Welcome to Francis Academic Press

Academic Journal of Computing & Information Science, 2023, 6(11); doi: 10.25236/AJCIS.2023.061104.

Segmentation of nodules in CT images of the lung using an improved V-Net network model


Xiaoru Xu1,2, Lingyan Du1,2, Dongsheng Yin1,2

Corresponding Author:
Lingyan Du

1School of Automation and Information Engineering, Sichuan University of Science and Engineering, Zigong, China

2Artificial Intelligence Key Laboratory of Sichuan Province, Yibin, China


Lung cancer has traditionally exhibited high incidence and mortality rates, making the early detection and treatment of the disease essential in reducing mortality rates. We present a new network design to enhance the separation of lung nodules in CT lung images. The design reduces the problem of missed or incorrect segmentation due to unclear nodule morphology, varying shapes, and attachment to the pleura. In this paper, the V-Net serves as the foundation network, paired with a multi-scale feature network to enhance the original's leap connections. Two distinct attention mechanism modules are integrated into the network encoding and decoding for increased feature extraction of lung nodules. The Log-Cosh Dice Loss replaces the original loss function to address the issue of non-convexity in the Dice loss function. Additionally, the lung 3D images are cropped to resolve the problem of imbalanced distribution of positive and negative samples within lung CT images. We evaluated the performance of the model on the LUNA16 dataset. The evaluation results demonstrate the superiority of the model. We observed objective improvements compared to the initial network; 6.9% improvement in DSC values, 11.5% improvement in MIoU values, 8.5% improvement in accuracy and 99.8% pixel accuracy. This method has been found to effectively prevent missed lung nodules and produce satisfactory segmentation results.


Deep learning, V-Net network, Attention mechanism, LIDC-IDRI

Cite This Paper

Xiaoru Xu, Lingyan Du, Dongsheng Yin. Segmentation of nodules in CT images of the lung using an improved V-Net network model. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 11: 26-34. https://doi.org/10.25236/AJCIS.2023.061104.


[1] Taher F, Prakash N, Shaffie A, et al. An overview of lung cancer classification algorithms and their performances [J]. IAENG International Journal of Computer Science, 2021, 48(4). 

[2] Dong T, Wei L, Nie S D. Research progress of lung nodule segmentation based on CT images. Journal of Image and Graphics, 2021, 26(4): 751-765.

[3] Tammemagi M C, Mayo J R, Lam S. Cancer in pulmonary nodules detected on first screening CT[J]. The New England journal of medicine, 2013, 369(21): 2060-2061.

[4] Vachani A, Carroll N M, Simoff M J, et al. Stage migration and lung cancer incidence after initiation of low-dose computed tomography screening[J]. Journal of Thoracic Oncology, 2022, 17(12): 1355-1364.

[5] Wang S, Zhou M, Liu Z, et al. Central focused convolutional neural networks: Developing a data-driven model for lung nodule segmentation[J]. Medical image analysis, 2017, 40: 172-183.

[6] Aresta G, Jacobs C, Araújo T, et al. iW-Net: an automatic and minimalistic interactive lung nodule segmentation deep network[J]. Scientific reports, 2019, 9(1): 11591.

[7] Shi J, Ye Y, Zhu D, et al. Comparative analysis of pulmonary nodules segmentation using multiscale residual U-Net and fuzzy C-means clustering[J]. Computer Methods and Programs in Biomedicine, 2021, 209: 106332.

[8] Wu W, Gao L, Duan H, et al. Segmentation of pulmonary nodules in CT images based on 3D‐UNET combined with three‐dimensional conditional random field optimization[J]. Medical Physics, 2020, 47(9): 4054-4063.

[9] Kido S, Kidera S, Hirano Y, et al. Segmentation of lung nodules on ct images using a nested three-dimensional fully connected convolutional network[J]. Frontiers in artificial intelligence, 2022, 5: 782225.

[10] Zhou Z, Siddiquee M M R, Tajbakhsh N, et al. Unet++: Redesigning skip connections to exploit multiscale features in image segmentation[J]. IEEE transactions on medical imaging, 2019, 39(6): 1856-1867.

[11] Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[C]//International conference on machine learning. pmlr, 2015: 448-456.

[12] Li X, Wang W, Hu X, et al. Selective kernel networks[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2019: 510-519.

[13] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.

[14] Jadon S. A survey of loss functions for semantic segmentation[C]//2020 IEEE conference on computational intelligence in bioinformatics and computational biology (CIBCB). IEEE, 2020: 1-7.

[15] Armato III S G, McLennan G, Bidaut L, et al. The lung image database consortium (LIDC) and image database resource initiative (IDRI): a completed reference database of lung nodules on CT scans [J]. Medical physics, 2011, 38(2): 915-931.

[16] Huang G, Liu Z, Van Der Maaten L, et al. Densely connected convolutional networks[C] //Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4700-4708.