Cardiac Left Ventricle Segmentation using Recurrent U-Net and Level Set with Short Axis MRI Images
A. Karthik, Bojja Sharanya, Koorapati Reshma, Shashi Kiran Vunguturi · 2022 International Conference on Advances in Computing, Communication and Applied Informatics (ACCAI) · 2022
Semantic segmentation methods based on Deep Learning (DL). There are a number of techniques that exist to accomplish this task but Level Set (LS) consider to be the most popular one to perform MRI Segmentation. Also notice that the result of the segmentation results depends on two factors one is parameter initializations and set of iterations. To solve these problems and propose a hybrid method that combines classic LS methods with deep learning approaches (ie Recurrent U- Net Model) called Recurrent U-Net Level Set (RULS). First, a U-Net architecture allows for training. Next, the recurrent networks are used to allow to get best features to complete the segmentation of MRI images. At final, the level set method is used to refine the segmentation process. To test the segmentation result for Left Ventricle MRI images, real database is to be considered the MICCAI database.