激光与光电子学进展, 2020, 57 (22): 221004, 网络出版: 2020-10-24
基于条件生成对抗网络的低级别胶质瘤MR图像分割 下载: 910次
Low-Grade Gliomas MR Image Segmentation Based on Conditional Generative Adversarial Networks
图像处理 低级别胶质瘤分割 条件生成对抗网络 深度学习 磁共振图像 U-Net image processing low-grade gliomas segmentation conditional generative adversarial networks deep learning magnetic resonance image U-Net
摘要
针对深度学习算法在脑肿瘤分割中存在标记数据不足的问题,提出了一种基于条件生成对抗网络(CGAN)的低级别胶质瘤(LGG)磁共振(MR)图像自动分割方法。首先,使用原始数据集训练CGAN并生成LGG图像以扩充原始数据集;然后,利用生成图像预训练分割网络;最后,在预训练模型的基础上训练分割模型。实验结果表明,相比常规数据扩充方法,本方法的Dice系数提高了4.39%,Jaccard指数提高了4.42%,为基于MR图像的LGG分割计算机铺助诊断系统提供了参考。
Abstract
In order to solve the problem that deep learning algorithm has insufficient labeled data in brain tumor segmentation, in this paper, an automatic segmentation method of low-grade gliomas (LGG) magnetic resonance (MR) images based on conditional generative adversarial networks (CGAN) is proposed. First, the original dataset is used to train the CGAN and generate LGG images to expand the original dataset. Then, the generated images are used to pre-train a segment network. Finally, the segmentation model is trained on the basis of the pre-training model. Experimental results show that compared with traditional data augmentation methods, the proposed method improves the Dice coefficient by 4.39% and Jaccard index by 4.42%. The method provides a reference for the development of a computer assisted diagnosis system for LGG segmentation based on MR images.
艾玲梅, 石康珍. 基于条件生成对抗网络的低级别胶质瘤MR图像分割[J]. 激光与光电子学进展, 2020, 57(22): 221004. Lingmei Ai, Kangzhen Shi. Low-Grade Gliomas MR Image Segmentation Based on Conditional Generative Adversarial Networks[J]. Laser & Optoelectronics Progress, 2020, 57(22): 221004.