Heart failure & cardiomyopathiesResearch article
Artificial Intelligence-Assisted cardiovascular magnetic resonance (CMR) Scanning vs Standard-of-Care
- nonrandomized comparison
- tertiary-care center
- consecutive patients
Authors' conclusion
In this nonrandomized comparison, AIA improved scan time, scan time consistency, image quality, and 90-day scan failure across top CMR indications, compared to ODS.
Study context
- Population
- Patients undergoing CMR at a tertiary-care center, 2020 to 2024consecutive patients, 6,545 studies · AI-assisted (AIA) studies 1,635 · ODS studies 4,910
- Participants
- 6,080 patients
- Comparison
- Artificial intelligence (AI)-assisted vs conventional, operator-directed CMR scanning
- AIAn = 1,635
- ODSn = 4,910
Main finding
As reported in the abstract
Scan time
- AIA vs ODS
- reduction in average scan time 23%; P < 0.001
- AIA vs ODS
- reduction in coefficient of variation of scan time 31%
- Reduction consistent across indications and technologists' experience; significant after adjustment
- P < 0.0001
Further findings
Scan failure leading to repeat CMR · 90-day · lower in AIA
P < 0.001
| Outcome | AIA | ODS | P value |
|---|---|---|---|
| Scan quality score (5-point scale) | |||
| cine | 4.09 ± 0.8 | 4.03 ± 0.8 | |
| late gadolinium enhancement | 4.08 ± 0.8 | 3.98 ± 0.7 | |
| perfusion | 4.28 ± 0.6 | 4.14 ± 0.6 | |
| Patient survival · median follow-up of 29 months · did not differ | log-rank P = 0.500 |
- Scan quality score (5-point scale): P < 0.001 for each of cine, late gadolinium enhancement and perfusion; higher in AIA
AIA uptake · by 2023
- AIA use after introduction in 2021
- 78%
Original publication
Artificial Intelligence-Assisted CMR Scanning vs Standard-of-Care. JACC Adv. doi:10.1016/j.jacadv.2026.102677
Summarised from the published abstract; every number is checked against it