AI-Driven Quantitative Coronary CT Angiography in Suspected Coronary Artery Disease: Multicenter CONFIRM2 Registry

Alexander van Rosendael; Rine Nakanishi; Jeroen J. Bax; Gianluca Pontone; Saima Mushtaq; Ronny R. Buechel; Christoph Gräni; Gudrun Feuchtner; Pietro G. Lacaita; Amit R. Patel; Cristiane C. Singulane; Andrew D. Choi; Mouaz Al-Mallah; Daniele Andreini; Ronald P. Karlsberg; Geoffrey W. Cho; Carlos E. Rochitte; Mirvat Alasnag; Ashraf Hamdan; Filippo Cademartiri; Erica Maffei; Hugo Marques; Pedro de Araújo Gonçalves; Himanshu Gupta; Martin Hadamitzky; Omar Khalique; Dinesh Kalra; James D. Mills; Nick S. Nurmohamed; Paul Knaapen; Matthew Budoff; Kashif Shaikh; Enrico Martin; David M. German; Maros Ferencik; Andrew C. Oehler; Roderick Deaño; Prashant Nagpal; Marly van Assen; Carlo N. De Cecco; Vasileios Kamperidis; Borek Foldyna; Jan M. Brendel; Victor Y. Cheng; Kelley R. Branch; Marcio Bittencourt; Sabha Bhatti; Venkateshwar Polsani; George Wesbey; Rhanderson Cardoso; Ron Blankstein; Augustin Delago; Amit Pursnani; Amro Alsaid; Stephen Bloom; Melissa Aquino; Ibrahim Danad

doi:10.1016/j.jacadv.2026.102618

AI-Driven Quantitative Coronary CT Angiography in Suspected Coronary Artery Disease: Multicenter CONFIRM2 Registry

Alexander van Rosendael, Rine Nakanishi, Jeroen J. Bax, Gianluca Pontone, Saima Mushtaq, Ronny R. Buechel, Christoph Gräni, Gudrun Feuchtner, Pietro G. Lacaita, Amit R. Patel, Cristiane C. Singulane, Andrew D. Choi, Mouaz Al-Mallah, Daniele Andreini, Ronald P. Karlsberg, Geoffrey W. Cho, Carlos E. Rochitte, Mirvat Alasnag, Ashraf Hamdan, Filippo CademartiriErica Maffei, Hugo Marques, Pedro de Araújo Gonçalves, Himanshu Gupta, Martin Hadamitzky, Omar Khalique, Dinesh Kalra, James D. Mills, Nick S. Nurmohamed, Paul Knaapen, Matthew Budoff, Kashif Shaikh, Enrico Martin, David M. German, Maros Ferencik, Andrew C. Oehler, Roderick Deaño, Prashant Nagpal, Marly van Assen, Carlo N. De Cecco, Vasileios Kamperidis, Borek Foldyna, Jan M. Brendel, Victor Y. Cheng, Kelley R. Branch, Marcio Bittencourt, Sabha Bhatti, Venkateshwar Polsani, George Wesbey, Rhanderson Cardoso, Ron Blankstein, Augustin Delago, Amit Pursnani, Amro Alsaid, Stephen Bloom, Melissa Aquino, Ibrahim Danad

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

2 Scopus citations

Abstract

BACKGROUND: Plaque assessment by quantitative coronary CT angiography has demonstrated to correlate highly with intravascular ultrasound and optical coherence tomography, and these modalities have shown strong prognostic value.

OBJECTIVES: The purpose of this study was to identify the prognostic value of artificial intelligence-guided quantitative CCTA (AI-QCT) for major adverse cardiovascular events (MACE) against the risk factor-weighted clinical likelihood model.

