Deployment and validation of an AI system for detecting abnormal chest radiographs in clinical settings.
Source check: Source checked. Not independent scientific confirmation.
Finding / tested result: The deployed VinDr-CXR system was compared with reports at Phu Tho General Hospital; it achieved F1 0.653 (95% CI 0.635–0.671), accuracy 79.6%, sensitivity 68.6% and specificity 83.9% for any abnormality. This establishes real-world performance with a substantial drop from laboratory results.
Technology: CAD combines convolutional image models for abnormality detection/classification within the hospital PACS workflow.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- CAD combines convolutional image models for abnormality detection/classification within the hospital PACS workflow.
- Limitation / still unknown
- Deployment was assessed at one hospital against local reports; performance was lower than laboratory results and does not establish performance at other sites or improved outcomes.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://www.frontiersin.org/journals/digital-health/articles/10.3389/fdgth.2022.890759/full
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.