Brain Tumor Segmentation
Brain tumor segmentation involves automatically identifying and outlining tumor regions in medical images, primarily MRI scans, to aid in diagnosis and treatment planning. Current research focuses on improving segmentation accuracy and robustness using advanced deep learning architectures like U-Net and its variants (e.g., Swin UNETR, nnU-Net), often incorporating attention mechanisms and multi-scale feature extraction to better handle the complex heterogeneity of brain tumors. These advancements are crucial for improving the speed and accuracy of clinical diagnosis, facilitating personalized treatment strategies, and potentially leading to better patient outcomes. Furthermore, significant effort is dedicated to addressing challenges like missing modalities and imbalanced datasets.
Papers
Memory Consistent Unsupervised Off-the-Shelf Model Adaptation for Source-Relaxed Medical Image Segmentation
Xiaofeng Liu, Fangxu Xing, Georges El Fakhri, Jonghye Woo
Hybrid Window Attention Based Transformer Architecture for Brain Tumor Segmentation
Himashi Peiris, Munawar Hayat, Zhaolin Chen, Gary Egan, Mehrtash Harandi
Region-Based Evidential Deep Learning to Quantify Uncertainty and Improve Robustness of Brain Tumor Segmentation
Hao Li, Yang Nan, Javier Del Ser, Guang Yang
PA-Seg: Learning from Point Annotations for 3D Medical Image Segmentation using Contextual Regularization and Cross Knowledge Distillation
Shuwei Zhai, Guotai Wang, Xiangde Luo, Qiang Yue, Kang Li, Shaoting Zhang