What happened
The National Institute for Health and Care Excellence (NICE) assessed AI-derived software that detects and measures lung nodules on chest CT scans for a clinician to review. In five screening studies of four products, NICE's assessors found that review with the software was nearly always more sensitive but less specific than review without it. In all nine studies of reading time, review was faster with the software, though the comparisons were partly in laboratory-like conditions. The assessors also built an exploratory model. In the model's targeted screening scenario, the software with clinician review detected more people with nodules needing follow-up and slightly more lung cancers, at slightly lower cost. In guidance published on 5 July 2023, NICE let targeted screening centers use the software while they generate evidence. For scans done for other reasons, the software may flag more harmless nodule-like structures and lead to unneeded surveillance. NICE limited that use to research. Five of the thirteen products assessed had no published evidence.
Sources
-
Assessment report, by authors with no tie to the system.
In the EAG's model, for targeted screening, using CAD software alongside clinician review of CT scan images detected a larger number of people with nodules that needed follow up compared with clinician review alone. It also detected slightly more lung cancers. Using the software alongside clinician review was slightly less costly and more effective than clinician review alone.
-
Assessment report, by authors with no tie to the system.
National Institute for Health and Care Excellence. HTG687, section 1: Recommendations. 5 July 2023.
Centres using AI-derived CAD software alongside clinician review as part of targeted lung cancer screening should generate further evidence.
Details
AI system
- Workflow
-
A clinician reviewed every scan along with the software's marks. Targeted screening used low-dose CT scans without contrast, read by radiologists specialized in chest CT.
Follow-up
- Conditions for replication
-
The screened population has few other lung conditions that the software could mistake for nodules. The software fits into the radiologists' image archive and reporting workflow.
Entry details
- Sector
-
Healthcare (lung cancer screening).
- Purpose
-
Siemens Healthineers, Coreline Soft, Riverain Technologies, contextflow, Infervision, Arterys, Qure.ai, SenseTime, MeVis, Aidence and VUNO built their software to detect and measure lung nodules on chest CT scans for clinicians to review.
Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.
Cite this record
Peter H. Vartanian. (2026). Lung screening centers may use AI nodule detection under NICE guidance (AIBD-0736). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0736.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)