Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Aug 30:2022:3836539.
doi: 10.1155/2022/3836539. eCollection 2022.

Lightweight Deep Learning Classification Model for Identifying Low-Resolution CT Images of Lung Cancer

Shanmugasundaram Marappan  1 , Muhammad Danish Mujib  2 , Adnan Ahmed Siddiqui  3 , Abdul Aziz  4 , Samiullah Khan  5 , Mahesh Singh  6
Affiliations

Lightweight Deep Learning Classification Model for Identifying Low-Resolution CT Images of Lung Cancer

Shanmugasundaram Marappan et al. Comput Intell Neurosci. .

Abstract

With an astounding five million fatal cases every year, lung cancer is among the leading causes of mortality worldwide for both men and women. The diagnosis of lung illnesses can benefit from the information a computed tomography (CT) scan can offer. The major goals of this study are to diagnose lung cancer and its seriousness and to identify malignant lung nodules from the provided input lung picture. This paper applies unique deep learning techniques to identify the exact location of the malignant lung nodules. Using a DenseNet model, mixed ground glass is analyzed in low-dose, low-resolution CT scan images of nodules (mGGNs) with a slice thickness of 5 mm in this study. This was done to categorize and identify many histological subtypes of lung cancer. Low-resolution CT scans are used to pathologically classify invasive adenocarcinoma (IAC) and minimally invasive adenocarcinoma (MIA). 105 low-resolution CT images with 5 mm thick slices from 105 patients at Lishui Central Hospital were selected. To detect and distinguish, IAC and MIA, extend and enhance deep learning two- and three-dimensional DenseNet models are used. The two-dimensional DenseNet model was shown to perform much better than the three-dimensional DenseNet model in terms of classification accuracy (76.67%), sensitivity (63.3%), specificity (100%), and area under the receiver operating characteristic curve (0.88). Finding the histological subtypes of persons with lung cancer should aid doctors in making a more precise diagnosis, even if the image quality is not outstanding.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no conflicts of interest.

Figures

Figure 1
Figure 1
CT images with confirmed IAC and MIA [2]. (a) IAC. (b) MIA.
Figure 2
Figure 2
Distribution of the actual number of pixels occupied by the diameter of the long axis of the nodule [4].
Figure 3
Figure 3
DenseNet basic network model structure.
Figure 4
Figure 4
Two-dimensional and three-dimensional DenseNet network model experimental process.
Figure 5
Figure 5
Performance comparison of two-dimensional and three-dimensional DenseNet network models.
Figure 6
Figure 6
Confusion element of different parameters of the two-dimensional DenseNet model.
Figure 7
Figure 7
AUC of different parameters of the two-dimensional DenseNet model.
Figure 8
Figure 8
Performance comparison of confusion element of the proposed model with other deep learning network models.
Figure 9
Figure 9
Performance comparison of AUC of the proposed model with other deep learning network models.
See this image and copyright information in PMC

References

    1. Qu H., Gao P. The effect of square wave stretching on the proliferation of lung adenocarcinoma cells and Squamous Carcinoma of Tongue cells. Proceedings of the 2010 3rd International Conference on Biomedical Engineering and Informatics; October 2010; Yantai, China. pp. 2122–2124. - DOI
    1. Khan A., Saha G., Dasgupta S., Datta S. K. Prediction of dominant genes responsible for lung adenocarcinoma using rough set theory. Proceedings of the 2011 E-Health and Bioengineering Conference (EHB); November 2011; Iasi, Romania. pp. 1–4.
    1. Ma C., Yue S., Li Q. Deep transfer learning strategy for invasive lung adenocarcinoma classification appearing as ground glass nodules. Proceedings of the 2021 IEEE International Instrumentation and Measurement Technology Conference (I2MTC); May 2021; Glasgow, United Kingdom. pp. 1–5. - DOI
    1. Zheng S.-Y., Zhao J., Jiang D., Ge J.-f. Relationship between expression of caspase-3 and apoptosis by FasL transfected in lung adenocarcinoma cell. Proceedings of the 2009 International Conference on Future BioMedical Information Engineering; December 2009; FBIE; pp. 219–221. - DOI
    1. Zhai Y., Chen Y., Jiang Y., Li Q. Weighted gene Co-expression network analysis of gene modules for lung adenocarcinoma. Proceedings of the 2018 5th International Conference on Systems and Informatics; November 2018; Nanjing, China. ICSAI; pp. 473–477. - DOI