Improvements in Brain Tumor Segmentation Methods Based on Convolutional Neural Networks

Yuzhuo Li, Lihong Zhang, Yingbo Liang, Chongxin Xu, Tong Liu — Fri, 13 Dec 2024 00:00:00 +0000

Convolutional Neural Networks (CNNs) have emerged as a prominent research area in deep learning in recent years. U-Net, an essential model within CNNs, has gradually become a research focus in the field of medical image segmentation due to its remarkable segmentation performance. This paper presents a comprehensive overview of brain tumor segmentation methods based on CNNs. Firstly, it introduces common medical image datasets in the field of brain tumor segmentation. Secondly, it offers detailed reviews on the common improvements to 2D U-Net, 3D U-Net, and improvements based on other CNNs for brain tumor segmentation. Finally, it discusses the future development directions of CNNs for brain tumor segmentation.

EAI Endorsed Transactions on e-Learning

Improvements in Brain Tumor Segmentation Methods Based on Convolutional Neural Networks