METHODS: The CONFIRM2 (COroNary CT Angiography Evaluation For Evaluation of Clinical Outcomes: An InteRnational, Multicenter Registry) is a multicenter, international, observational cohort study that included patients with clinically indicated CCTA and follow-up for MACE. Patients without cardiac symptoms and prior coronary artery disease (CAD) were excluded. Across the entire coronary artery tree, the presence, extent, and composition of CAD were analyzed by an AI-QCT software, and 24 variables at a patient, vessel, and plaque level were derived, including percent luminal narrowing, remodeling index, plaque volumes (total, calcified, noncalcified, low attenuation), and plaque composition. The primary MACE endpoint was defined as a composite of all-cause death, myocardial infarction (MI), stroke, congestive heart failure, late revascularizations, and hospitalization for unstable angina. The secondary MACE endpoint was defined as all-cause death and MI.

RESULTS: A total of 3,551 patients (age 58.8 ± 12.5 years, 50.5% male) were followed for a median of 4.27 (IQR: 3.47-5.08) years during which 167 (4.7%) events occurred. After excluding collinear variables, diameter stenosis (HR: 1.25 [95% CI: 1.18-1.32]) per 10% increase and noncalcified plaque volume (HR: 1.07 [95% CI: 1.03-1.11]) per 50 mm 3 increase were the only independent predictors for MACE. In multivariable modeling, the discriminatory value defined by area under the curve (AUC) improved from 0.63 (95% CI: 0.58-0.67) based on the risk factor-weighted clinical likelihood model to 0.76 (95% CI: 0.77-0.80), P < 0.001 when adding AI-QCT-based diameter stenosis and noncalcified plaque volume. A similar improvement in risk prediction was seen when adding AI-QCT (AUC 0.77; P < 0.001) to a model with traditional risk factors, age, and sex (AUC: 0.67). In addition, AI-QCT significantly improved discrimination compared to the atherosclerotic cardiovascular disease risk score (AUC: 0.63; 95% CI: 0.58-0.68) to 0.75 (95% CI: 0.69-0.80; P < 0.001). Similar results were seen for the secondary MACE endpoint of death/MI.

CONCLUSIONS: This first multicenter global registry with AI-guided quantitative CT identified noncalcified plaque burden and increment in stenosis severity as the most powerful predictors of MACE, demonstrating the interplay between traditional and novel measures of the severity of CAD. Standardized and rapid quantitative assessment of CAD may improve clinical implementation of multidimensional assessment of CAD as a cornerstone for risk assessment.

Original language	English (US)
Article number	102618
Pages (from-to)	102618
Journal	JACC: Advances
Volume	5
Issue number	3
DOIs	https://doi.org/10.1016/j.jacadv.2026.102618
State	Published - Mar 2026

Keywords

artificial intelligence
coronary CT
coronary artery disease
noncalcified plaque volume
plaque quantification
quantitative coronary CT
risk stratification
total plaque volume

ASJC Scopus subject areas

Cardiology and Cardiovascular Medicine

Access to Document

10.1016/j.jacadv.2026.102618

Cite this

van Rosendael, A., Nakanishi, R., Bax, J. J., Pontone, G., Mushtaq, S., Buechel, R. R., Gräni, C., Feuchtner, G., Lacaita, P. G., Patel, A. R., Singulane, C. C., Choi, A. D., Al-Mallah, M., Andreini, D., Karlsberg, R. P., Cho, G. W., Rochitte, C. E., Alasnag, M., Hamdan, A., ... Danad, I. (2026). AI-Driven Quantitative Coronary CT Angiography in Suspected Coronary Artery Disease: Multicenter CONFIRM2 Registry. JACC: Advances, 5(3), 102618. Article 102618. https://doi.org/10.1016/j.jacadv.2026.102618

