Scale-specific auxiliary multi-task contrastive learning for deep liver vessel segmentation

TL;DR

This paper introduces a novel deep learning approach that uses scale-specific auxiliary tasks and contrastive learning to improve the segmentation of hepatic vessels across different scales, addressing the challenge of preserving complex vascular geometry.

Contribution

It proposes a new multi-task contrastive learning framework with scale decomposition for better multi-scale vessel segmentation in medical images.

Findings

Improved segmentation accuracy on 3D-IRCADb dataset

Effective discrimination between vessel scales

Enhanced representation of vascular tree geometry

Abstract

Extracting hepatic vessels from abdominal images is of high interest for clinicians since it allows to divide the liver into functionally-independent Couinaud segments. In this respect, an automated liver blood vessel extraction is widely summoned. Despite the significant growth in performance of semantic segmentation methodologies, preserving the complex multi-scale geometry of main vessels and ramifications remains a major challenge. This paper provides a new deep supervised approach for vessel segmentation, with a strong focus on representations arising from the different scales inherent to the vascular tree geometry. In particular, we propose a new clustering technique to decompose the tree into various scale levels, from tiny to large vessels. Then, we extend standard 3D UNet to multi-task learning by incorporating scale-specific auxiliary tasks and contrastive learning to encourage the discrimination between scales in the shared representation. Promising results, depicted in several evaluation metrics, are revealed on the public 3D-IRCADb dataset.