AI Evaluation of Stenosis on Coronary CTA, Comparison With Quantitative Coronary Angiography and Fractional Flow Reserve: A CREDENCE Trial Substudy

William F. Griffin; Andrew D. Choi; Joanna S. Riess; Hugo Marques; Hyuk Jae Chang; Jung Hyun Choi; Joon Hyung Doh; Ae Young Her; Bon Kwon Koo; Chang Wook Nam; Hyung Bok Park; Sang Hoon Shin; Jason Cole; Alessia Gimelli; Muhammad Akram Khan; Bin Lu; Yang Gao; Faisal Nabi; Ryo Nakazato; U. Joseph Schoepf; Roel S. Driessen; Michiel J. Bom; Randall Thompson; James J. Jang; Michael Ridner; Chris Rowan; Erick Avelar; Philippe Généreux; Paul Knaapen; Guus A. de Waard; Gianluca Pontone; Daniele Andreini; James P. Earls

doi:10.1016/j.jcmg.2021.10.020

AI Evaluation of Stenosis on Coronary CTA, Comparison With Quantitative Coronary Angiography and Fractional Flow Reserve: A CREDENCE Trial Substudy

William F. Griffin, Andrew D. Choi, Joanna S. Riess, Hugo Marques, Hyuk Jae Chang, Jung Hyun Choi, Joon Hyung Doh, Ae Young Her, Bon Kwon Koo, Chang Wook Nam, Hyung Bok Park, Sang Hoon Shin, Jason Cole, Alessia Gimelli, Muhammad Akram Khan, Bin Lu, Yang Gao, Faisal Nabi, Ryo Nakazato, U. Joseph SchoepfRoel S. Driessen, Michiel J. Bom, Randall Thompson, James J. Jang, Michael Ridner, Chris Rowan, Erick Avelar, Philippe Généreux, Paul Knaapen, Guus A. de Waard, Gianluca Pontone, Daniele Andreini, James P. Earls

Research output: Contribution to journal › Article › peer-review

40 Scopus citations

Abstract

BACKGROUND: Clinical reads of coronary computed tomography angiography (CTA), especially by less experienced readers, may result in overestimation of coronary artery disease stenosis severity compared with expert interpretation. Artificial intelligence (AI)-based solutions applied to coronary CTA may overcome these limitations.

OBJECTIVES: This study compared the performance for detection and grading of coronary stenoses using artificial intelligence-enabled quantitative coronary computed tomography (AI-QCT) angiography analyses to core lab-interpreted coronary CTA, core lab quantitative coronary angiography (QCA), and invasive fractional flow reserve (FFR).

METHODS: Coronary CTA, FFR, and QCA data from 303 stable patients (64 ± 10 years of age, 71% male) from the CREDENCE (Computed TomogRaphic Evaluation of Atherosclerotic DEtermiNants of Myocardial IsChEmia) trial were retrospectively analyzed using an Food and Drug Administration-cleared cloud-based software that performs AI-enabled coronary segmentation, lumen and vessel wall determination, plaque quantification and characterization, and stenosis determination.

RESULTS: Disease prevalence was high, with 32.0%, 35.0%, 21.0%, and 13.0% demonstrating ≥50% stenosis in 0, 1, 2, and 3 coronary vessel territories, respectively. Average AI-QCT analysis time was 10.3 ± 2.7 minutes. AI-QCT evaluation demonstrated per-patient sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 94%, 68%, 81%, 90%, and 84%, respectively, for ≥50% stenosis, and of 94%, 82%, 69%, 97%, and 86%, respectively, for detection of ≥70% stenosis. There was high correlation between stenosis detected on AI-QCT evaluation vs QCA on a per-vessel and per-patient basis (intraclass correlation coefficient = 0.73 and 0.73, respectively; P < 0.001 for both). False positive AI-QCT findings were noted in in 62 of 848 (7.3%) vessels (stenosis of ≥70% by AI-QCT and QCA of <70%); however, 41 (66.1%) of these had an FFR of <0.8.

CONCLUSIONS: A novel AI-based evaluation of coronary CTA enables rapid and accurate identification and exclusion of high-grade stenosis and with close agreement to blinded, core lab-interpreted quantitative coronary angiography. (Computed TomogRaphic Evaluation of Atherosclerotic DEtermiNants of Myocardial IsChEmia [CREDENCE]; NCT02173275).

