Deep Learning and Fluid Dynamics On-Site CT-FFR Solution Compared to Off-Site FFRct and Invasive Fractional Flow Reserve (FFR)
Performance · Diagnostic performance for functionally significant CADResult — go to the resultAUC 0.91 vs 0.89Authors' conclusion
This study establishes xFFR as a robust and efficient on-site tool for assessing CAD, demonstrating high diagnostic accuracy, reproducibility, and agreement with invasive methods.
Study context
- Population
- Symptomatic patients at intermediate-to-high coronary artery disease (CAD) risk undergoing coronary computed tomography angiography (CTA) and invasive coronary angiography
- Participants
- 250 patients
- Comparison
- On-site xFFR vs off-site FFRct, with invasive FFR as reference
- Outcome
- Diagnostic performance for functionally significant CAD vs iFFR (AUC)
Methodological detailssingle-centre, prospective
- Single-centre
- Prospective
Performance
As reported in the abstract
Diagnostic performance for functionally significant CAD
AUC
0.91 vs 0.89
P = 0.274
By group
Further findings
| Outcome | xFFR | FFRct | iFFR | AUC | P value | |
|---|---|---|---|---|---|---|
| AUC, left anterior descending artery | 0.96 vs 0.84 | P = 0.001 | ||||
| AUC | 0.96 | 0.84 | Not reported in the abstract | iFFR: not reported in the abstract |
- xFFR analysis time
- 8 · mean · SD 3.4 · minutes
Correlation of xFFR (Spearman)
- xFFR vs iFFR
- ρ = 0.67
- xFFR vs FFRct
- ρ = 0.53
Original publication
Deep Learning and Fluid Dynamics On-Site CT-FFR Solution Compared to Off-Site FFRct and Invasive FFR. JACC Cardiovasc Imaging. Published 27 February 2026. doi:10.
Summarised from the published abstract; every number is checked against it