激光与光电子学进展, 2021, 58 (8): 0810020, 网络出版: 2021-04-12
一种级联改进U-Net网络的脑肿瘤分割方法 下载: 1265次
A Method for Brain Tumor Segmentation Using Cascaded Modified U-Net
图像处理 脑肿瘤分割 多层特征融合 空洞卷积 条件随机场 image processing gliomas segmentation multi-levels feature fusing dilated convolution conditional random fields
摘要
提出了一种基于深度学习的3D脑肿瘤核磁共振图像(MRI)自动分割方法。为了降低分割难度,采用三级级联网络的策略分割脑肿瘤的三个子区域;为了进一步提高三维分割的精度,采用帧间卷积和帧内卷积,加入额外的多层特征融合机制和空洞卷积;为了进一步细化分割结果,将条件随机场构建的循环神经网络整合到网络结构中。在模型训练中结合了两种损失函数,进一步提高了准确率。该方法在BraTS 2018 数据集上进行验证,对于脑肿瘤整体、肿瘤核以及增强肿瘤,其分割结果的Dice系数分别达到了0.9093、0.8254 和 0.7855,Hausdorff距离达到3.8188、7.8487和4.3264,优于大多数脑肿瘤图像分割方法。
Abstract
We proposed a deep learning approach for automatic segmentation of three-dimensional gliomas magnetic resonance images(MRI). First, we used a three-stage cascaded strategy to sequentially segment the subregions of gliomas. Second, to further improve the segmentation accuracy, we used intraslice and interslice convolutions, introduced additional multi-levels feature fusing, and implemented dilated convolution. Third, to produce a fine-grained output, conditional random fields as recurrent neural network were adopted as a part of network structure. Finally, we combined two types of loss functions in the training procedure to further improve the segmentation accuracy. We applied our method on the BraTS 2018 dataset and achieved a Dice score of 0.9093, 0.8254, and 0.7855 and the Hausdorff distance of 3.8188, 7.8487, and 4.3264 for the whole tumor, tumor core, and enhanced tumor, respectively. The proposed methods achieved better performance than most brain tumor segmentation methods.
褚晶辉, 黄凯隆, 吕卫. 一种级联改进U-Net网络的脑肿瘤分割方法[J]. 激光与光电子学进展, 2021, 58(8): 0810020. Jinghui Chu, Kailong Huang, Wei Lü. A Method for Brain Tumor Segmentation Using Cascaded Modified U-Net[J]. Laser & Optoelectronics Progress, 2021, 58(8): 0810020.