Cardiology · Coronary CT FFR
HeartFlow FFRCT Analysis
HeartFlow, Inc.
HeartFlow's FFRCT Analysis is the most-published AI in interventional cardiology — anchored by multiple randomised controlled trials (PRECISE, FORECAST), 600+ peer-reviewed publications accumulated over a decade, and a De Novo regulatory pathway that places it in a distinct regulatory class from the 510(k) field.
Performance Metrics
Clinical Evidence
HeartFlow FFRCT has the broadest clinical evidence base of any AI medical device in cardiology. The NXT study (n=254, 4 countries; *JAMA* 2014) provided the pivotal De Novo dataset, showing FFRCT diagnostic accuracy of 86% versus 65% for CCTA alone against invasive FFR as reference. The PLATFORM prospective comparative-effectiveness study enrolled patients referred for invasive coronary angiography (ICA) and showed FFRCT-guided strategy cancelled ICA in 61% of planned procedures; the proportion of ICA finding no obstructive CAD fell from 73% (usual care) to 12% (FFRCT-guided), with zero adverse events at one-year follow-up in 117 patients whose ICA was cancelled. The FORECAST RCT demonstrated a 22% reduction in overall invasive testing versus usual care, and a 70% reduction in the composite of death, MI, or ICA without obstructive CAD at one year. The PRECISE trial — the first global RCT of a CCTA+FFRCT precision pathway — enrolled 2,103 patients across 65 sites and showed a 4× reduction in unnecessary invasive catheterisation and a 75% higher likelihood of identifying patients who required intervention compared with standard evaluation. Results were presented as late-breaking data at the AHA Scientific Sessions. The ADVANCE registry (n=5,083 patients; 38 sites; North America, Europe, Japan) provided prospective real-world evidence; seven-year ADVANCE-DK follow-up confirmed prognostic value: FFRCT-defined incomplete revascularisation carried 16.2% MACE vs 5.7% for normal FFRCT (AUC 0.73 with combined plaque analysis vs 0.63 for baseline risk variables + CCTA). The FISH&CHIPS study (NHS England; 27 hospital sites; 90,553 patients; ≥2-year follow-up) was published in *Nature Medicine* in May 2025 — the largest FFRCT study ever conducted — confirming reduced unnecessary invasive and non-invasive testing at national scale. A December 2025 analysis of the same cohort documented £1,042 cost savings per patient and prognostic superiority. HeartFlow FFRCT is covered by Medicare and >99% of commercially insured lives in the US; a Category I CPT code took effect January 1, 2024, cementing FFRCT as a reimbursed standard-of-care pathway for stable suspected CAD.
| Study | Design | n | Sensitivity | Specificity | AUC | Published |
|---|---|---|---|---|---|---|
| Nørgaard BL, Leipsic J, Gaur S, et al. (NXT Trial) | ProspectiveProspective | 254 | 86% | 79% | 0.90 | JACC, Apr 2014 |
| Douglas PS, Pontone G, Hlatky MA, et al. (PLATFORM) | ProspectiveProspective | 584 | — | — | — | JACC, Aug 2016 |
| Patel MR, Nørgaard BL, Fairbairn TA, et al. (ADVANCE Registry) | ProspectiveProspective | 5,083 | — | — | — | JACC Cardiovasc Imaging, Aug 2020 |
| NHS FISH&CHIPS Study Group (27 NHS sites, England) | RetrospectiveRetrospective | 90,553 | — | — | — | Nature Medicine, May 2025 |
Clinical Pulse
Verified clinician sentiment
Verified clinician reviews launching · Apply to contribute →
Reviews are signed by clinicians verified against their regulator's public register.
Inside the Auris+ Listing
Six more sections complete this device’s Auris+ Listing.
Inside the algorithm
ProPro unlocks the five-stage editorial walkthrough of how this device's AI reaches a verdict — model architecture, training data, inference latency, and where the clinician stays in the loop.
Decision Ledger
ProPro unlocks a private, cross-vendor log of every case you read on this device — what the AI called, what you concluded, and a one-line reason if you overrode it. Ready for the EU AI Act's deployer logging obligations when they land in 2028.
Clinical Evidence Deep Dive
ProPro unlocks the structured clinical-evidence summary — study count, target patient population, and a tabular accuracy-metrics view drawn from peer-reviewed sources.
Peer-Reviewed Publications
ProPro unlocks the curated peer-reviewed publication list with PubMed cross-links — the citation backbone of every editorial verdict.
Post-Market & Regulatory Conditions
ProPro unlocks the post-market surveillance summary, recall record, and the conditions of approval that bound real-world use.
AI Algorithm Version History
ProPro unlocks the chronological record of algorithm version changes — what changed when, drawn from manufacturer changelogs and regulatory filings.
Regulatory Approvals
Class IIb
K213857
Class II
DEN130045
Class II
Safety Record
No device recalls or Class I/II FDA safety communications have been identified for HeartFlow FFRCT as of the review date. The MAUDE database contains a small number of isolated adverse-event reports, including a false-negative case where FFRCT was reported negative but the patient later presented to the emergency department with chest pain, and one analyst-error case in which a distal left main segment was modelled larger than indicated by the CT data. These are characteristic of decision-support software AEs rather than systematic device failures. The PLATFORM trial confirmed no adverse events in 117 patients whose invasive angiography was deferred on the basis of FFRCT. The technology is image-analysis software only — it does not alter CT acquisitions, deliver therapy, or interact with implanted hardware — which substantially limits the harm surface.
Intended Use & Indications
HeartFlow FFRCT Analysis is a post-processing software service that ingests standard coronary CCTA DICOM images from clinically stable symptomatic patients with suspected coronary artery disease and generates a patient-specific 3-D computational fluid dynamics simulation of coronary blood flow under conditions of hyperaemia. The output is an FFRCT value at every point along the coronary tree, displayed as a colour-coded vessel map and a numeric FFRCT pull-back curve. An FFRCT ≤ 0.80 is consistent with haemodynamically significant ischaemia. The analysis does not require additional pharmacological stress, separate radiation exposure, or invasive access; it is performed by HeartFlow analysts and delivered electronically to the ordering clinician. The device is a decision-support tool; downstream revascularisation decisions remain with the treating physician.