van Rosendael, A, Nakanishi, R, Bax, JJ, Pontone, G, Mushtaq, S, Buechel, RR, Gräni, C, Feuchtner, G, Lacaita, PG, Patel, AR, Singulane, CC, Choi, AD, Al-Mallah, M, Andreini, D, Karlsberg, RP, Cho, GW, Rochitte, CE, Alasnag, M, Hamdan, A, Cademartiri, F, Maffei, E, Marques, H, de Araújo Gonçalves, P, Gupta, H, Hadamitzky, M, Khalique, O, Kalra, D, Mills, JD, Nurmohamed, NS, Knaapen, P, Budoff, M, Shaikh, K, Martin, E, German, DM, Ferencik, M, Oehler, AC, Deaño, R, Nagpal, P, van Assen, M, De Cecco, CN, Kamperidis, V, Foldyna, B, Brendel, JM, Cheng, VY, Branch, KR, Bittencourt, M, Bhatti, S, Polsani, V, Wesbey, G, Cardoso, R, Blankstein, R, Delago, A, Pursnani, A, Alsaid, A, Bloom, S, Aquino, M & Danad, I 2026, 'AI-Driven Quantitative Coronary CT Angiography in Suspected Coronary Artery Disease: Multicenter CONFIRM2 Registry', JACC: Advances, vol. 5, no. 3, 102618, pp. 102618. https://doi.org/10.1016/j.jacadv.2026.102618

@article{c7c5aa55955142d390b0c8d7a1a8232f,

title = "AI-Driven Quantitative Coronary CT Angiography in Suspected Coronary Artery Disease: Multicenter CONFIRM2 Registry",

abstract = "BACKGROUND: Plaque assessment by quantitative coronary CT angiography has demonstrated to correlate highly with intravascular ultrasound and optical coherence tomography, and these modalities have shown strong prognostic value.OBJECTIVES: The purpose of this study was to identify the prognostic value of artificial intelligence-guided quantitative CCTA (AI-QCT) for major adverse cardiovascular events (MACE) against the risk factor-weighted clinical likelihood model.METHODS: The CONFIRM2 (COroNary CT Angiography Evaluation For Evaluation of Clinical Outcomes: An InteRnational, Multicenter Registry) is a multicenter, international, observational cohort study that included patients with clinically indicated CCTA and follow-up for MACE. Patients without cardiac symptoms and prior coronary artery disease (CAD) were excluded. Across the entire coronary artery tree, the presence, extent, and composition of CAD were analyzed by an AI-QCT software, and 24 variables at a patient, vessel, and plaque level were derived, including percent luminal narrowing, remodeling index, plaque volumes (total, calcified, noncalcified, low attenuation), and plaque composition. The primary MACE endpoint was defined as a composite of all-cause death, myocardial infarction (MI), stroke, congestive heart failure, late revascularizations, and hospitalization for unstable angina. The secondary MACE endpoint was defined as all-cause death and MI.RESULTS: A total of 3,551 patients (age 58.8 ± 12.5 years, 50.5\% male) were followed for a median of 4.27 (IQR: 3.47-5.08) years during which 167 (4.7\%) events occurred. After excluding collinear variables, diameter stenosis (HR: 1.25 [95\% CI: 1.18-1.32]) per 10\% increase and noncalcified plaque volume (HR: 1.07 [95\% CI: 1.03-1.11]) per 50 mm 3 increase were the only independent predictors for MACE. In multivariable modeling, the discriminatory value defined by area under the curve (AUC) improved from 0.63 (95\% CI: 0.58-0.67) based on the risk factor-weighted clinical likelihood model to 0.76 (95\% CI: 0.77-0.80), P < 0.001 when adding AI-QCT-based diameter stenosis and noncalcified plaque volume. A similar improvement in risk prediction was seen when adding AI-QCT (AUC 0.77; P < 0.001) to a model with traditional risk factors, age, and sex (AUC: 0.67). In addition, AI-QCT significantly improved discrimination compared to the atherosclerotic cardiovascular disease risk score (AUC: 0.63; 95\% CI: 0.58-0.68) to 0.75 (95\% CI: 0.69-0.80; P < 0.001). Similar results were seen for the secondary MACE endpoint of death/MI. CONCLUSIONS: This first multicenter global registry with AI-guided quantitative CT identified noncalcified plaque burden and increment in stenosis severity as the most powerful predictors of MACE, demonstrating the interplay between traditional and novel measures of the severity of CAD. Standardized and rapid quantitative assessment of CAD may improve clinical implementation of multidimensional assessment of CAD as a cornerstone for risk assessment.",

