20 million chest CTs. Why do not we using all the information they contain?

ESC Congress 2026 · Scientific Poster

Nearly 20 million chest CT scans are performed each year in the U.S.—yet fewer than 5% are ECG-gated.

What if AI could extract meaningful cardiac information from the routine non-gated chest CTs already being performed every day?

In a new HeartLung.AI scientific poster at ESC Congress 2026, researchers evaluated AI-based cardiac chamber volumetry from non-gated chest CT and compared its ability to predict heart failure (HF) and atrial fibrillation (AF) with ECG-gated cardiac CT.

Non-gated CT. Comparable predictive value.

The study evaluated 2,053 MESA participants, comparing AI-derived chamber volumetry from non-gated and ECG-gated CT.

Across key cardiac measurements:

• LVVI: C-index 0.77 vs. 0.77 for HF

• LVMI: 0.77 vs. 0.78 for HF

• LV/RV ratio: 0.74 vs. 0.74 for HF

• LAVI: 0.74 vs. 0.75 for AF

• 
  

All four non-gated measurements met the study’s non-inferiority criterion.

Turning routine imaging into a cardiovascular opportunity

AI-derived chamber volumetry from routine non-gated chest CT was non-inferior to ECG-gated CT for predicting incident heart failure and atrial fibrillation.

And because non-gated scans vastly outnumber gated cardiac CTs, this could expand opportunities for opportunistic HF and AF risk assessment using imaging that already exists.

No additional scan. No additional cost. No additional radiation.

One Scan. Many Answers.

📍 To be presented at ESC Congress 2026 · Munich

By Seyed Reza Mirjalili, MD