A Topology-Attention ConvLSTM Network and Its Application to EM Images

TL;DR

This paper introduces TACNet, a novel 3D image segmentation network that leverages topology-aware modules and ConvLSTM to improve structural accuracy in biomedical EM images.

Contribution

The paper presents a new Topology-Attention ConvLSTM Network with innovative modules for topology preservation in 3D segmentation tasks.

Findings

Outperforms baseline methods in topology-aware metrics

Effectively captures contextual information across slices

Provides stable topology-critical maps for segmentation

Abstract

Structural accuracy of segmentation is important for finescale structures in biomedical images. We propose a novel TopologyAttention ConvLSTM Network (TACNet) for 3D image segmentation in order to achieve high structural accuracy for 3D segmentation tasks. Specifically, we propose a Spatial Topology-Attention (STA) module to process a 3D image as a stack of 2D image slices and adopt ConvLSTM to leverage contextual structure information from adjacent slices. In order to effectively transfer topology-critical information across slices, we propose an Iterative-Topology Attention (ITA) module that provides a more stable topology-critical map for segmentation. Quantitative and qualitative results show that our proposed method outperforms various baselines in terms of topology-aware evaluation metrics.