Attention-Enhanced Hybrid Feature Aggregation Network for 3D Brain Tumor Segmentation
Utilizing a hybrid U-Net-shaped model with attention-guided features, the GLIMS approach enhances 3D brain tumor segmentation performance.
Brain Tumor Segmentation Using Deep Learning and Attention Mechanisms
Automated brain tumor segmentation using deep learning and attention mechanisms improves accuracy and reduces computation.
Optimizing Brain Tumor Segmentation Through CNN U-Net with Hybrid Image Enhancement Techniques
Hybrid image enhancement techniques, combining Histogram Equalization (HE) and Contrast Limited Adaptive Histogram Equalization (CLAHE), consistently outperform individual methods in improving the accuracy of CNN-based brain tumor segmentation using the U-Net architecture.
LATUP-Net: A Lightweight 3D Attention U-Net with Parallel Convolutions for Efficient Brain Tumor Segmentation
LATUP-Net, a lightweight 3D U-Net variant, incorporates parallel convolutions and attention mechanisms to achieve high brain tumor segmentation performance with significantly reduced computational costs compared to state-of-the-art models.
Re-DiffiNet: Improving Brain Tumor Segmentation by Modeling Discrepancies with Diffusion Models
By explicitly modeling the discrepancy between the outputs of a segmentation model like U-Net and the ground truth using Denoising Diffusion Probabilistic Models (DDPMs), Re-DiffiNet can improve brain tumor segmentation performance, especially on boundary-distance metrics like Hausdorff Distance.
SEDNet: A Computationally Efficient Shallow Encoder-Decoder Network for Accurate Brain Tumor Segmentation
The proposed SEDNet architecture, with its shallow encoder-decoder design and selective skip paths, achieves impressive brain tumor segmentation performance while being computationally efficient for real-time clinical diagnosis.
Enhancing Brain Tumor Segmentation with Multiscale Attention and Omni-Dimensional Dynamic Convolution in nnU-Net
The proposed algorithm utilizes a modified nnU-Net architecture with multiscale attention and Omni-Dimensional Dynamic Convolution (ODConv3D) layers to enhance the accuracy of brain tumor segmentation across diverse datasets, including Brain Metastases and BraTS-Africa.
Automated and Editable Prompt Learning (AEPL) for Enhanced Brain Tumor Segmentation Using MRI
Integrating tumor grade prediction as automatically generated and editable prompts in a multi-task learning framework improves the accuracy and flexibility of brain tumor segmentation in MRI.