Towards Simultaneous Segmentation of Liver Tumors and Intrahepatic Vessels via Cross-attention Mechanism

TL;DR

This paper introduces UCA-Net, a novel 3D segmentation model using cross-attention mechanisms to simultaneously segment liver tumors and intrahepatic vessels, achieving state-of-the-art results on a new benchmark dataset.

Contribution

The paper presents a new 3D U-shaped network with cross-attention modules for improved simultaneous segmentation of liver tumors and vessels, filling a research gap.

Findings

UCA-Net outperforms existing methods in segmentation accuracy.

The cross-attention modules effectively reduce semantic gaps.

The dataset provides a new benchmark for liver tumor and vessel segmentation.

Abstract

Accurate visualization of liver tumors and their surrounding blood vessels is essential for noninvasive diagnosis and prognosis prediction of tumors. In medical image segmentation, there is still a lack of in-depth research on the simultaneous segmentation of liver tumors and peritumoral blood vessels. To this end, we collect the first liver tumor, and vessel segmentation benchmark datasets containing 52 portal vein phase computed tomography images with liver, liver tumor, and vessel annotations. In this case, we propose a 3D U-shaped Cross-Attention Network (UCA-Net) that utilizes a tailored cross-attention mechanism instead of the traditional skip connection to effectively model the encoder and decoder feature. Specifically, the UCA-Net uses a channel-wise cross-attention module to reduce the semantic gap between encoder and decoder and a slice-wise cross-attention module to enhance the contextual semantic learning ability among distinct slices. Experimental results show that the proposed UCA-Net can accurately segment 3D medical images and achieve state-of-the-art performance on the liver tumor and intrahepatic vessel segmentation task.