A Transformer-based Generative Adversarial Network for Brain Tumor Segmentation

TL;DR

This paper introduces a novel transformer-based GAN architecture for brain tumor segmentation in multi-modal MRI, combining transformer blocks with CNNs to improve global context learning and segmentation accuracy.

Contribution

It proposes a new transformer-based GAN model with a U-shaped generator and CNN discriminator, enhancing brain tumor segmentation performance over existing methods.

Findings

Achieved comparable or better results than state-of-the-art methods on BRATS2015 dataset.

Demonstrated the effectiveness of transformer blocks in medical image segmentation.

Validated the model's ability to learn global dependencies in MRI data.

Abstract

Brain tumor segmentation remains a challenge in medical image segmentation tasks. With the application of transformer in various computer vision tasks, transformer blocks show the capability of learning long-distance dependency in global space, which is complementary with CNNs. In this paper, we proposed a novel transformer-based generative adversarial network to automatically segment brain tumors with multi-modalities MRI. Our architecture consists of a generator and a discriminator, which are trained in min-max game progress. The generator is based on a typical "U-shaped" encoder-decoder architecture, whose bottom layer is composed of transformer blocks with resnet. Besides, the generator is trained with deep supervision technology. The discriminator we designed is a CNN-based network with multi-scale $L_{1}$ loss, which is proved to be effective for medical semantic image segmentation. To validate the effectiveness of our method, we conducted experiments on BRATS2015 dataset, achieving comparable or better performance than previous state-of-the-art methods.