Original language	English (US)
Pages (from-to)	193-205
Number of pages	13
Journal	JACC: Cardiovascular Imaging
Volume	16
Issue number	2
DOIs	https://doi.org/10.1016/j.jcmg.2021.10.020
State	Published - Feb 2023

Keywords

artificial intelligence
atherosclerosis
coronary CTA
coronary artery disease
coronary computed tomography
fractional flow reserve
quantitative coronary angiography

ASJC Scopus subject areas

Radiology Nuclear Medicine and imaging
Cardiology and Cardiovascular Medicine

Access to Document

10.1016/j.jcmg.2021.10.020

Cite this

Griffin, W. F., Choi, A. D., Riess, J. S., Marques, H., Chang, H. J., Choi, J. H., Doh, J. H., Her, A. Y., Koo, B. K., Nam, C. W., Park, H. B., Shin, S. H., Cole, J., Gimelli, A., Khan, M. A., Lu, B., Gao, Y., Nabi, F., Nakazato, R., ... Earls, J. P. (2023). AI Evaluation of Stenosis on Coronary CTA, Comparison With Quantitative Coronary Angiography and Fractional Flow Reserve: A CREDENCE Trial Substudy. JACC: Cardiovascular Imaging, 16(2), 193-205. https://doi.org/10.1016/j.jcmg.2021.10.020

Griffin, WF, Choi, AD, Riess, JS, Marques, H, Chang, HJ, Choi, JH, Doh, JH, Her, AY, Koo, BK, Nam, CW, Park, HB, Shin, SH, Cole, J, Gimelli, A, Khan, MA, Lu, B, Gao, Y, Nabi, F, Nakazato, R, Schoepf, UJ, Driessen, RS, Bom, MJ, Thompson, R, Jang, JJ, Ridner, M, Rowan, C, Avelar, E, Généreux, P, Knaapen, P, de Waard, GA, Pontone, G, Andreini, D & Earls, JP 2023, 'AI Evaluation of Stenosis on Coronary CTA, Comparison With Quantitative Coronary Angiography and Fractional Flow Reserve: A CREDENCE Trial Substudy', JACC: Cardiovascular Imaging, vol. 16, no. 2, pp. 193-205. https://doi.org/10.1016/j.jcmg.2021.10.020

@article{c5ce36ff50d3499d900e8a7254236acb,

title = "AI Evaluation of Stenosis on Coronary CTA, Comparison With Quantitative Coronary Angiography and Fractional Flow Reserve: A CREDENCE Trial Substudy",

abstract = "BACKGROUND: Clinical reads of coronary computed tomography angiography (CTA), especially by less experienced readers, may result in overestimation of coronary artery disease stenosis severity compared with expert interpretation. Artificial intelligence (AI)-based solutions applied to coronary CTA may overcome these limitations.OBJECTIVES: This study compared the performance for detection and grading of coronary stenoses using artificial intelligence-enabled quantitative coronary computed tomography (AI-QCT) angiography analyses to core lab-interpreted coronary CTA, core lab quantitative coronary angiography (QCA), and invasive fractional flow reserve (FFR).METHODS: Coronary CTA, FFR, and QCA data from 303 stable patients (64 ± 10 years of age, 71% male) from the CREDENCE (Computed TomogRaphic Evaluation of Atherosclerotic DEtermiNants of Myocardial IsChEmia) trial were retrospectively analyzed using an Food and Drug Administration-cleared cloud-based software that performs AI-enabled coronary segmentation, lumen and vessel wall determination, plaque quantification and characterization, and stenosis determination.RESULTS: Disease prevalence was high, with 32.0%, 35.0%, 21.0%, and 13.0% demonstrating ≥50% stenosis in 0, 1, 2, and 3 coronary vessel territories, respectively. Average AI-QCT analysis time was 10.3 ± 2.7 minutes. AI-QCT evaluation demonstrated per-patient sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 94%, 68%, 81%, 90%, and 84%, respectively, for ≥50% stenosis, and of 94%, 82%, 69%, 97%, and 86%, respectively, for detection of ≥70% stenosis. There was high correlation between stenosis detected on AI-QCT evaluation vs QCA on a per-vessel and per-patient basis (intraclass correlation coefficient = 0.73 and 0.73, respectively; P < 0.001 for both). False positive AI-QCT findings were noted in in 62 of 848 (7.3%) vessels (stenosis of ≥70% by AI-QCT and QCA of <70%); however, 41 (66.1%) of these had an FFR of <0.8.CONCLUSIONS: A novel AI-based evaluation of coronary CTA enables rapid and accurate identification and exclusion of high-grade stenosis and with close agreement to blinded, core lab-interpreted quantitative coronary angiography. (Computed TomogRaphic Evaluation of Atherosclerotic DEtermiNants of Myocardial IsChEmia [CREDENCE]; NCT02173275).",

