LU-Net: a multi-stage attention network to improve the robustness of segmentation of left ventricular structures in 2D echocardiography

Segmentation of cardiac structures is one of the fundamental steps to estimate volumetric indices of the heart. This step is still performed semi-automatically in clinical routine, and is thus prone to inter and intra observer variability. Recent studies have shown that deep learning has the potential to perform fully automatic segmentation. However, the current best solutions still suffer from a lack of robustness, in terms of accuracy and number of outliers. The goal of this work is to introduce a novel network designed to improve the overall segmentation accuracy of left ventricular structures (endocardial and epicardial borders) while enhancing the estimation of the corresponding clinical indices and reducing the number of outliers. This network is based on a multi-stage framework where both the localization and segmentation steps are optimized jointly through an end-to-end scheme. Results obtained on a large open access dataset show that our method outperforms the current best performing deep learning solution with a lighter architecture and achieved an overall segmentation accuracy lower than the intra observer variability for the epicardial border (i.e. on average a mean absolute error of 1.5 mm and a Hausdorff distance of 5.1 mm) with 11% of outliers. Moreover, we demonstrate that our method can closely reproduce the expert analysis for the end-diastolic and end-systolic left ventricular volumes, with a mean correlation of 0.96 and a mean absolute error of 7.6 ml. Concerning the ejection fraction of the left ventricle, results are more contrasted with a mean correlation coefficient of 0.83 and an absolute mean error of 5.0%, producing scores that are slightly below the intra observer margin. Based on this observation, areas for improvement are suggested.