11.1 Artificial Intelligence in Interventional Medicine: Beyond Imaging to Procedures. Where are we Headed?

Problem Presenter: Larry Dean

These proceedings summarize the educational activity of the 18th Biennial Meeting of the International Andreas Gruentzig Society held January 27 to 30, 2026, in Puerto Ayora, Santa Cruz, Ecuador.

Faculty Disclosures     Sponsors

2026 IAGS Summary Document

Statement of problem or issue

Coronary computed tomography angiography (CCTA) has evolved dramatically as an imaging technology, and quantitative methods of assessment, including several based on artificial intelligence (AI) programs, are being applied to these systems.[1] One specific area of interest is use of CCTA for planning PCI procedures, especially PCI of chronic total occlusions (CTO).[2,3] These analytic programs can inform operators about many features of the planned procedure, including luminal dimensions, lesion (or occlusion) length, anticipated stent length, plaque composition including calcium burden, case complexity, guide catheter selection, optimal fluoroscopic angles, and others.

Gaps in current knowledge

Everything in this field is a knowledge gap area. One important aspect to be explored is the existence of proprietary technologies in CCTA and AI. Ready access to images and the systems by which they can be manipulated and combined must be available.

Possible solutions or future directions

First, interventional cardiologists will need to become familiar with CCTA and interpretation of the images. Credentialing likely will be required, and in some institutions already is, and interventional cardiologists must be part of this process. Additionally, cardiologists will need to become more familiar will AI and what it can and cannot do.

Multimodality imaging has been used in structural heart interventions for years. [4] And, as another example, certain electrophysiology ablation procedures already utilize multimodality imaging. [5] With AI-enhanced CCTA it may become possible to have real-time integrated multimodality imaging as a routine component of PCI procedures. When this was combined with a hands-free, wearable, computer screen with reconstructed 3-dimensional CCTA images of CTO lesions, it has been demonstrated and found feasible for guiding PCI.[6] These and many other systems will have to be examined to determine their ultimate utility. Interventional cardiologists must be involved in leading AI use and its appropriate application in our field.

One concern with AI applied to imaging and “AI coaching” during procedures is the possibility for AI-generated hallucinations.[7] The ultimate question is: Will these systems really make us better?

References

1. Jo JI, et al. Artificial Intelligence-Driven Assessment of Coronary Computed Tomography Angiography for Intermediate Stenosis: Comparison With Quantitative Coronary Angiography and Fractional Flow Reserve. Am J Cardiol. 2025 Mar 15;239:82-89. doi: 10.1016/j.amjcard.2024.12.011. Epub 2024 Dec 11. PMID: 39672486.
2. Sandoval Y, et al. Coronary Computed Tomography Angiography to Guide Percutaneous Coronary Intervention: Expert Opinion from a SCAI/SCCT Roundtable. J Soc Cardiovasc Angiogr Interv. 2025 May 1;4(6):103664. doi: 10.1016/j.jscai.2025.103664. PMID: 40630246.
3. Kumar S, et al. The Role of Coronary Computed Tomography Angiography in Chronic Total Occlusion Percutaneous Coronary Intervention. JACC Cardiovasc Interv. 2026 Jan 12;19(1):1-14. doi: 10.1016/j.jcin.2025.10.055. PMID: 41534971.
4. Agricola E, et al. Multimodality imaging for patient selection, procedural guidance, and follow-up of transcatheter interventions for structural heart disease: a consensus document of the EACVI Task Force on Interventional Cardiovascular Imaging: part 1: access routes, transcatheter aortic valve implantation, and transcatheter mitral valve interventions. Eur Heart J Cardiovasc Imaging. 2023 Aug 23;24(9):e209-e268. doi: 10.1093/ehjci/jead096. PMID: 37283275.
5. Ninni S, et al. Stereotactic Radioablation for Ventricular Tachycardia in the Setting of Electrical Storm. Circ Arrhythm Electrophysiol. 2022;15(9):e010955. doi: 10.1161/CIRCEP.122.010955. Epub 2022 Sep 8. PMID: 36074658.
6. Opolski MP, et al. Feasibility and safety of augmented-reality glass for computed tomography-assisted percutaneous revascularization of coronary chronic total occlusion: A single center prospective pilot study. J Cardiovasc Comput Tomogr. 2017;11(6):489-496. doi: 10.1016/j.jcct.2017.09.013. Epub 2017 Sep 21. PMID: 28964751.
7. Xia M, et al. On Hallucinations in Artificial Intelligence-Generated Content for Nuclear Medicine Imaging (the DREAM Report). J Nucl Med. 2026;67(2):166-174. doi: 10.2967/jnumed.125.270653. PMID: 41198241.