Prospective deep learning–based quantitative assessment of coronary plaque by computed tomography angiography compared with intravascular ultrasound: the REVEALPLAQUE study
Authors' conclusion
AI-enabled CCTA quantification and characterization of atherosclerosis demonstrated strong agreement with IVUS reference standard measurements.
Study context
- Population
- Clinically stable patients with known CAD, 15 centres in the US and Japan
- Participants
- 237 patients · 432 lesions
- Comparison
- Artificial intelligence (AI)-enabled coronary computed tomography angiography (CCTA) plaque analysis vs intravascular ultrasound (IVUS)
- Outcome
- Agreement of AI-derived CCTA plaque volumes (TPV, vessel, lumen, CP, NCP, low-attenuation plaque (LAP)) with IVUS
Methodological detailsprospective, multicentre, international, blinded, core laboratory-adjudicated
- Prospective
- Multicentre
- International
- Blinded
- Core laboratory-adjudicated
Performance
As reported in the abstract
Correlation of AI-CCTA with IVUS plaque volumes
TPV: correlation coefficient, slope, Y intercept
0.91, 0.99, 1.87
Calcified plaque (CP) volume: correlation coefficient, slope, Y intercept
0.91, 1.05, 5.32
Non-calcified plaque (NCP) volume: correlation coefficient, slope, Y intercept
0.87, 0.98, 15.24
Original publication
Prospective deep learning–based quantitative assessment of coronary plaque by computed tomography angiography compared with intravascular ultrasound: the REVEALPLAQUE study. Eur Heart J Cardiovasc Imaging. Published 3 May 2024. doi:10.
Summarised from the published abstract; every number is checked against it