Coronary plaque imaging keeps improving thanks to advanced AI, photon-counting CT
Quantitative plaque analysis using coronary CT angiography (CCTA) is gaining momentum as artificial intelligence (AI) tools, growing clinical evidence and advances in scanner technology expand the ability of clinicians to characterize coronary artery disease and better guide treatment.
Jonathan Leipsic, MD, chairman of radiology and a professor of radiology and cardiology at the University of British Columbia in Canada, said growing interest in AI-driven plaque analysis is fueling investments and technical innovation across the cardiovascular imaging field. He spoke with Cardiovascular Business about this topic, saying he is looking forward to new data being presented in this area at TCT 2026 Oct. 31- Nov. 3 in San Diego.
Research increasingly supports the feasibility, reproducibility and prognostic value of quantitative plaque analysis, with emerging opportunities to track disease over time and assess responses to medical therapy. However, Leipsic emphasized that clinicians must understand the strengths and limitations of individual software products, even as evidence suggests a broadly consistent relationship between quantitative plaque measurements and prognosis across vendors.
AI enables more detailed plaque quantification
He said AI addresses a fundamental limitation of conventional CCTA. While physicians can identify atherosclerosis, consistently measuring the extent and composition of plaque is difficult and time-consuming to perform manually.
“Human intelligence isn't made for the segmentation of quantitative plaque in a very highly reproducible way,” Leipsic explained.
AI can accurately quantify total plaque burden and distinguish calcified from noncalcified plaque, providing information that may help physicians better assess cardiovascular risk and tailor preventive treatment. The findings also can help clinicians determine whether a patient's atherosclerosis burden is unusually high for their age. Such information could support more individualized decisions about the intensity of lipid-lowering therapy, including appropriate low-density lipoprotein (LDL) cholesterol targets.
Leipsic said the goal is to move beyond simply identifying coronary disease toward using quantitative measurements to guide a more personalized approach to medical management.
Over time, researchers hope serial imaging will help determine whether treatment stabilizes noncalcified plaque, or produces changes in plaque composition associated with treatment response. These capabilities could eventually help clinicians evaluate the effectiveness of therapies and identify patients who may benefit from more intensive or expensive medications.
"I think this is definitely where the field is going," he said.
Clinical evidence remains essential in advancing CCTA AI
Although U.S. Food and Drug Administration (FDA) clearance and reimbursement pathways have helped advance quantitative plaque analysis into clinical practice, Leipsic said regulatory clearance should be viewed as a starting point rather than the end of clinical validation. He pointed to fractional flow reserve derived from CT (FFR-CT) as an example of how a technology can progress through additional clinical evidence after an initial clearance. Randomized trials, large registries and extended follow-up have helped establish its role in clinical decision-making.
Similar investment is needed in prospective studies of quantitative plaque analysis, including research into how the technology affects treatment decisions, patient outcomes and healthcare costs. Leipsic said collaboration among technology vendors, pharmaceutical companies, the National Institutes of Health and clinical investigators will be important to generating that evidence.
Photon-counting CT could improve plaque characterization
Hardware advances also may strengthen the capabilities of quantitative plaque analysis. Leipsic highlighted photon-counting CT as a promising development that could improve spatial resolution and tissue characterization.
Potential benefits include a better visualization of structures obscured by coronary calcium, reduced image noise and improved evaluations of coronary stents. Greater image fidelity could provide more detailed information for plaque quantification and other coronary assessments.
Leipsic also sees growing interest in combining coronary anatomy, plaque characteristics and physiology to guide treatment. In addition to helping determine which patients should undergo invasive coronary angiography, FFR-CT and plaque analysis is starting to help plan complex percutaneous coronary intervention procedures.
Despite the rapid pace of innovation, Leipsic stressed that adoption must remain grounded in evidence demonstrating accuracy, reproducibility, clinical utility and cost-effectiveness.
"I think that all of this is coming together in a very positive way. It's now just a matter of generating the data, as was seen with FFR-CT with a stepwise approach," he said. "Enthusiasm is important, but that enthusiasm should be translated into investment in science and data to make sure that we first do no harm."
Dave Fornell has covered healthcare for more than 17 years, with a focus in cardiology and radiology. Fornell is a 5-time winner of a Jesse H. Neal Award, the most prestigious editorial honors in the field of specialized journalism. The wins included best technical content, best use of social media and best COVID-19 coverage. Fornell was also a three-time Neal finalist for best range of work by a single author. He produces more than 100 editorial videos each year, most of them interviews with key opinion leaders in medicine. He also writes technical articles, covers key trends, conducts video hospital site visits, and is very involved with social media. E-mail: [email protected]