Cleerly Coronary Analysis

Performance Metrics

Clinical Evidence

The published evidence base is substantial in volume and largely manufacturer-authored — a pattern worth carrying openly, since it shapes how much the diagnostic-accuracy figures can be leaned on. The strongest diagnostic-accuracy result is a CREDENCE trial substudy (Griffin et al., JACC: Cardiovascular Imaging 2022; n=303), which reported per-patient [[sensitivity]] and [[specificity]] of 94 percent and 68 percent for ≥50 percent stenosis against quantitative coronary angiography, with an [[npv]] of 90 percent — authors are Cleerly-affiliated. A companion analysis of CREDENCE patients (Lipkin et al., AJR 2022; n=301) found AI-QCT achieved an AUC of 0.88 for ≥50 percent stenosis versus 0.66 for stress myocardial perfusion imaging, with invasive angiography as reference — also manufacturer-authored. On prospective evidence, the CERTAIN study (Nurmohamed et al., European Heart Journal – Cardiovascular Imaging 2024) enrolled 750 consecutive patients across five expert CCTA sites in a cross-over design and found that AI-QCT was associated with a change in diagnosis or management in 57.1 percent of patients (p<0.001), with a 37.1 percent reduction in downstream testing after the ischaemia assessment; the authorship is again Cleerly-affiliated. A post-hoc analysis of the independent PACIFIC-1 cohort (Bernardo et al., Open Heart 2025; n=208) reported a per-patient AUC of 0.91 for AI-QCT versus 0.77 for expert readers (p<0.002) — manufacturer-authored, with independent site investigators. The foundational CREDENCE dataset paper (Stuijfzand et al., JAMA Cardiology 2020; n=612) is largely independent and established that comprehensive anatomical CCTA interpretation outperformed functional imaging against invasive FFR — the substrate the later Cleerly analyses were built on, rather than a test of the Cleerly software itself.

Clinical Pulse

Verified clinician sentiment

Verified clinician reviews launching soon · Apply to contribute →

Reviews are signed by clinicians verified against their regulator's public register.

Inside the Auris+ Listing

Regulatory Approvals

Safety Record

Two adverse-event reports were filed by the manufacturer to the FDA MAUDE database in 2024, both describing discrepancies between the stenosis severity reported by Cleerly on CCTA and a later invasive coronary angiogram — one case in which a patient declined recommended further testing, and one in which a patient later presented with acute chest pain and high-grade blockages. The manufacturer states both were submitted in an abundance of caution, and each notes it could not be determined whether the software underreported severity or the disease progressed in the interval. No FDA recall, enforcement action or safety communication for Cleerly was found in public reporting as of July 2026. The MAUDE detail pages could not be read directly from this environment; the case narratives above are corroborated through indexed copies. As adjunctive decision-support software, Cleerly does not deliver therapy or alter image acquisition, and the interpreting physician owns the verdict — a discrepancy between the software and invasive findings is a reason the reports exist, not a device malfunction.

Intended Use & Indications

Cleerly Labs is a web-based post-processing tool that ingests coronary CT angiography images from patients evaluated for coronary artery disease and produces a per-vessel account of atherosclerotic plaque volume, composition and stenosis severity. Cleerly ISCHEMIA, cleared as an adjunctive cardiovascular status indicator (21 CFR 870.2200), adds a machine-learned estimate of the functional significance of coronary lesions to support non-invasive assessment. Both products are adjunctive decision-support software per their indications: the outputs inform, but do not replace, the interpreting physician's judgment, and downstream diagnostic and treatment decisions remain with the treating clinician.