Unified Attentional Generative Adversarial Network for Brain Tumor Segmentation From Multimodal Unpaired Images

TL;DR

This paper introduces UAGAN, a novel model that performs unpaired multimodal image translation and segmentation for brain tumors, overcoming the need for registered image pairs in clinical settings.

Contribution

The proposed UAGAN enables simultaneous multimodal image translation and segmentation using unpaired data, incorporating attention mechanisms for improved feature extraction.

Findings

UAGAN outperforms existing methods in three-modality brain tumor segmentation.

The model effectively captures modality-invariant features for accurate segmentation.

Attention blocks enhance segmentation performance by focusing on relevant features.

Abstract

In medical applications, the same anatomical structures may be observed in multiple modalities despite the different image characteristics. Currently, most deep models for multimodal segmentation rely on paired registered images. However, multimodal paired registered images are difficult to obtain in many cases. Therefore, developing a model that can segment the target objects from different modalities with unpaired images is significant for many clinical applications. In this work, we propose a novel two-stream translation and segmentation unified attentional generative adversarial network (UAGAN), which can perform any-to-any image modality translation and segment the target objects simultaneously in the case where two or more modalities are available. The translation stream is used to capture modality-invariant features of the target anatomical structures. In addition, to focus on segmentation-related features, we add attentional blocks to extract valuable features from the translation stream. Experiments on three-modality brain tumor segmentation indicate that UAGAN outperforms the existing methods in most cases.