Lung nodule detection based on 3D convolutional neural networks

Lei Fan, Zhaoqiang Xia, Xiaobiao Zhang, Xiaoyi Feng

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

24 Scopus citations

Abstract

With the increasing number of lung cancer patients, the CAD system is playing an increasingly important rule in the field of automatic identification for medical images. Since the 3D characteristics of low-dose CT images make the 3D convolution more suitable than 2D convolution, in this paper, we propose a method to detect lung nodule of lung CT images using 3D convolutional neural networks. Combined with the traditional morphological preprocessing methods, 3D convolutional neural networks are applied to lung CT images. The experimental accuracy indicates that this method is suitable for the problem of lung nodule detection and has great room to improve. The experimental results also demonstrate that the application of deep learning in the medical field will bring great progress for medical development.

Original languageEnglish
Title of host publicationConference Proceedings - 2017 International Conference on the Frontiers and Advances in Data Science, FADS 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages7-10
Number of pages4
ISBN (Electronic)9781538631485
DOIs
StatePublished - 1 Jul 2017
Event2017 International Conference on the Frontiers and Advances in Data Science, FADS 2017 - Xian, China
Duration: 23 Oct 201725 Oct 2017

Publication series

NameConference Proceedings - 2017 International Conference on the Frontiers and Advances in Data Science, FADS 2017
Volume2018-January

Conference

Conference2017 International Conference on the Frontiers and Advances in Data Science, FADS 2017
Country/TerritoryChina
CityXian
Period23/10/1725/10/17

Keywords

  • 3D convolutional network
  • CT lung image
  • deep learning
  • nodule detection

Fingerprint

Dive into the research topics of 'Lung nodule detection based on 3D convolutional neural networks'. Together they form a unique fingerprint.

Cite this