Artificial intelligence in coronary plaque characterization and risk assessment: from images to impact

Abstract

Accurate characterization of coronary atherosclerotic plaque and individualized cardiovascular risk assessment remain active challenges in clinical and interventional cardiology. In recent times, Artificial Intelligence (AI) has emerged as a powerful new tool able to support clinicians in determining diagnoses and prognoses from coronary imaging. This commentary focuses on the current applications of AI in coronary plaque imaging, particularly on coronary computed tomography angiography (CCTA), intravascular ultrasound (IVUS), and optical coherence tomography (OCT), evaluating its role in identifying high-risk plaque features and predicting future adverse cardiovascular events. We discuss limitations of conventional assessment methods, illustrating how AI algorithms can improve reproducibility, reduce operator dependence, and examine current evidence from registries and clinical studies. Furthermore, some key challenges remain to be addressed, including data quality, model generalizability, clinical integration, and regulatory concerns. We argue that AI’s promise lies not in replacing clinical expertise, but in empowering coronary risk stratification and characterization. Ongoing validation and clinician-AI collaboration will be essential to ensure meaningful patient outcomes.

Keywords

Artificial intelligence / coronary plaque / risk stratification

References

[1] Nedkoff L,Zemedikun D,Wright FL.Global trends in atherosclerotic cardiovascular disease.Clin Ther2023;45:1087-91

[2] Labrecque Langlais É,Tastet O.Evaluation of stenoses using AI video models applied to coronary angiography.NPJ Digit Med2024;7:138 PMCID:PMC11116436

[3] Abdelrahman KM,Dey AK.Coronary computed tomography angiography from clinical uses to emerging technologies: JACC state-of-the-art review.J Am Coll Cardiol2020;76:1226-43 PMCID:PMC7480405

[4] Dzaye O,Blaha MJ.Evaluation of coronary stenosis versus plaque burden for atherosclerotic cardiovascular disease risk assessment and management.Curr Opin Cardiol2021;36:769-75 PMCID:PMC8547346

[5] Veelen A, van der Sangen NMR, Henriques JPS, Claessen BEPM. Identification and treatment of the vulnerable coronary plaque.Rev Cardiovasc Med2022;23:39

[6] Chen Q,Wang YN.A coronary CT angiography radiomics model to identify vulnerable plaque and predict cardiovascular events.Radiology2023;307:e221693

[7] Tearney GJ,Akasaka T.Consensus standards for acquisition, measurement, and reporting of intravascular optical coherence tomography studies: a report from the international working group for intravascular optical coherence tomography standardization and validation.J Am Coll Cardiol2012;59:1058-72

[8] Radu MD,García-García HM.Variability in the measurement of minimum fibrous cap thickness and reproducibility of fibroatheroma classification by optical coherence tomography using manual versus semi-automatic assessment.EuroIntervention2016;12:e987-97

[9] Huisman J,Rdzanek A.Multicenter assessment of the reproducibility of volumetric radiofrequency-based intravascular ultrasound measurements in coronary lesions that were consecutively stented.Int J Cardiovasc Imaging2012;28:1867-78 PMCID:PMC3485535

[10] Klüner LV,Antoniades C.Using artificial intelligence to study atherosclerosis from computed tomography imaging: a state-of-the-art review of the current literature.Atherosclerosis2024;398:117580 PMCID:PMC11579307

[11] Xu P,Zhou F.Artificial intelligence in coronary computed tomography angiography.Medicine Plus2024;1:100001

[12] Rinehart S,Ng N.Utility of artificial intelligence plaque quantification: results of the DECODE study.J Soc Cardiovasc Angiogr Interv2024;3:101296 PMCID:PMC11308844

[13] Cho GW,Quesada CG.Serial analysis of coronary artery disease progression by artificial intelligence assisted coronary computed tomography angiography: early clinical experience.BMC Cardiovasc Disord2022;22:506 PMCID:PMC9701371

[14] Chandramohan N,O’Kane P.Artificial intelligence for the interventional cardiologist: powering and enabling OCT image interpretation.Interv Cardiol2024;19:e03 PMCID:PMC10964291

[15] Zhang C,Guo X.Machine learning model comparison for automatic segmentation of intracoronary optical coherence tomography and plaque cap thickness quantification.Comput Model Eng Sci2020;123:631-46

[16] Nurmohamed NS,Jukema RA.AI-guided quantitative plaque staging predicts long-term cardiovascular outcomes in patients at risk for atherosclerotic CVD.JACC Cardiovasc Imaging2024;17:269-80

[17] Greenland P,Budoff MJ,Watson KE.Coronary calcium score and cardiovascular risk.J Am Coll Cardiol2018;72:434-47 PMCID:PMC6056023

[18] Criqui MH,Ix JH.Calcium density of coronary artery plaque and risk of incident cardiovascular events.JAMA2014;311:271-8 PMCID:PMC4091626

[19] Muzammil MA,Afridi AK.Artificial intelligence-enhanced electrocardiography for accurate diagnosis and management of cardiovascular diseases.J Electrocardiol2024;83:30-40

[20] Feuchtner GM,Bax JJ.AI-Quantitative CT coronary plaque features associate with a higher relative risk in women: CONFIRM2 registry.Circ Cardiovasc Imaging2025;18:e018235

[21] Williams MC,Doris M.Low-attenuation noncalcified plaque on CCTA predicts myocardial infarction: results from the multicenter SCOT-HEART trial (Scottish computed tomography of the HEART).Circulation2020;141:1452-62