keywords = "artificial intelligence, atherosclerosis, coronary CTA, coronary artery disease, coronary computed tomography, fractional flow reserve, quantitative coronary angiography",

author = "Griffin, {William F.} and Choi, {Andrew D.} and Riess, {Joanna S.} and Hugo Marques and Chang, {Hyuk Jae} and Choi, {Jung Hyun} and Doh, {Joon Hyung} and Her, {Ae Young} and Koo, {Bon Kwon} and Nam, {Chang Wook} and Park, {Hyung Bok} and Shin, {Sang Hoon} and Jason Cole and Alessia Gimelli and Khan, {Muhammad Akram} and Bin Lu and Yang Gao and Faisal Nabi and Ryo Nakazato and Schoepf, {U. Joseph} and Driessen, {Roel S.} and Bom, {Michiel J.} and Randall Thompson and Jang, {James J.} and Michael Ridner and Chris Rowan and Erick Avelar and Philippe G{\'e}n{\'e}reux and Paul Knaapen and {de Waard}, {Guus A.} and Gianluca Pontone and Daniele Andreini and Earls, {James P.}",

year = "2023",

month = feb,

doi = "10.1016/j.jcmg.2021.10.020",

language = "English (US)",

volume = "16",

pages = "193--205",

journal = "JACC: Cardiovascular Imaging",

issn = "1936-878X",

publisher = "Elsevier",

number = "2",

}

TY - JOUR

T1 - AI Evaluation of Stenosis on Coronary CTA, Comparison With Quantitative Coronary Angiography and Fractional Flow Reserve

T2 - A CREDENCE Trial Substudy

AU - Griffin, William F.

AU - Choi, Andrew D.

AU - Riess, Joanna S.

AU - Marques, Hugo

AU - Chang, Hyuk Jae

AU - Choi, Jung Hyun

AU - Doh, Joon Hyung

AU - Her, Ae Young

AU - Koo, Bon Kwon

AU - Nam, Chang Wook

AU - Park, Hyung Bok

AU - Shin, Sang Hoon

AU - Cole, Jason

AU - Gimelli, Alessia

AU - Khan, Muhammad Akram

AU - Lu, Bin

AU - Gao, Yang

AU - Nabi, Faisal

AU - Nakazato, Ryo

AU - Schoepf, U. Joseph

AU - Driessen, Roel S.

AU - Bom, Michiel J.

AU - Thompson, Randall

AU - Jang, James J.

AU - Ridner, Michael

AU - Rowan, Chris

AU - Avelar, Erick

AU - Généreux, Philippe

AU - Knaapen, Paul

AU - de Waard, Guus A.

AU - Pontone, Gianluca

AU - Andreini, Daniele

AU - Earls, James P.

PY - 2023/2

Y1 - 2023/2

N2 - BACKGROUND: Clinical reads of coronary computed tomography angiography (CTA), especially by less experienced readers, may result in overestimation of coronary artery disease stenosis severity compared with expert interpretation. Artificial intelligence (AI)-based solutions applied to coronary CTA may overcome these limitations.OBJECTIVES: This study compared the performance for detection and grading of coronary stenoses using artificial intelligence-enabled quantitative coronary computed tomography (AI-QCT) angiography analyses to core lab-interpreted coronary CTA, core lab quantitative coronary angiography (QCA), and invasive fractional flow reserve (FFR).METHODS: Coronary CTA, FFR, and QCA data from 303 stable patients (64 ± 10 years of age, 71% male) from the CREDENCE (Computed TomogRaphic Evaluation of Atherosclerotic DEtermiNants of Myocardial IsChEmia) trial were retrospectively analyzed using an Food and Drug Administration-cleared cloud-based software that performs AI-enabled coronary segmentation, lumen and vessel wall determination, plaque quantification and characterization, and stenosis determination.RESULTS: Disease prevalence was high, with 32.0%, 35.0%, 21.0%, and 13.0% demonstrating ≥50% stenosis in 0, 1, 2, and 3 coronary vessel territories, respectively. Average AI-QCT analysis time was 10.3 ± 2.7 minutes. AI-QCT evaluation demonstrated per-patient sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 94%, 68%, 81%, 90%, and 84%, respectively, for ≥50% stenosis, and of 94%, 82%, 69%, 97%, and 86%, respectively, for detection of ≥70% stenosis. There was high correlation between stenosis detected on AI-QCT evaluation vs QCA on a per-vessel and per-patient basis (intraclass correlation coefficient = 0.73 and 0.73, respectively; P < 0.001 for both). False positive AI-QCT findings were noted in in 62 of 848 (7.3%) vessels (stenosis of ≥70% by AI-QCT and QCA of <70%); however, 41 (66.1%) of these had an FFR of <0.8.CONCLUSIONS: A novel AI-based evaluation of coronary CTA enables rapid and accurate identification and exclusion of high-grade stenosis and with close agreement to blinded, core lab-interpreted quantitative coronary angiography. (Computed TomogRaphic Evaluation of Atherosclerotic DEtermiNants of Myocardial IsChEmia [CREDENCE]; NCT02173275).

