In vivo comparison of key quantitative parameters measured with 3D peripheral angiography, 2D peripheral quantitative angiography and intravascular ultrasound.
Wojciech ZasadaMagdalena SlezakElzbieta PociaskKrzysztof Piotr MalinowskiKlaudia ProniewskaPiotr BuszmanKrzysztof MilewskiJuan F GranadaGrzegorz L KaluzaPublished in: The international journal of cardiovascular imaging (2019)
The aim of this study was to compare the measures of luminal stenosis between the two-dimensional (2D) and three-dimensional (3D) Quantitative Vessel Analysis (QVA) generated by CAAS QVA software and intravascular ultrasound (IVUS). Invasive contrast angiography is considered gold standard for diagnostic imaging and intervention in both coronary and peripheral arterial disease. However, it is based on 2D images depicting complicated 3D arterial anatomy. To overcome these limitations, 3D QVA has been developed to bridge the gap between 2D QVA and endovascular imaging. Thirty porcine femoral angiograms (common, profunda and superficial) with matching intravascular ultrasound (IVUS) pullbacks featuring variable degree of stenosis were analysed by 2D QVA, 3D QVA and quantitative IVUS. All 3 modalities provided similar data regarding the length of the investigated segment. Median lumen diameter was nearly identical in IVUS (4.69 mm) and in 3D QVA (4.76 mm) but quite a bit lower in 2D QVA (4.47 mm, Kruskal-Wallis test p = 0.1648). Lumen area measured in 2D QVA was lower than in IVUS and in 3D QVA. Lumen areas rendered by IVUS and 3D QVA were similar. Bland-Altman plots showed that the lowest differences were observed between IVUS and 3D QVA. IVUS and 3D QVA results were consistently higher than 2D QVA. 3D QVA is a useful surrogate of IVUS for precise luminal morphology measurements of peripheral arteries, rendering results that are much closer to IVUS than 2D QVA can provide.
Keyphrases
- high resolution
- optical coherence tomography
- coronary artery
- ultrasound guided
- magnetic resonance imaging
- computed tomography
- randomized controlled trial
- magnetic resonance
- deep learning
- machine learning
- artificial intelligence
- big data
- aortic valve
- optic nerve
- transcatheter aortic valve replacement
- aortic stenosis