Detection and segmentation of the Left Ventricle in Cardiac MRI using Deep Learning
This addresses a time-consuming and variable task for cardiac diagnosticians, but the work is incremental as it builds on existing methods.
The paper tackles automating left ventricle segmentation in cardiac MRI to replace tedious manual delineation, achieving improved accuracy through a combined deep learning approach.
Manual segmentation of the Left Ventricle (LV) is a tedious and meticulous task that can vary depending on the patient, the Magnetic Resonance Images (MRI) cuts and the experts. Still today, we consider manual delineation done by experts as being the ground truth for cardiac diagnosticians. Thus, we are reviewing the paper - written by Avendi and al. - who presents a combined approach with Convolutional Neural Networks, Stacked Auto-Encoders and Deformable Models, to try and automate the segmentation while performing more accurately. Furthermore, we have implemented parts of the paper (around three quarts) and experimented both the original method and slightly modified versions when changing the architecture and the parameters.