Validation of aortic valve pressure gradient quantification using semi-automated 4D flow CMR pipeline.
Ciaran Grafton-ClarkePaul NjokuJean-Paul AbenLeon LedouxLiang ZhongJos WestenbergAndrew SwiftGareth ArcherJames WildRod HoseMarcus FlatherVassilios S VassiliouPankaj GargPublished in: BMC research notes (2022)
We enrolled 11 patients with severe AS on echocardiography from the EurValve programme. All patients had pre-intervention doppler echocardiography, invasive cardiac catheterisation with peak pressure drop assessment across the AV and 4D flow CMR. The peak AVPG was 51.9 ± 35.2 mmHg using the invasive pressure drop method and 52.2 ± 29.2 mmHg for the 4D flow CMR method (semi-automated pipeline), with good correlation between the two methods (r = 0.70, p = 0.017). Assessment of AVPG by 4D flow CMR using the novel semi-automated pipeline method shows excellent agreement to invasive assessment when compared to doppler-based methods and advocate for its use as complementary to echocardiography.
Keyphrases
- aortic valve
- left ventricular
- machine learning
- deep learning
- computed tomography
- high throughput
- pulmonary hypertension
- end stage renal disease
- ejection fraction
- aortic stenosis
- transcatheter aortic valve replacement
- randomized controlled trial
- chronic kidney disease
- newly diagnosed
- aortic valve replacement
- transcatheter aortic valve implantation
- blood flow
- prognostic factors
- clinical trial
- study protocol
- patient reported outcomes
- double blind