A Machine Learning-Based Approach for Automatic Coronary Sinus Vein Segmentation and Anatomy Reconstruction

The purpose of this study was to develop an approach for reconstructing the anatomy of coronary sinus veins using the proprietary cardiac computer tomography (
$CT$
) protocol. The method used in this study is based on a state-of-the-art neural network (NN) for medical image segmentation called Swin-Unet and the 5-fold cross-validation ensemble. The trained NNs were fuzzed using the mean ensemble and the final quality was estimated. The results of the study indicate that the neural network ensemble achieved a quality of 0.76 in terms of the dice score. Furthermore, the study confirmed the efficiency of the proposed cardiac
$CT$
protocol for the reconstruction of
$CS$
veins in patients with chronic HF and a reduced LV ejection fraction
$(LV\, EF < 40\%)$
.