Design of lung nodules segmentation and recognition algorithm based on deep learning

Abstract Background Accurate segmentation and recognition algorithm of lung nodules has great important value of reference for early diagnosis of lung cancer. An algorithm is proposed for 3D CT sequence images in this paper based on 3D Res U-Net segmentation network and 3D ResNet50 classification ne...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Hui Yu, Jinqiu Li, Lixin Zhang, Yuzhen Cao, Xuyao Yu, Jinglai Sun
Formato:	article
Lenguaje:	EN
Publicado:	BMC 2021
Materias:	Lung nodule Convolutional neural network U-Net Residual learning Image segmentation Image classification Computer applications to medicine. Medical informatics R858-859.7 Biology (General) QH301-705.5
Acceso en línea:	https://doaj.org/article/ea95d1654c53428fa1eb5fc5cc931d56
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:ea95d1654c53428fa1eb5fc5cc931d56
record_format	dspace
spelling	oai:doaj.org-article:ea95d1654c53428fa1eb5fc5cc931d562021-11-14T12:13:03ZDesign of lung nodules segmentation and recognition algorithm based on deep learning10.1186/s12859-021-04234-01471-2105https://doaj.org/article/ea95d1654c53428fa1eb5fc5cc931d562021-11-01T00:00:00Zhttps://doi.org/10.1186/s12859-021-04234-0https://doaj.org/toc/1471-2105Abstract Background Accurate segmentation and recognition algorithm of lung nodules has great important value of reference for early diagnosis of lung cancer. An algorithm is proposed for 3D CT sequence images in this paper based on 3D Res U-Net segmentation network and 3D ResNet50 classification network. The common convolutional layers in encoding and decoding paths of U-Net are replaced by residual units while the loss function is changed to Dice loss after using cross entropy loss to accelerate network convergence. Since the lung nodules are small and rich in 3D information, the ResNet50 is improved by replacing the 2D convolutional layers with 3D convolutional layers and reducing the sizes of some convolution kernels, 3D ResNet50 network is obtained for the diagnosis of benign and malignant lung nodules. Results 3D Res U-Net was trained and tested on 1044 CT subcases in the LIDC-IDRI database. The segmentation result shows that the Dice coefficient of 3D Res U-Net is above 0.8 for the segmentation of lung nodules larger than 10 mm in diameter. 3D ResNet50 was trained and tested on 2960 lung nodules in the LIDC-IDRI database. The classification result shows that the diagnostic accuracy of 3D ResNet50 is 87.3% and AUC is 0.907. Conclusion The 3D Res U-Net module improves segmentation performance significantly with the comparison of 3D U-Net model based on residual learning mechanism. 3D Res U-Net can identify small nodules more effectively and improve its segmentation accuracy for large nodules. Compared with the original network, the classification performance of 3D ResNet50 is significantly improved, especially for small benign nodules.Hui YuJinqiu LiLixin ZhangYuzhen CaoXuyao YuJinglai SunBMCarticleLung noduleConvolutional neural networkU-NetResidual learningImage segmentationImage classificationComputer applications to medicine. Medical informaticsR858-859.7Biology (General)QH301-705.5ENBMC Bioinformatics, Vol 22, Iss S5, Pp 1-21 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Lung nodule Convolutional neural network U-Net Residual learning Image segmentation Image classification Computer applications to medicine. Medical informatics R858-859.7 Biology (General) QH301-705.5
spellingShingle	Lung nodule Convolutional neural network U-Net Residual learning Image segmentation Image classification Computer applications to medicine. Medical informatics R858-859.7 Biology (General) QH301-705.5 Hui Yu Jinqiu Li Lixin Zhang Yuzhen Cao Xuyao Yu Jinglai Sun Design of lung nodules segmentation and recognition algorithm based on deep learning
description	Abstract Background Accurate segmentation and recognition algorithm of lung nodules has great important value of reference for early diagnosis of lung cancer. An algorithm is proposed for 3D CT sequence images in this paper based on 3D Res U-Net segmentation network and 3D ResNet50 classification network. The common convolutional layers in encoding and decoding paths of U-Net are replaced by residual units while the loss function is changed to Dice loss after using cross entropy loss to accelerate network convergence. Since the lung nodules are small and rich in 3D information, the ResNet50 is improved by replacing the 2D convolutional layers with 3D convolutional layers and reducing the sizes of some convolution kernels, 3D ResNet50 network is obtained for the diagnosis of benign and malignant lung nodules. Results 3D Res U-Net was trained and tested on 1044 CT subcases in the LIDC-IDRI database. The segmentation result shows that the Dice coefficient of 3D Res U-Net is above 0.8 for the segmentation of lung nodules larger than 10 mm in diameter. 3D ResNet50 was trained and tested on 2960 lung nodules in the LIDC-IDRI database. The classification result shows that the diagnostic accuracy of 3D ResNet50 is 87.3% and AUC is 0.907. Conclusion The 3D Res U-Net module improves segmentation performance significantly with the comparison of 3D U-Net model based on residual learning mechanism. 3D Res U-Net can identify small nodules more effectively and improve its segmentation accuracy for large nodules. Compared with the original network, the classification performance of 3D ResNet50 is significantly improved, especially for small benign nodules.
format	article
author	Hui Yu Jinqiu Li Lixin Zhang Yuzhen Cao Xuyao Yu Jinglai Sun
author_facet	Hui Yu Jinqiu Li Lixin Zhang Yuzhen Cao Xuyao Yu Jinglai Sun
author_sort	Hui Yu
title	Design of lung nodules segmentation and recognition algorithm based on deep learning
title_short	Design of lung nodules segmentation and recognition algorithm based on deep learning
title_full	Design of lung nodules segmentation and recognition algorithm based on deep learning
title_fullStr	Design of lung nodules segmentation and recognition algorithm based on deep learning
title_full_unstemmed	Design of lung nodules segmentation and recognition algorithm based on deep learning
title_sort	design of lung nodules segmentation and recognition algorithm based on deep learning
publisher	BMC
publishDate	2021
url	https://doaj.org/article/ea95d1654c53428fa1eb5fc5cc931d56
work_keys_str_mv	AT huiyu designoflungnodulessegmentationandrecognitionalgorithmbasedondeeplearning AT jinqiuli designoflungnodulessegmentationandrecognitionalgorithmbasedondeeplearning AT lixinzhang designoflungnodulessegmentationandrecognitionalgorithmbasedondeeplearning AT yuzhencao designoflungnodulessegmentationandrecognitionalgorithmbasedondeeplearning AT xuyaoyu designoflungnodulessegmentationandrecognitionalgorithmbasedondeeplearning AT jinglaisun designoflungnodulessegmentationandrecognitionalgorithmbasedondeeplearning
_version_	1718429379510403072

Design of lung nodules segmentation and recognition algorithm based on deep learning

Ejemplares similares