AI-enabled left atrial volumetry in coronary artery calcium scans (AI-CAC

Artificial intelligence Atrial fibrillation CHARGE-AF Coronary artery calcium Left atrial volume NT-proBNP

Journal

Journal of cardiovascular computed tomography
ISSN: 1876-861X
Titre abrégé: J Cardiovasc Comput Tomogr
Pays: United States
ID NLM: 101308347

Informations de publication

Date de publication:
23 Apr 2024
Historique:
received: 07 02 2024
revised: 04 04 2024
accepted: 13 04 2024
medline: 24 4 2024
pubmed: 24 4 2024
entrez: 23 4 2024
Statut: aheadofprint

Résumé

Coronary artery calcium (CAC) scans contain actionable information beyond CAC scores that is not currently reported. We have applied artificial intelligence-enabled automated cardiac chambers volumetry to CAC scans (AI-CAC At 1,2,3,4, and 5 years follow-up 36, 77, 123, 182, and 236 cases of AF were identified, respectively. The AUC for AI-CAC LA volume was significantly higher than CHARGE-AF for Years 1, 2, and 3 (0.83 vs. 0.74, 0.84 vs. 0.80, and 0.81 vs. 0.78, respectively, all p ​< ​0.05), but similar for Years 4 and 5, and significantly higher than NT-proBNP at Years 1-5 (all p ​< ​0.01), but not for combined CHARGE-AF and NT-proBNP at any year. AI-CAC LA significantly improved the continuous Net Reclassification Index for prediction of AF over years 1-5 when added to CHARGE-AF Risk Score (0.60, 0.28, 0.32, 0.19, 0.24), and NT-proBNP (0.68, 0.44, 0.42, 0.30, 0.37) (all p ​< ​0.01). AI-CAC LA volume enabled prediction of AF as early as one year and significantly improved on risk classification of CHARGE-AF Risk Score and NT-proBNP.

Sections du résumé

BACKGROUND BACKGROUND
Coronary artery calcium (CAC) scans contain actionable information beyond CAC scores that is not currently reported.
METHODS METHODS
We have applied artificial intelligence-enabled automated cardiac chambers volumetry to CAC scans (AI-CAC
RESULTS RESULTS
At 1,2,3,4, and 5 years follow-up 36, 77, 123, 182, and 236 cases of AF were identified, respectively. The AUC for AI-CAC LA volume was significantly higher than CHARGE-AF for Years 1, 2, and 3 (0.83 vs. 0.74, 0.84 vs. 0.80, and 0.81 vs. 0.78, respectively, all p ​< ​0.05), but similar for Years 4 and 5, and significantly higher than NT-proBNP at Years 1-5 (all p ​< ​0.01), but not for combined CHARGE-AF and NT-proBNP at any year. AI-CAC LA significantly improved the continuous Net Reclassification Index for prediction of AF over years 1-5 when added to CHARGE-AF Risk Score (0.60, 0.28, 0.32, 0.19, 0.24), and NT-proBNP (0.68, 0.44, 0.42, 0.30, 0.37) (all p ​< ​0.01).
CONCLUSION CONCLUSIONS
AI-CAC LA volume enabled prediction of AF as early as one year and significantly improved on risk classification of CHARGE-AF Risk Score and NT-proBNP.

Identifiants

pubmed: 38653606
pii: S1934-5925(24)00079-0
doi: 10.1016/j.jcct.2024.04.005
pii:
doi:

Types de publication

Journal Article

Langues

eng

Sous-ensembles de citation

IM

Informations de copyright

Copyright © 2024 The Authors. Published by Elsevier Inc. All rights reserved.

Auteurs

Morteza Naghavi (M)

HeartLung.AI, Houston, TX, USA. Electronic address: mn@vp.org.

David Yankelevitz (D)

Mount Sinai Hospital, New York, NY, USA.

Anthony P Reeves (AP)

Department of Computer Engineering, Cornell University, Ithaca, NY, USA.

Matthew J Budoff (MJ)

The Lundquist Institute, Torrance, CA, USA.

Dong Li (D)

The Lundquist Institute, Torrance, CA, USA.

Kyle Atlas (K)

HeartLung.AI, Houston, TX, USA.

Chenyu Zhang (C)

HeartLung.AI, Houston, TX, USA.

Thomas L Atlas (TL)

Tustin Teleradiology, Tustin, CA, USA.

Seth Lirette (S)

HeartLung.AI, Houston, TX, USA.

Jakob Wasserthal (J)

Universität Basel, Basel, Switzerland.

Sion K Roy (SK)

The Lundquist Institute, Torrance, CA, USA.

Claudia Henschke (C)

Mount Sinai Hospital, New York, NY, USA.

Nathan D Wong (ND)

Heart Disease Prevention Program, Division of Cardiology, University of California Irvine, CA, USA.

Christopher Defilippi (C)

Inova Heart and Vascular Institute, Falls Church, VA, USA.

Susan R Heckbert (SR)

University of Washington, Seattle, WA, USA.

Philip Greenland (P)

Northwestern University, Evanston, IL, USA.

Classifications MeSH