AB - BACKGROUND: Clinical reads of coronary computed tomography angiography (CTA), especially by less experienced readers, may result in overestimation of coronary artery disease stenosis severity compared with expert interpretation. Artificial intelligence (AI)-based solutions applied to coronary CTA may overcome these limitations.OBJECTIVES: This study compared the performance for detection and grading of coronary stenoses using artificial intelligence-enabled quantitative coronary computed tomography (AI-QCT) angiography analyses to core lab-interpreted coronary CTA, core lab quantitative coronary angiography (QCA), and invasive fractional flow reserve (FFR).METHODS: Coronary CTA, FFR, and QCA data from 303 stable patients (64 ± 10 years of age, 71% male) from the CREDENCE (Computed TomogRaphic Evaluation of Atherosclerotic DEtermiNants of Myocardial IsChEmia) trial were retrospectively analyzed using an Food and Drug Administration-cleared cloud-based software that performs AI-enabled coronary segmentation, lumen and vessel wall determination, plaque quantification and characterization, and stenosis determination.RESULTS: Disease prevalence was high, with 32.0%, 35.0%, 21.0%, and 13.0% demonstrating ≥50% stenosis in 0, 1, 2, and 3 coronary vessel territories, respectively. Average AI-QCT analysis time was 10.3 ± 2.7 minutes. AI-QCT evaluation demonstrated per-patient sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 94%, 68%, 81%, 90%, and 84%, respectively, for ≥50% stenosis, and of 94%, 82%, 69%, 97%, and 86%, respectively, for detection of ≥70% stenosis. There was high correlation between stenosis detected on AI-QCT evaluation vs QCA on a per-vessel and per-patient basis (intraclass correlation coefficient = 0.73 and 0.73, respectively; P < 0.001 for both). False positive AI-QCT findings were noted in in 62 of 848 (7.3%) vessels (stenosis of ≥70% by AI-QCT and QCA of <70%); however, 41 (66.1%) of these had an FFR of <0.8.CONCLUSIONS: A novel AI-based evaluation of coronary CTA enables rapid and accurate identification and exclusion of high-grade stenosis and with close agreement to blinded, core lab-interpreted quantitative coronary angiography. (Computed TomogRaphic Evaluation of Atherosclerotic DEtermiNants of Myocardial IsChEmia [CREDENCE]; NCT02173275).

KW - artificial intelligence

KW - atherosclerosis

KW - coronary CTA

KW - coronary artery disease

KW - coronary computed tomography

KW - fractional flow reserve

KW - quantitative coronary angiography

UR - http://www.scopus.com/inward/record.url?scp=85147128903&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85147128903&partnerID=8YFLogxK

U2 - 10.1016/j.jcmg.2021.10.020

DO - 10.1016/j.jcmg.2021.10.020

M3 - Article

C2 - 35183478

AN - SCOPUS:85147128903

SN - 1936-878X

VL - 16

SP - 193

EP - 205

JO - JACC: Cardiovascular Imaging

JF - JACC: Cardiovascular Imaging

IS - 2

ER -

AI Evaluation of Stenosis on Coronary CTA, Comparison With Quantitative Coronary Angiography and Fractional Flow Reserve: A CREDENCE Trial Substudy

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this