). The expert reader and deep learning method was not significantly different for MLA (median 4.3 mm2 for both, p=0.68) and CDD (11.6 vs 11.1%, p=0.30), and was significantly different for DS (26.0 vs 26.6%, p<0.05); however, the ranges of all the quantitative measures were within inter-observer variability between 2 expert readers. Conclusions: Our deep learning-based method allows quantitative measurement of coronary artery disease segments accurately from CTA and may enhance clinical reporting.

AB - Coronary computed tomography angiography (CTA) allows quantification of stenosis. However, such quantitative analysis is not part of clinical routine. We evaluated the feasibility of utilizing deep learning for quantifying coronary artery disease from CTA. Methods: A total of 716 diseased segments in 156 patients (66 ± 10 years) who underwent CTA were analyzed. Minimal luminal area (MLA), percent diameter stenosis (DS), and percent contrast density difference (CDD) were measured using semi-automated software (Autoplaque) by an expert reader. Using the expert annotations, deep learning was performed with convolutional neural networks using 10-fold cross-validation to segment CTA lumen and calcified plaque. MLA, DS and CDD computed using deep-learning-based approach was compared to expert reader measurements. Results: There was excellent correlation between the expert reader and deep learning for all quantitative measures (r=0.984 for MLA; r=0.957 for DS; and r=0.975 for CDD, p<0.001 for all). The expert reader and deep learning method was not significantly different for MLA (median 4.3 mm2 for both, p=0.68) and CDD (11.6 vs 11.1%, p=0.30), and was significantly different for DS (26.0 vs 26.6%, p<0.05); however, the ranges of all the quantitative measures were within inter-observer variability between 2 expert readers. Conclusions: Our deep learning-based method allows quantitative measurement of coronary artery disease segments accurately from CTA and may enhance clinical reporting.

UR - http://www.scopus.com/inward/record.url?scp=85068309283&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068309283&partnerID=8YFLogxK

U2 - 10.1117/12.2512168

DO - 10.1117/12.2512168

M3 - Conference contribution

AN - SCOPUS:85068309283

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2019

A2 - Angelini, Elsa D.

A2 - Angelini, Elsa D.

A2 - Angelini, Elsa D.

A2 - Landman, Bennett A.

PB - SPIE

ER -

Hong Y, Commandeur F, Cadet S, Goeller M, Doris MK, Chen X et al. Deep learning-based stenosis quantification from coronary CT angiography. In Angelini ED, Angelini ED, Angelini ED, Landman BA, editors, Medical Imaging 2019: Image Processing. SPIE. 2019. 109492I. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). https://doi.org/10.1117/12.2512168