Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (7): 2117-2125.DOI: 10.11772/j.issn.1001-9081.2019112012

• Virtual reality and multimedia computing • Previous Articles     Next Articles

Pulmonary nodule segmentation method based on deep transfer learning

MA Jinlin, WEI Meng, MA Ziping   

  1. School of Computer Science and Engineering, North Minzu University, Yinchuan Ningxia 750021, China
  • Received:2019-11-28 Revised:2020-04-01 Online:2020-07-10 Published:2020-06-29
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61462002), the National Natural Science Foundation of Ningxia Province (2020AAC3215), the "Image and Intelligent Information Processing" Innovation Team Project of State Ethnic Affairs Commission, the Innovation Team Project of "Computer Vision and Virtual Reality" of North Minzu University, the Major Project of North Minzu University (2018XYZJK02).

基于深度迁移学习的肺结节分割方法

马金林, 魏萌, 马自萍   

  1. 北方民族大学 计算机科学与工程学院, 银川 750021
  • 通讯作者: 马金林
  • 作者简介:马金林(1976-),男,宁夏银川人,副教授,博士,主要研究方向:图像处理、智能信息处理;魏萌(1996-),女,陕西咸阳人,硕士研究生,主要研究方向:图像处理;马自萍(1977-),女,宁夏银川人,副教授,博士,主要研究方向:智能信息处理。
  • 基金资助:
    国家自然科学基金资助项目(61462002);宁夏自然科学基金资助项目(2020AAC3215);国家民委“图像与智能信息处理”创新团队项目;北方民族大学“计算机视觉与虚拟现实”创新团队项目;北方民族大学重大专项(ZDZX201801)。

Abstract: Focused on the issue that U-Net has a poor segmentation effect for small-volume pulmonary nodules, a segmentation method based on deep transfer learning was proposed, and Block Superimposed Fine-Tuning (BSFT) strategy was used to assist the segmentation of pulmonary nodules. Firstly, convolutional neural network was used to learn the feature information of large natural image datasets. Then, the learned features were transferred to the network for the segmentation of small pulmonary nodule image datasets. From the last sampling layer of the network, the network was released block by block and fine-tuned until the network completed the superimposition of the last layer. Finally, the similarity coefficient of Dice was quantitatively analyzed to determine the optimal segmentation network. The experimental results show that the Dice value of BSFT on LUNA16 pulmonary nodule open dataset reaches 0.917 9, which is obviously better than that of the mainstream pulmonary nodule segmentation algorithms.

Key words: pulmonary nodule segmentation, deep transfer learning, Convolutional Neural Network (CNN), U-Net segmentation network

摘要: 针对U-Net分割小体积肺结节效果较差的问题,提出一种基于深度迁移学习的分割方法,利用分块式叠加微调(BSFT)策略辅助分割肺结节。首先,利用卷积神经网络学习自然图像大数据集的特征信息;然后,将所学特征迁移到进行肺结节图像小数据集分割的网络,从该网络最后一个下采样层开始逐块释放、微调训练,直到网络完成最后一层的叠加;最后,定量分析Dice相似性系数,以确定最佳分割网络。实验结果表明,BSFT在LUNA16肺结节公开数据集上的Dice值达到0.917 9,该策略的性能明显优于主流肺结节分割算法。

关键词: 肺结节分割, 深度迁移学习, 卷积神经网络, U-Net分割网络

CLC Number: