Can the Combination of Photon-Counting CT and AI Change Our Approach to CAC Scoring?

For more than three decades, the power of zero has been one of the most influential concepts in preventive cardiology.

No coronary calcium.Low risk. Case closed. Or is it?

A recently published editorial in the American Journal of Preventive Cardiology challenges a common misconception.¹ A coronary artery calcium (CAC) score of zero does not mean absence of atherosclerosis. It simply means there is no calcium detectable by the Agatston method.

Can AI Enhance Detection of Subtle Calcification?

The authors of the editorial discuss a new AI-based approach (Agatston 2.0) that moves beyond the traditional 130-HU threshold by combining deep-learning coronary segmentation with voxel-level probabilistic analysis.

The results are intriguing.

Among nearly 4,000 individuals classified as having a CAC 0 score, AI detected subtle calcification in 21.7 percent. These individuals had a 1.7-fold higher risk of coronary heart disease and nearly double the likelihood of progressing to conventional CAC positivity during follow-up.

However, the editorial makes an even more important point. This isn't really about AI. It's about abandoning the idea that calcium is a binary phenomenon. The Agatston score transformed preventive cardiology because it simplified biology into a single number.

Now biology is becoming more nuanced. Low-density calcium, spatial distribution, lesion morphology, topological organization and sub-threshold calcification may all represent different stages of the same disease process.

For years, we've treated calcium scoring as a test. Perhaps it should be viewed as plaque phenotyping instead.

What Does the Future Hold for the Agatston Method and CAC Scoring?

While future developments are unlikely to replace the Agatston score, we will probably build upon it by integrating multiple calcium phenotypes into a more comprehensive assessment of coronary biology.

This is also where photon-counting computed tomography (PCCT) becomes particularly relevant.

If AI is trying to extract more information from subtle calcium patterns, PCCT provides the hardware capable of detecting those patterns with greater spatial resolution, lower noise, and improved reproducibility.

Better algorithms need better data. Photon-counting provides exactly that.

Embracing a New Evolution with CAC Scoring

The power of zero is not disappearing. It is evolving.

In the next era of calcium scoring, we probably will not ask:“Is calcium present?”

The new question will be: “What kind of calcium is present, and what does it tell us about the biology of the plaque?”

That is a far more interesting question.

Dr. Cademartiri is the director of advanced cardiovascular imaging and photon-counting CT at the Scientific Institute for Research, Hospitalization, and Healthcare Synlab Diagnostic Network in Naples, Italy. He is also a consultant in advanced cardiovascular imaging at CDI/Centro Diagnostico Italiano in Milan, Italy.

(Editor’s note: This blog is adapted with permission from Dr. Cademartiri’s original LinkedIn post at:

https://www.linkedin.com/feed/update/urn:li:activity:7485726363480543232/

Reference

1. Al-Kindi S, Villines TC. Beyond the “power of zero”: rethinking coronary calcium scoring in the AI era. Am J Prev Cardiol. 2026 July 11, 101744. https://doi.org/10.1016/j.ajpc.2026.101744