Using deep learning method to identify left ventricular hypertrophy on echocardiography

Abstract

Background: Left ventricular hypertrophy (LVH) is an independent prognostic factor for cardiovascular events and it can be detected by echocardiography in the early stage. In this study, we aim to develop a semi-automatic diagnostic network based on deep learning algorithms to detect LVH.

Methods: We retrospectively collected 1610 transthoracic echocardiograms, included 724 patients [189 hypertensive heart disease (HHD), 218 hypertrophic cardiomyopathy (HCM), and 58 cardiac amyloidosis (CA), along with 259 controls]. The diagnosis of LVH was defined by two experienced clinicians. For the deep learning architecture, we introduced ResNet and U-net++ to complete classification and segmentation tasks respectively. The models were trained and validated independently. Then, we connected the best-performing models to form the final framework and tested its capabilities.

Results: In terms of individual networks, the view classification model produced AUC = 1.0. The AUC of the LVH detection model was 0.98 (95% CI 0.94-0.99), with corresponding sensitivity and specificity of 94.0% (95% CI 85.3-98.7%) and 91.6% (95% CI 84.6-96.1%) respectively. For etiology identification, the independent model yielded good results with AUC = 0.90 (95% CI 0.82-0.95) for HCM, AUC = 0.94 (95% CI 0.88-0.98) for CA, and AUC = 0.88 (95% CI 0.80-0.93) for HHD. Finally, our final integrated framework automatically classified four conditions (Normal, HCM, CA, and HHD), which achieved an average of AUC 0.91, with an average sensitivity and specificity of 83.7% and 90.0%.

Conclusion: Deep learning architecture has the ability to detect LVH and even distinguish the latent etiology of LVH.

Keywords: Deep learning; Echocardiography; Left ventricular hypertrophy.

Fig.1

The architecture of the integrated framework. ^aLeft ventricular hypertrophy, ^bhypertrophic cardiomyopathy, ^ccardiac amyloidosis ^dhypertensive heart disease

Fig. 2

Basic architecture of LVH detection and etiology classification models. This figure shows the inner architecture of the green block in the Fig. 1. ^aparasternal long-axis, ^bapical four-chamber

Fig. 3

The results of individual classification network. LVH left ventricular hypertrophy, HCM hypertrophic cardiomyopathy, CA cardiac amyloidosis, HHD hypertensive heart disease

Fig. 4

The masks created by manual annotation and auto-segmentation, and the newly generated images of left ventricular myocardium. ^acardiac amyloidosis, ^bhypertrophic cardiomyopathy, ^chypertensive heart disease

Fig. 5

The results of classification networks on automatically or manually segmented images. LVH left ventricular hypertrophy, HCM hypertrophic cardiomyopathy, CA cardiac amyloidosis, HHD hypertensive heart disease

Fig. 6

The results of the final integrated framework. HCM hypertrophic cardiomyopathy, CA cardiac amyloidosis, HHD hypertensive heart disease

Fig. 7

Class activation maps. The heatmaps created by Grad-CAM method highlight the regions which mainly influence the decisions made by the network