Pixel-wise Segmentation of Right Ventricle of Heart

TL;DR

This paper introduces a deep learning method for precise, automated segmentation of the right ventricle in cardiac MRI images, achieving state-of-the-art accuracy without post-processing.

Contribution

It presents a novel adaptive loss function and a comprehensive analysis of segmentation architectures and techniques for improved biomedical image segmentation.

Findings

Achieved 0.86 Dice coefficient on RVSC-MICCAI 2012 dataset

Reduced Hausdorff distance to 6.73 mm

Demonstrated effectiveness of adaptive loss function

Abstract

One of the first steps in the diagnosis of most cardiac diseases, such as pulmonary hypertension, coronary heart disease is the segmentation of ventricles from cardiac magnetic resonance (MRI) images. Manual segmentation of the right ventricle requires diligence and time, while its automated segmentation is challenging due to shape variations and illdefined borders. We propose a deep learning based method for the accurate segmentation of right ventricle, which does not require post-processing and yet it achieves the state-of-the-art performance of 0.86 Dice coefficient and 6.73 mm Hausdorff distance on RVSC-MICCAI 2012 dataset. We use a novel adaptive cost function to counter extreme class-imbalance in the dataset. We present a comprehensive comparative study of loss functions, architectures, and ensembling techniques to build a principled approach for biomedical segmentation tasks.