AI Science Tracker
EN

Hebei Yiling Hospital

1 discovery from 2026

Discoveries

Identifying a Critical Blind Spot: How Commercial AI (CAD) Systems Fail to Detect Faint Ground-Glass Opacities at -730 HU on Low-Dose CT.

Source check: Source checked. Not independent scientific confirmation.

Finding: Primary cohort false negatives were 13/100 and 14/100; missed GGOs averaged −737 and −727 HU. In the external cohort, missed-GGO means were −741 and −733 HU. Vendor A used a 3D CNN CAD system and Vendor B a hybrid machine-learning/deep-learning system; both processed LDCT cases against three-radiologist consensus. Testing: A separate retrospective 50-patient cohort and three-radiologist consensus reference corroborated failures; findings are limited to the two tested products and study sites.. Two commercial CT CAD systems missed 13–14% of reference nodules in a 100-case cohort; faint GGOs missed near −730 HU were also observed in a separate 50-case external cohort. This is a bounded failure-analysis result, not a general estimate for all systems. Source affiliation was not stated in the reviewed case checks and was unavailable in the local bibliographic cache.

Technology: Vendor A used a 3D CNN CAD system and Vendor B a hybrid machine-learning/deep-learning system; both processed LDCT cases against three-radiologist consensus.

Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
Vendor A used a 3D CNN CAD system and Vendor B a hybrid machine-learning/deep-learning system; both processed LDCT cases against three-radiologist consensus.
Test and result
A separate retrospective 50-patient cohort and three-radiologist consensus reference corroborated failures; findings are limited to the two tested products and study sites.
Limitation / still unknown
Two commercial CT CAD systems missed 13–14% of reference nodules in a 100-case cohort; faint GGOs missed near −730 HU were also observed in a separate 50-case external cohort. This is a bounded failure-analysis result, not a general estimate for all systems.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://www.mdpi.com/2075-4418/16/7/1014

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open the interactive profile for the timeline, or browse all organizations and model families.