Abstract:The accurate segmentation of brain tumor is significant in clinical practice. Convolutional Neural Network (CNN)-based methods have made great progress in brain tumor segmentation due to powerful local modeling ability. However, brain tumors are frequently pattern-agnostic, i.e. variable in shape, size and location, which can not be effectively matched by traditional CNN-based methods with local and regular receptive fields. To address the above issues, we propose a shape-scale co-awareness network (S2CA-Net) for brain tumor segmentation, which can efficiently learn shape-aware and scale-aware features simultaneously to enhance pattern-agnostic representations. Primarily, three key components are proposed to accomplish the co-awareness of shape and scale. The Local-Global Scale Mixer (LGSM) decouples the extraction of local and global context by adopting the CNN-Former parallel structure, which contributes to obtaining finer hierarchical features. The Multi-level Context Aggregator (MCA) enriches the scale diversity of input patches by modeling global features across multiple receptive fields. The Multi-Scale Attentive Deformable Convolution (MS-ADC) learns the target deformation based on the multiscale inputs, which motivates the network to enforce feature constraints both in terms of scale and shape for optimal feature matching. Overall, LGSM and MCA focus on enhancing the scale-awareness of the network to cope with the size and location variations, while MS-ADC focuses on capturing deformation information for optimal shape matching. Finally, their effective integration prompts the network to perceive variations in shape and scale simultaneously, which can robustly tackle the variations in patterns of brain tumors. The experimental results on BraTS 2019, BraTS 2020, MSD BTS Task and BraTS2023-MEN show that S2CA-Net has superior overall performance in accuracy and efficiency compared to other state-of-the-art methods. Code: https://github.com/jiangyu945/S2CA-Net.

Automatic Brain Tumor Segmentation Based on Cascaded Convolutional Neural Networks with Uncertainty Estimation

Shape-Scale Co-Awareness Network for 3D Brain Tumor Segmentation.

Automatic Brain Tumor Segmentation Using Cascaded Anisotropic Convolutional Neural Networks

Automatic Brain Tumor Segmentation By Exploring The Multi-Modality Complementary Information And Cascaded 3d Lightweight Cnns

Multi-step Cascaded Networks for Brain Tumor Segmentation

Brain Tumor Segmentation Using Automatic 3D Multi-channel Feature Selection Convolutional Neural Network

MRI Brain Tumor Segmentation and Uncertainty Estimation Using 3D-UNet Architectures

Brain Tumor Segmentation using 3D-CNNs with Uncertainty Estimation

Multiscale Cnns For Brain Tumor Segmentation And Diagnosis

TuNet: End-to-End Hierarchical Brain Tumor Segmentation Using Cascaded Networks

Automatic Segmentation of MRI of Brain Tumor Using Deep Convolutional Network

Automatic Segmentation Of Brain Tumor Using 3d Se-Inception Networks With Residual Connections

Segmenting Brain Tumor Using Cascaded V-Nets in Multimodal MR Images

Cascaded Coarse-to-Fine Neural Network for Brain Tumor Segmentation.

Cascaded Global Context Convolutional Neural Network for Brain Tumor Segmentation.

Asymmetric Ensemble of Asymmetric U-Net Models for Brain Tumor Segmentation With Uncertainty Estimation

3d Brain Tumor Image Segmentation Integrating Cascaded Anisotropic Fully Convolutional Neural Network And Hybrid Level Set Method

Automatic Semantic Segmentation of Brain Gliomas from MRI Images Using a Deep Cascaded Neural Network

Glioma Segmentation with Cascaded Unet

Brain tumour segmentation using cascaded 3D densely-connected U-net

Cascade Residual Multiscale Convolution and Mamba-Structured UNet for Advanced Brain Tumor Image Segmentation