[22] Kim Y,Suh J.FLASH Trial InvestigatorsArtificial intelligence-based fully automated quantitative coronary angiography vs optical coherence tomography-guided PCI: The FLASH Trial.JACC Cardiovasc Interv2025;18:187-97

[23] Rosenbacke R,McKee M.How Explainable artificial intelligence can increase or decrease clinicians’ trust in AI applications in health care: systematic review.JMIR AI2024;3:e53207

[24] Pinna A,Mannelli L.Machine learning for coronary plaque characterization: a multimodal review of OCT, IVUS, and CCTA.Diagnostics2025;15:1822 PMCID:PMC12293362

[25] Tejani AS,Hussain M,O’Donnell KP.Integrating and adopting AI in the radiology workflow: a primer for standards and integrating the healthcare enterprise (IHE) profiles.Radiology2024;311:e232653

[26] Dobrolińska MM,Pociask E.Performance of integrated near-infrared spectroscopy and intravascular Ultrasound (NIRS-IVUS) system against quantitative flow ratio (QFR).Diagnostics2021;11:1148 PMCID:PMC8305529

[27] Wikimedia Commons. CCTA CAD-RADS 4a.png. Available from: https://commons.wikimedia.org/wiki/File:CCTA_CAD-RADS_4a.png [Last accessed on 15 Sep 2025]

[28] Viscusi MM,Migliaro G.Current applications and new perspectives in optical coherence tomography (OCT) coronary atherosclerotic plaque assessment: from PCI optimization to pharmacological treatment guidance.Photonics2023;10:158

[29] Lin A,McElhinney P.Deep learning-enabled coronary CT angiography for plaque and stenosis quantification and cardiac risk prediction: an international multicentre study.Lancet Digit Health2022;4:e256-65

[30] Choudhary G,Ritter N.Interobserver reliability in the assessment of coronary stenoses by multidetector computed tomography.J Comput Assist Tomogr2011;35:126-34

[31] Zhuang B,Zhao S.Computed tomography angiography-derived fractional flow reserve (CT-FFR) for the detection of myocardial ischemia with invasive fractional flow reserve as reference: systematic review and meta-analysis.Eur Radiol2020;30:712-25

[32] Lipkin I,Kim Y.Coronary CTA With AI-QCT interpretation: comparison with myocardial perfusion imaging for detection of obstructive stenosis using invasive angiography as reference standard.AJR Am J Roentgenol2022;219:407-19

[33] Nurmohamed NS,Jukema RA.CREDENCE and PACIFIC-1 InvestigatorsDevelopment and validation of a quantitative coronary CT angiography model for diagnosis of vessel-specific coronary ischemia.JACC Cardiovasc Imaging2024;17:894-906

[34] Maclean E,Newby DE,Williams MC.Prognostic utility of semi-quantitative coronary computed tomography angiography scores in the SCOT-HEART trial.J Cardiovasc Comput Tomogr2023;17:393-400

[35] Johnson KM,Zhao Y,Staib LH.Scoring of coronary artery disease characteristics on coronary CT angiograms by using machine learning.Radiology2019;292:354-62

[36] Abbott. Ultreon^TM 1.0 software: experience the power of automation. Available from: https://www.cardiovascular.abbott/int/en/hcp/products/percutaneous-coronary-intervention/intravascular-imaging/ultreon-software/ultreon-1-0.html [Last accessed on 15 Sep 2025]

[37] Boston Scientific. AVVIGO^TM+ Multi-Modality Guidance System. Available from: https://www.bostonscientific.com/en-US/products/ffr-ivus-systems/avvigo-guidance-system.html [Last accessed on 15 Sep 2025]

[38] Galo J,Al-Qaraghuli A.Machine learning in intravascular ultrasound: validating automated lesion assessment for complex coronary interventions.Catheter Cardiovasc Interv2025;105:1320-8

[39] In Kim Y,Kweon J.Artificial intelligence-based quantitative coronary angiography of major vessels using deep-learning.Int J Cardiol2024;405:131945

[40] HeartFlow. Heartflow introduces next generation interactive plaque analysis platform to assess patient risk in suspected coronary artery disease. Available from: https://www.heartflow.com/press-release/Heartflow-plaque-interactive/ [Last accessed on 15 Sep 2025]

[41] Cleerly. Not all plaque analysis software are the same: roviding multi-modality imaging decision support. Available from: https://cleerlyhealth.com/plaque-analysis [Last accessed on 15 Sep 2025]

[42] Siemens Healthineers. AI-Rad Companion: providing multi-modality imaging decision support. Available from: https://www.siemens-healthineers.com/en-us/digital-health-solutions/ai-rad-companion [Last accessed on 15 Sep 2025]

[43] Kay FU,Kukkar V.Diagnostic accuracy of on-premise automated coronary CT Angiography analysis based on coronary artery disease reporting and data system 2.0.Radiology2025;315:e242087

[44] NANOX. Nanox receives FDA clearance for HealthCCSng V2.0, upgraded version of advanced AI cardiac solution empowering physicians in assessment of coronary artery calcium. Available from: https://investors.nanox.vision/news-releases/news-release-details/nanox-receives-fda-clearance-healthccsng-v20-upgraded-version [Last accessed on 15 Sep 2025]

[45] Kerndt CC,Weber P.Using artificial intelligence to semi-quantitate coronary calcium as an ‘Incidentaloma’ on non-gated, non-contrast CT scans, a single-center descriptive study in West Michigan.Spartan Med Res J2023;8:89132 PMCID:PMC10702149