Deep reconstructed coronary MRI is compared with CT angiography
Source check: Source checked. Not independent scientific confirmation.
Finding: Adaptive-CS-Net reconstruction enabled noncontrast coronary MR angiography in a paired 64-patient study using CT angiography as the reference. Patient-level sensitivity and specificity for significant coronary narrowing were 92.0% and 84.6%, respectively. These diagnostic measurements came from a selected single-centre cohort; they do not establish replacement of CT angiography or improved patient outcomes.
Technology: Adaptive-CS-Net; convolutional neural network
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- Adaptive-CS-Net combines learned CNN image representations with compressed-sensing reconstruction to make accelerated coronary MRI interpretable for stenosis assessment.
- Test and result
- New paired coronary MRI and CT examinations, independently interpreted with CT angiography as the stenosis reference.
- Limitation / still unknown
- Larger external diagnostic studies and clinical outcome benefit remain unestablished; prospective acquisition alone does not prove prospective frozen-model outcome prediction.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.1002/jmri.28653
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: 21 Nov 2026. Review scope: Primary relevant Methods/results/limitations and exact front/canonical identity; current primary PubMed linked correction/retraction metadata checked..
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 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.