keywords = "artificial intelligence, coronary CT, coronary artery disease, noncalcified plaque volume, plaque quantification, quantitative coronary CT, risk stratification, total plaque volume",

author = "\{van Rosendael\}, Alexander and Rine Nakanishi and Bax, \{Jeroen J.\} and Gianluca Pontone and Saima Mushtaq and Buechel, \{Ronny R.\} and Christoph Gr{\"a}ni and Gudrun Feuchtner and Lacaita, \{Pietro G.\} and Patel, \{Amit R.\} and Singulane, \{Cristiane C.\} and Choi, \{Andrew D.\} and Mouaz Al-Mallah and Daniele Andreini and Karlsberg, \{Ronald P.\} and Cho, \{Geoffrey W.\} and Rochitte, \{Carlos E.\} and Mirvat Alasnag and Ashraf Hamdan and Filippo Cademartiri and Erica Maffei and Hugo Marques and \{de Ara{\'u}jo Gon{\c c}alves\}, Pedro and Himanshu Gupta and Martin Hadamitzky and Omar Khalique and Dinesh Kalra and Mills, \{James D.\} and Nurmohamed, \{Nick S.\} and Paul Knaapen and Matthew Budoff and Kashif Shaikh and Enrico Martin and German, \{David M.\} and Maros Ferencik and Oehler, \{Andrew C.\} and Roderick Dea{\~n}o and Prashant Nagpal and \{van Assen\}, Marly and \{De Cecco\}, \{Carlo N.\} and Vasileios Kamperidis and Borek Foldyna and Brendel, \{Jan M.\} and Cheng, \{Victor Y.\} and Branch, \{Kelley R.\} and Marcio Bittencourt and Sabha Bhatti and Venkateshwar Polsani and George Wesbey and Rhanderson Cardoso and Ron Blankstein and Augustin Delago and Amit Pursnani and Amro Alsaid and Stephen Bloom and Melissa Aquino and Ibrahim Danad",

year = "2026",

month = mar,

doi = "10.1016/j.jacadv.2026.102618",

language = "English (US)",

volume = "5",

pages = "102618",

journal = "JACC: Advances",

issn = "2772-963X",

publisher = "Elsevier B.V.",

number = "3",

}

TY - JOUR

T1 - AI-Driven Quantitative Coronary CT Angiography in Suspected Coronary Artery Disease

T2 - Multicenter CONFIRM2 Registry

AU - van Rosendael, Alexander

AU - Nakanishi, Rine

AU - Bax, Jeroen J.

AU - Pontone, Gianluca

AU - Mushtaq, Saima

AU - Buechel, Ronny R.

AU - Gräni, Christoph

AU - Feuchtner, Gudrun

AU - Lacaita, Pietro G.

AU - Patel, Amit R.

AU - Singulane, Cristiane C.

AU - Choi, Andrew D.

AU - Al-Mallah, Mouaz

AU - Andreini, Daniele

AU - Karlsberg, Ronald P.

AU - Cho, Geoffrey W.

AU - Rochitte, Carlos E.

AU - Alasnag, Mirvat

AU - Hamdan, Ashraf

AU - Cademartiri, Filippo

AU - Maffei, Erica

AU - Marques, Hugo

AU - de Araújo Gonçalves, Pedro

AU - Gupta, Himanshu

AU - Hadamitzky, Martin

AU - Khalique, Omar

AU - Kalra, Dinesh

AU - Mills, James D.

AU - Nurmohamed, Nick S.

