Radiology · Vessel suppression and lung-nodule detection on chest CT
ClearRead CT
Riverain Technologies
ClearRead CT is one of the most broadly cleared chest-CT AI suites in the field — an FDA lineage stretching back to a 2016 vessel-suppression and detection clearance, an EU MDR certificate, active Health Canada licences and an Australian TGA clearance — and it is embedded across the US Veterans Affairs network and Department of Defense radiology. What keeps it at two marks rather than three is the shape of its evidence. The independent literature is real and consistent — multi-reader studies report roughly a fifth more nodules detected and about a fifth less reading time with vessel suppression — but it is reader-study and diagnostic-accuracy evidence, not a randomised trial, and the same independent work flags a high per-series false-positive rate for the suppressed images. The newest module, coronary-calcium scoring, has no peer-reviewed independent validation beyond conference material as of the review date.
Performance Metrics
Clinical Evidence
ClearRead CT's evidence base is unusually independent for a vendor imaging product, but it is built from reader and diagnostic-accuracy studies rather than randomised outcomes trials, and it should be read that way. The strongest independent study is a Zurich multi-reader evaluation (Academic Radiology 2021, PMID 32037256) in which six readers assessed 100 oncologic chest CTs with and without vessel-suppressed images: nodule detection rose by about 21 percent and reading time fell by about 21 percent with vessel suppression. The same study is candid about the trade-off — nearly every vessel-suppressed series carried at least one false-positive mark — which is the central caution for the suppression workflow. Independent corroboration comes from a Japanese reader study (Jpn J Radiol 2021, PMID 32940850; sensitivity rising from roughly 72 to 84 percent), a UK reader study (Clin Radiol 2021, PMID 34023068; sensitivity from roughly 64 to 82 percent), and a stand-alone CAD evaluation (J Med Syst 2019, PMID 30706143; sensitivity around 84 percent). The manufacturer-associated foundational study (AJR 2018, PMID 29336601) is a twelve-reader localisation-ROC study across 324 cases in which the mean LROC area under the curve for lung-cancer detection rose from 0.633 unaided to 0.773 with the vessel-suppression and CAD system, with the stand-alone system detecting about 89.5 percent of malignant nodules at 0.58 false positives per study. As vendor-associated work it establishes plausibility; the independent reader studies carry the catalogue judgment. The coronary-calcium module (ClearRead CT CAC) is a separate case. Its cleared claim is automated Agatston scoring on ungated chest CT, and a Mass General Brigham-led evaluation reported high agreement with ground-truth radiologists (about 96 percent) and with paired gated cardiac CT (about 90 percent) — but as of the review date that result is conference material (RSNA 2024) rather than a peer-reviewed publication, and no independent Agatston-accuracy paper for the module was located. Its performance claims should be treated as not yet peer-reviewed.
| Study | Design | n | Sensitivity | Specificity | AUC | Published |
|---|---|---|---|---|---|---|
| Zurich multi-reader study, 6 readers (independent) | RetrospectiveRetrospective | 100 | +21% detection | — | — | Academic Radiology, 2021; PMID 32037256; read time -21%; frequent false positives on VS series |
| Japanese reader study (independent) | RetrospectiveRetrospective | 0 | 72% → 84% | — | — | Jpn J Radiol, 2021; PMID 32940850 |
| UK reader study (independent) | RetrospectiveRetrospective | 0 | 64% → 82% | — | — | Clin Radiol, 2021; PMID 34023068 |
| Riverain-associated 12-reader LROC study | RetrospectiveRetrospective | 324 | 89.5% malignant (stand-alone) | 0.58 FP/study | LROC 0.633 → 0.773 | AJR, 2018; PMID 29336601; manufacturer-associated |
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
Five more sections complete this device’s Auris+ Listing.
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
112822
K242188
Class II
Safety Record
No ClearRead CT recalls or FDA safety communications were identified in publicly available sources as of July 2026. The MAUDE adverse-event database and the openFDA recall endpoint could not be queried directly from this environment, so this should be read as "none found in public reporting" rather than an exhaustive vigilance audit. The dominant documented risk is not a hardware failure but the false-positive behaviour of the vessel-suppression images — the Zurich reader study observed at least one false-positive mark on nearly every suppressed series — which places the burden of dismissing spurious marks on the reading radiologist. As an adjunctive concurrent-read tool that neither alters the original CT nor issues a diagnosis, a missed mark is not exclusionary and does not substitute for the standard read.
Intended Use & Indications
ClearRead CT is a Class II medical image management and processing system used as an adjunct to the radiologist's read of chest CT. The Vessel Suppress module produces a second image series in which pulmonary vasculature is digitally suppressed, improving the conspicuity of pulmonary nodules; the Detect module is a computer-aided detection (CADe) tool that marks candidate nodules; the Compare module registers a current study against a prior to support nodule follow-up. These modules require the original CT and are used concurrently with the standard read — they do not remove the study from the worklist, do not replace the radiologist's interpretation, and do not issue a diagnosis. A distinct product, ClearRead CT CAC, is cleared under a different regulation and product code to quantify coronary-artery calcium (an Agatston score) on non-contrast, non-gated chest CT for cardiovascular-risk assessment; it is a quantification tool, not the nodule pipeline, and its own evidence base is described separately below.