Radiology · Pulmonary nodule detection and tracking on chest CT
Veye Lung Nodules
Aidence B.V. (DeepHealth, a RadNet company)
The interesting thing about Veye Lung Nodules is where the evidence comes from: nearly all of it is independent. A 2024 European Journal of Radiology analysis of 167 surgical patients found radiologists with Veye detected 92.4 percent of resected lung metastases against 80.4 percent unassisted; a Dutch teaching-hospital evaluation reported 88 percent sensitivity at about one false positive per scan; and the software sits inside the NHS England Targeted Lung Health Checks as its screening AI of choice, with a 2025 NIHR/NICE assessment counting five Veye studies. What holds it at one mark is the register: a single verified jurisdiction (EU MDR Class IIb), no FDA clearance, and no prospective outcome trial published.
Performance Metrics
Clinical Evidence
The peer-reviewed record is unusually independent for a commercial nodule-AI product. The strongest result is the 2024 European Journal of Radiology retrospective analysis of 167 patients undergoing metastasectomy (475 resected nodules): radiologists reading with Veye Lung Nodules v3.9.2 reached 92.4 percent [[sensitivity]] for lung metastases versus 80.4 percent unassisted (p < 0.001), and 46 of the 50 nodules absent from the preoperative radiology report proved to be metastases. The same study maps the failure modes honestly — cavitation and pleural contact were associated with AI misses, vascular contact with radiologist misses — a genuinely complementary error profile. A 2021 Clinical Radiology evaluation at a large Dutch teaching hospital (145 patients) reported 88 percent sensitivity, roughly one false positive per scan and a negative predictive value of 95 percent, while flagging a tendency to overestimate nodule size — a caution that matters when volume thresholds trigger follow-up. (The journal anonymises the vendor; the attribution to Veye rests on the manufacturer's product sheet and the study's inclusion in the NICE assessment's Veye evidence set.) A 2023 JAMIA qualitative study from Edinburgh (39 stakeholder interviews) found the tool easy to use with context-dependent performance. The 2025 NIHR/NICE systematic review and economic evaluation of lung-nodule AI counted five Veye studies (UK, Netherlands, US) and concluded, programme-wide, that AI-assisted reading may improve consistency of nodule measurement and management while the comparative evidence base remains immature. No prospective outcome trial of Veye has been published; the device-specific independent base — three primary studies and a synthesis — is real but small.
| Study | Design | n | Sensitivity | Specificity | AUC | Published |
|---|---|---|---|---|---|---|
| Independent retrospective analysis, 167 metastasectomy patients (Eur J Radiol) | RetrospectiveRetrospective | 167 | 92.4% (radiologist + AI) | not reported as spec | — | Eur J Radiol 2024; vs 80.4% unassisted (p<0.001); 475 resected nodules; Veye v3.9.2; cavitation/pleural contact linked to AI misses |
| Martins Jarnalo CO, et al. (independent, Dutch teaching hospital) | RetrospectiveRetrospective | 145 | 88% | NPV 95%, ~1 FP/scan | — | Clin Radiol 2021;76:838-45; flags nodule-size overestimation; vendor anonymised in-journal, attribution per manufacturer product sheet + NICE evidence set |
| Farič N, et al. (independent qualitative, Edinburgh; 39 interviews) | ExpertExpert review | 39 | n/a (qualitative) | n/a (qualitative) | — | J Am Med Inform Assoc 2023; early-integration experiences; easy to use, context-dependent performance |
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 algorithm
Editorial featureHow Veye reads a chest CT.
Five stages — from raw input to verdict — drawn from manufacturer documentation and the public regulatory record.
- INGEST
- NORMALISE
- DETECT
- LOCALISE
- VERDICT
Stage 01 · INGEST
The chest CT series arrives, with priors when available.
Veye receives the chest CT study from PACS or the imaging gateway. When a prior study exists, it is retrieved too — growth measurement across time is the product's defining output, because change, not size alone, drives nodule-management decisions.
Deployment is server-side; versions observed in the independent literature (v3.9.2) indicate a locked, versioned release lineage.
Input
Chest CT
Inference
Per-study
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
Safety Record
No field safety notice, recall or regulator safety communication specific to Veye Lung Nodules was found in public reporting as of July 2026 (the device has no US distribution, so FDA MAUDE is not an applicable channel; EU and MHRA sources could not be queried directly from this environment). The documented clinical cautions come from the independent literature: a tendency to overestimate nodule size (Clinical Radiology 2021), which can push borderline nodules over follow-up thresholds, and AI misses associated with cavitation and pleural contact (European Journal of Radiology 2024). The software is a concurrent/second reader — every finding passes through the reporting radiologist, and absence of a Veye mark is not an exclusion of malignancy.
Intended Use & Indications
Veye Lung Nodules analyses chest CT studies and marks suspected pulmonary nodules, measures their diameter and volume, characterises composition, and — given a prior study — registers the two series to compute growth and volume-doubling time, the quantities that drive nodule-management protocols such as the British Thoracic Society and Lung-RADS pathways. The product began life as Veye Chest (CE-marked 2019) and was re-architected under its current name; its certification was upgraded from Class IIa to Class IIb under the EU MDR in December 2022, with DEKRA Certification B.V. (notified body 0344) as certifier. It operates strictly as a [[second-reader]]-style aid: findings are presented to the reporting radiologist, who accepts, edits or rejects them. In England it is deployed across the Targeted Lung Health Checks programme, the NHS's national lung-cancer screening rollout, with a reporting module that prepopulates the programme's structured template. It has no US marketing authorisation: DeepHealth states that the US version of its chest portfolio excludes the Veye components.