AU - Knaapen, Paul

AU - Budoff, Matthew

AU - Shaikh, Kashif

AU - Martin, Enrico

AU - German, David M.

AU - Ferencik, Maros

AU - Oehler, Andrew C.

AU - Deaño, Roderick

AU - Nagpal, Prashant

AU - van Assen, Marly

AU - De Cecco, Carlo N.

AU - Kamperidis, Vasileios

AU - Foldyna, Borek

AU - Brendel, Jan M.

AU - Cheng, Victor Y.

AU - Branch, Kelley R.

AU - Bittencourt, Marcio

AU - Bhatti, Sabha

AU - Polsani, Venkateshwar

AU - Wesbey, George

AU - Cardoso, Rhanderson

AU - Blankstein, Ron

AU - Delago, Augustin

AU - Pursnani, Amit

AU - Alsaid, Amro

AU - Bloom, Stephen

AU - Aquino, Melissa

AU - Danad, Ibrahim

PY - 2026/3

Y1 - 2026/3

N2 - BACKGROUND: Plaque assessment by quantitative coronary CT angiography has demonstrated to correlate highly with intravascular ultrasound and optical coherence tomography, and these modalities have shown strong prognostic value.OBJECTIVES: The purpose of this study was to identify the prognostic value of artificial intelligence-guided quantitative CCTA (AI-QCT) for major adverse cardiovascular events (MACE) against the risk factor-weighted clinical likelihood model.METHODS: The CONFIRM2 (COroNary CT Angiography Evaluation For Evaluation of Clinical Outcomes: An InteRnational, Multicenter Registry) is a multicenter, international, observational cohort study that included patients with clinically indicated CCTA and follow-up for MACE. Patients without cardiac symptoms and prior coronary artery disease (CAD) were excluded. Across the entire coronary artery tree, the presence, extent, and composition of CAD were analyzed by an AI-QCT software, and 24 variables at a patient, vessel, and plaque level were derived, including percent luminal narrowing, remodeling index, plaque volumes (total, calcified, noncalcified, low attenuation), and plaque composition. The primary MACE endpoint was defined as a composite of all-cause death, myocardial infarction (MI), stroke, congestive heart failure, late revascularizations, and hospitalization for unstable angina. The secondary MACE endpoint was defined as all-cause death and MI.RESULTS: A total of 3,551 patients (age 58.8 ± 12.5 years, 50.5% male) were followed for a median of 4.27 (IQR: 3.47-5.08) years during which 167 (4.7%) events occurred. After excluding collinear variables, diameter stenosis (HR: 1.25 [95% CI: 1.18-1.32]) per 10% increase and noncalcified plaque volume (HR: 1.07 [95% CI: 1.03-1.11]) per 50 mm 3 increase were the only independent predictors for MACE. In multivariable modeling, the discriminatory value defined by area under the curve (AUC) improved from 0.63 (95% CI: 0.58-0.67) based on the risk factor-weighted clinical likelihood model to 0.76 (95% CI: 0.77-0.80), P < 0.001 when adding AI-QCT-based diameter stenosis and noncalcified plaque volume. A similar improvement in risk prediction was seen when adding AI-QCT (AUC 0.77; P < 0.001) to a model with traditional risk factors, age, and sex (AUC: 0.67). In addition, AI-QCT significantly improved discrimination compared to the atherosclerotic cardiovascular disease risk score (AUC: 0.63; 95% CI: 0.58-0.68) to 0.75 (95% CI: 0.69-0.80; P < 0.001). Similar results were seen for the secondary MACE endpoint of death/MI. CONCLUSIONS: This first multicenter global registry with AI-guided quantitative CT identified noncalcified plaque burden and increment in stenosis severity as the most powerful predictors of MACE, demonstrating the interplay between traditional and novel measures of the severity of CAD. Standardized and rapid quantitative assessment of CAD may improve clinical implementation of multidimensional assessment of CAD as a cornerstone for risk assessment.

