Adaptive cascaded transformer U-Net for MRI brain tumor segmentation

Bonian Chen; Qiule Sun; Yutong Han; Bin Liu; Jianxin Zhang; Qiang Zhang

doi:10.1088/1361-6560/ad4081

Adaptive cascaded transformer U-Net for MRI brain tumor segmentation

Phys Med Biol. 2024 Apr 18. doi: 10.1088/1361-6560/ad4081. Online ahead of print.

Authors

Bonian Chen¹, Qiule Sun², Yutong Han¹, Bin Liu³, Jianxin Zhang¹, Qiang Zhang⁴

Affiliations

¹ School of Computer Science and Engineering, Dalian Minzu University, No.18, Liaohe West Road, Jinzhou New District, Dalian, Liaoning, 116600, CHINA.
² Faculty of Electronic and Information Engineering, Dalian University of Technology, No.2, Lingong Road, Ganjinzi District, Dalian, China, Dalian, Liaoning, 116024, CHINA.
³ International School of Information Science and Engineering (DUT-RUISE), Dalian University of Technology, No.2, Lingong Road, Ganjinzi District, Dalian, China, Dalian, Liaoning, 116024, CHINA.
⁴ Faculty of Electronic and Information Engineering, Dalian University of Technology, No.2, Lingong Road, Ganjinzi District, Dalian, Dalian, Liaoning, 116024, CHINA.

PMID: 38636503
DOI: 10.1088/1361-6560/ad4081

Abstract

Objective: Brain tumor segmentation on magnetic resonance imaging (MRI) plays an important role in assisting the diagnosis and treatment of cancer patients. Recently, cascaded U-Net models have achieved excellent performance via conducting coarse-to-fine segmentation of MRI brain tumors. However, they still suffer from obvious global and local differences among various brain tumors, which are difficult to solve with conventional convolutions.

Approach: To address the issue, this work proposes a novel Adaptive Cascaded Transformer U-Net (ACTransU-Net) for MRI brain tumor segmentation, which simultaneously integrates Transformer and dynamic convolution into a single cascaded U-Net architecture to adaptively capture global information and local details of brain tumors. ACTransU-Net first cascades two 3D U-Nets into a two-stage network to segment brain tumors from coarse to fine. Subsequently, it integrates omni-dimensional dynamic convolution modules into the second-stage shallow encoder and decoder, thereby enhancing the local detail representation of various brain tumors through dynamically adjusting convolution kernel parameters. Moreover, 3D Swin-Transformer modules are introduced into the second-stage deep encoder and decoder to capture image long-range dependencies, which helps adapt the global representation of brain tumors.

Main results: Extensive experiment results evaluated on the public BraTS 2020 and BraTS 2021 brain tumor datasets demonstrate the effectiveness of ACTransU-Net, with average DSC of 84.96% and 91.37%, and HD95 of 10.81 mm and 7.31 mm, proving competitiveness with the state-of-the-art methods.

Significance: The proposed method focuses on adaptively capturing both global information and local details of brain tumors, aiding physicians in their accurate diagnosis. Additionally, it has the potential to extend ACTransU-Net for segmenting other types of lesions. The source code is available at: https://github.com/chenbn266/ACTransUnet.

Keywords: Transformer; U-Net; brain tumor segmentation; cascaded network; dynamic convolution; magnetic resonance imaging.