What is the clinical evidence from this study?

Study design: Cohort. Population: severe aortic stenosis (n=366). Intervention: CNN-based feature extraction of Doppler tracings (VGG-16) vs. Cluster 1 vs Cluster 2. Primary outcome: 2-year mortality (HR 2.6, 95% CI 1.4-5.1, p=0.004).

What does this research mean for the field?

CNN-based feature extraction of Doppler tracings identifies a distinct aortic outflow velocity profile that reflects left ventricular compensation capacity and is associated with increased 2-year mortality after transcatheter aortic valve replacement. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

April 22, 2022Open Access

Harnessing feature extraction capacities from a pre-trained convolutional neural network (VGG-16) for the unsupervised distinction of aortic outflow velocity profiles in patients with severe aortic stenosis

Key Result

CNN-based feature extraction of Doppler tracings identified a distinct aortic outflow velocity profile associated with increased 2-year mortality after TAVR (HR 2.6; 95% CI 1.4-5.1; P=0.004).

Study Design

Type

Cohort (n=366)

Structured PICO

Does feature extraction of Doppler tracings using a convolutional neural network predict 2-year mortality in patients with severe aortic stenosis undergoing TAVR?

Population

366 patients with severe aortic stenosis (AS) undergoing pre-procedural echocardiography and right heart catheterization prior to transcatheter aortic valve replacement (TAVR), mean age 79.8 ± 6.77 years, 39.9% women.

Intervention

Feature extraction of Doppler tracings using a pre-trained convolutional neural network (VGG-16) combined with principal component analysis, k-means clustering, and an extreme gradient boosting algorithm.

Outcome

2-year mortalityhard clinical

Transfer learning and unsupervised clustering of Doppler tracings can identify distinct aortic outflow velocity profiles that predict 2-year mortality after TAVR.

Main Result

Effect estimate: HR 2.6 (95% CI 1.4-5.1)

p-value: p=0.004

Abstract

Abstract Aims Hypothesizing that aortic outflow velocity profiles contain more valuable information about aortic valve obstruction and left ventricular contractility than can be captured by the human eye, features of the complex geometry of Doppler tracings from patients with severe aortic stenosis (AS) were extracted by a convolutional neural network (CNN). Methods and results After pre-training a CNN (VGG-16) on a large data set (ImageNet data set; 14 million images belonging to 1000 classes), the convolutional part was employed to transform Doppler tracings to 1D arrays. Among 366 eligible patients age: 79.8 ± 6.77 years; 146 (39.9%) women with pre-procedural echocardiography and right heart catheterization prior to transcatheter aortic valve replacement (TAVR), good quality Doppler tracings from 101 patients were analysed. The convolutional part of the pre-trained VGG-16 model in conjunction with principal component analysis and k-means clustering distinguished two shapes of aortic outflow velocity profiles. Kaplan–Meier analysis revealed that mortality in patients from Cluster 2 (n = 40, 39.6%) was significantly increased hazard ratio (HR) for 2-year mortality: 3; 95% confidence interval (CI): 1–8.9. Apart from reduced cardiac output and mean aortic valve gradient, patients from Cluster 2 were also characterized by signs of pulmonary hypertension, impaired right ventricular function, and right atrial enlargement. After training an extreme gradient boosting algorithm on these 101 patients, validation on the remaining 265 patients confirmed that patients assigned to Cluster 2 show increased mortality (HR for 2-year mortality: 2.6; 95% CI: 1.4–5.1, P-value: 0.004). Conclusion Transfer learning enables sophisticated pattern recognition even in clinical data sets of limited size. Importantly, it is the left ventricular compensation capacity in the face of increased afterload, and not so much the actual obstruction of the aortic valve, that determines fate after TAVR.