AB - BACKGROUND: Plaque assessment by quantitative coronary CT angiography has demonstrated to correlate highly with intravascular ultrasound and optical coherence tomography, and these modalities have shown strong prognostic value.OBJECTIVES: The purpose of this study was to identify the prognostic value of artificial intelligence-guided quantitative CCTA (AI-QCT) for major adverse cardiovascular events (MACE) against the risk factor-weighted clinical likelihood model.METHODS: The CONFIRM2 (COroNary CT Angiography Evaluation For Evaluation of Clinical Outcomes: An InteRnational, Multicenter Registry) is a multicenter, international, observational cohort study that included patients with clinically indicated CCTA and follow-up for MACE. Patients without cardiac symptoms and prior coronary artery disease (CAD) were excluded. Across the entire coronary artery tree, the presence, extent, and composition of CAD were analyzed by an AI-QCT software, and 24 variables at a patient, vessel, and plaque level were derived, including percent luminal narrowing, remodeling index, plaque volumes (total, calcified, noncalcified, low attenuation), and plaque composition. The primary MACE endpoint was defined as a composite of all-cause death, myocardial infarction (MI), stroke, congestive heart failure, late revascularizations, and hospitalization for unstable angina. The secondary MACE endpoint was defined as all-cause death and MI.RESULTS: A total of 3,551 patients (age 58.8 ± 12.5 years, 50.5% male) were followed for a median of 4.27 (IQR: 3.47-5.08) years during which 167 (4.7%) events occurred. After excluding collinear variables, diameter stenosis (HR: 1.25 [95% CI: 1.18-1.32]) per 10% increase and noncalcified plaque volume (HR: 1.07 [95% CI: 1.03-1.11]) per 50 mm 3 increase were the only independent predictors for MACE. In multivariable modeling, the discriminatory value defined by area under the curve (AUC) improved from 0.63 (95% CI: 0.58-0.67) based on the risk factor-weighted clinical likelihood model to 0.76 (95% CI: 0.77-0.80), P < 0.001 when adding AI-QCT-based diameter stenosis and noncalcified plaque volume. A similar improvement in risk prediction was seen when adding AI-QCT (AUC 0.77; P < 0.001) to a model with traditional risk factors, age, and sex (AUC: 0.67). In addition, AI-QCT significantly improved discrimination compared to the atherosclerotic cardiovascular disease risk score (AUC: 0.63; 95% CI: 0.58-0.68) to 0.75 (95% CI: 0.69-0.80; P < 0.001). Similar results were seen for the secondary MACE endpoint of death/MI. CONCLUSIONS: This first multicenter global registry with AI-guided quantitative CT identified noncalcified plaque burden and increment in stenosis severity as the most powerful predictors of MACE, demonstrating the interplay between traditional and novel measures of the severity of CAD. Standardized and rapid quantitative assessment of CAD may improve clinical implementation of multidimensional assessment of CAD as a cornerstone for risk assessment.

KW - artificial intelligence

KW - coronary CT

KW - coronary artery disease

KW - noncalcified plaque volume

KW - plaque quantification

KW - quantitative coronary CT

KW - risk stratification

KW - total plaque volume

UR - https://www.scopus.com/pages/publications/105034224719

UR - https://www.scopus.com/inward/citedby.url?scp=105034224719&partnerID=8YFLogxK

U2 - 10.1016/j.jacadv.2026.102618

DO - 10.1016/j.jacadv.2026.102618

M3 - Article

C2 - 41906604

AN - SCOPUS:105034224719

SN - 2772-963X

VL - 5

SP - 102618

JO - JACC: Advances

JF - JACC: Advances

IS - 3

M1 - 102618

ER -

AI-Driven Quantitative Coronary CT Angiography in Suspected Coronary Artery Disease: Multicenter CONFIRM2 Registry

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this