FDA clears Elucid's plaque-aware FFR-CT software
Elucid's FFR-IQ just became the third commercially cleared FFR-CT platform in the US — and the first to fold coronary plaque composition directly into the hemodynamic estimate, a sign of how layered AI cardiac imaging software is getting.
Elucid received FDA 510(k) clearance (K262015) for FFR-IQ, its software for non-invasive estimation of fractional flow reserve (FFR) from coronary CT angiography (CCTA). The clearance was authorized September 29, 2026, with the commercial rollout announced October 21, 2026.
What sets FFR-IQ apart, per HIT Consultant's coverage, is BioIntegrated FFR-CT — the model folds coronary plaque composition into the physiology estimate alongside anatomy, rather than computing hemodynamics from vessel geometry alone. In the supporting ELUCIDATE FFR-CT validation study, the software showed 92.2% sensitivity and 93.1% specificity against invasive FFR, the reference standard that otherwise requires a catheterization lab visit.
Elucid is now the third vendor cleared to sell FFR-CT commercially in the US, following Heartflow (2016) and KeyaMed (2022). For teams building the software layer around a clearance like this, the harder problem usually isn't the imaging model itself — it's threading plaque segmentation, physiology simulation, and a clinician-facing read into one pipeline that stays auditable end to end. That's the kind of multi-model integration work a medical imaging software partner like YuSMP Group's GenAI integration team handles when stitching several AI components into a single regulated clinical workflow.
Why plaque biology changes the FFR-CT picture
Traditional FFR-CT estimates how much a coronary blockage restricts blood flow using vessel geometry derived from the CT scan. Folding in plaque composition — distinguishing calcified, fibrous, and lipid-rich plaque — gives clinicians a physiology estimate that accounts for how the plaque itself is likely to behave, not just how narrow the vessel looks. That matters for treatment decisions: a moderately narrowed vessel with unstable plaque reads differently than the same narrowing with stable, calcified plaque.
What it means for cardiac imaging software teams
Each new FFR-CT competitor raises the bar on what counts as a complete non-invasive cardiac workup. Software built around CCTA data now has to reconcile plaque characterization, hemodynamic simulation, and reporting in a single auditable chain — architecture decisions that get harder to retrofit the later they're made.
Building AI-assisted cardiac or imaging software? Talk to Epica →
More from the blog.
Why your DICOM viewer feels slow (and how we fix it)
The bottleneck is rarely the network. Here's what actually makes medical image viewers crawl — and the engineering that makes them fly.
FDA clears AI tool that reads MRI to flag Alzheimer's risk
NeuroXT's NEXT-Cue just became a Class II medical device — another SaMD clearance reshaping imaging AI.
Building regulated imaging software?
From model pipeline to clinical UI, we build the software layer that holds up under FDA review.