End-to-end Learning of Convolutional Neural Net and Dynamic Programming for Left Ventricle Segmentation

TL;DR

This paper introduces an end-to-end learning framework combining CNN and dynamic programming for heart MRI segmentation, utilizing synthetic gradients to enable differentiability and improve accuracy with fewer labeled images.

Contribution

It presents a novel method integrating CNN and DP via synthetic gradients for end-to-end training in medical image segmentation.

Findings

End-to-end CNN and DP improve segmentation accuracy.

Fewer labeled images needed for effective training.

Significant accuracy gains over CNN alone.

Abstract

Differentiable programming is able to combine different functions or programs in a processing pipeline with the goal of applying end-to-end learning or optimization. A significant impediment is the non-differentiable nature of some algorithms. We propose to use synthetic gradients (SG) to overcome this difficulty. SG uses the universal function approximation property of neural networks. We apply SG to combine convolutional neural network (CNN) with dynamic programming (DP) in end-to-end learning for segmenting left ventricle from short axis view of heart MRI. Our experiments show that end-to-end combination of CNN and DP requires fewer labeled images to achieve a significantly better segmentation accuracy than using only CNN.