External Attention Assisted Multi-Phase Splenic Vascular Injury Segmentation with Limited Data

TL;DR

This paper introduces a novel multi-phase splenic vascular injury segmentation framework that leverages external data and synthetic phase augmentation to improve accuracy with limited data, aiding clinical decision support.

Contribution

It proposes a new external attention mechanism using pseudo masks and a GAN-based synthetic phase augmentation to enhance segmentation performance with scarce data.

Findings

Outperforms baseline by over 7% in average DSC

Effectively leverages external data for attention guidance

Uses GANs for phase data augmentation

Abstract

The spleen is one of the most commonly injured solid organs in blunt abdominal trauma. The development of automatic segmentation systems from multi-phase CT for splenic vascular injury can augment severity grading for improving clinical decision support and outcome prediction. However, accurate segmentation of splenic vascular injury is challenging for the following reasons: 1) Splenic vascular injury can be highly variant in shape, texture, size, and overall appearance; and 2) Data acquisition is a complex and expensive procedure that requires intensive efforts from both data scientists and radiologists, which makes large-scale well-annotated datasets hard to acquire in general. In light of these challenges, we hereby design a novel framework for multi-phase splenic vascular injury segmentation, especially with limited data. On the one hand, we propose to leverage external data to mine pseudo splenic masks as the spatial attention, dubbed external attention, for guiding the segmentation of splenic vascular injury. On the other hand, we develop a synthetic phase augmentation module, which builds upon generative adversarial networks, for populating the internal data by fully leveraging the relation between different phases. By jointly enforcing external attention and populating internal data representation during training, our proposed method outperforms other competing methods and substantially improves the popular DeepLab-v3+ baseline by more than 7% in terms of average DSC, which confirms its effectiveness.