Automatic Left Ventricular Cavity Segmentation via Deep Spatial Sequential Network in 4D Computed Tomography Studies

TL;DR

This paper introduces a novel deep learning method that leverages spatial and sequential information in 4D cardiac CT and MRI images to improve the accuracy and consistency of left ventricular cavity segmentation across all cardiac phases.

Contribution

The proposed spatial-sequential network with bi-directional learning effectively captures deformation and motion, outperforming existing methods in LVC segmentation on 4D cardiac imaging datasets.

Findings

Outperforms existing methods in CT datasets

Demonstrates generalizability to MRI datasets

Improves segmentation accuracy during end-systole phase

Abstract

Automated segmentation of left ventricular cavity (LVC) in temporal cardiac image sequences (multiple time points) is a fundamental requirement for quantitative analysis of its structural and functional changes. Deep learning based methods for the segmentation of LVC are the state of the art; however, these methods are generally formulated to work on single time points, and fails to exploit the complementary information from the temporal image sequences that can aid in segmentation accuracy and consistency among the images across the time points. Furthermore, these segmentation methods perform poorly in segmenting the end-systole (ES) phase images, where the left ventricle deforms to the smallest irregular shape, and the boundary between the blood chamber and myocardium becomes inconspicuous. To overcome these limitations, we propose a new method to automatically segment temporal cardiac images where we introduce a spatial sequential (SS) network to learn the deformation and motion characteristics of the LVC in an unsupervised manner; these characteristics were then integrated with sequential context information derived from bi-directional learning (BL) where both chronological and reverse-chronological directions of the image sequence were used. Our experimental results on a cardiac computed tomography (CT) dataset demonstrated that our spatial-sequential network with bi-directional learning (SS-BL) method outperformed existing methods for LVC segmentation. Our method was also applied to MRI cardiac dataset and the results demonstrated the generalizability of our method.