Relational Modeling for Robust and Efficient Pulmonary Lobe Segmentation in CT Scans
- PMID: 32730216
- PMCID: PMC7393217
- DOI: 10.1109/TMI.2020.2995108
Relational Modeling for Robust and Efficient Pulmonary Lobe Segmentation in CT Scans
Abstract
Pulmonary lobe segmentation in computed tomography scans is essential for regional assessment of pulmonary diseases. Recent works based on convolution neural networks have achieved good performance for this task. However, they are still limited in capturing structured relationships due to the nature of convolution. The shape of the pulmonary lobes affect each other and their borders relate to the appearance of other structures, such as vessels, airways, and the pleural wall. We argue that such structural relationships play a critical role in the accurate delineation of pulmonary lobes when the lungs are affected by diseases such as COVID-19 or COPD. In this paper, we propose a relational approach (RTSU-Net) that leverages structured relationships by introducing a novel non-local neural network module. The proposed module learns both visual and geometric relationships among all convolution features to produce self-attention weights. With a limited amount of training data available from COVID-19 subjects, we initially train and validate RTSU-Net on a cohort of 5000 subjects from the COPDGene study (4000 for training and 1000 for evaluation). Using models pre-trained on COPDGene, we apply transfer learning to retrain and evaluate RTSU-Net on 470 COVID-19 suspects (370 for retraining and 100 for evaluation). Experimental results show that RTSU-Net outperforms three baselines and performs robustly on cases with severe lung infection due to COVID-19.
Figures
![Fig. 1.](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-f0001.gif)
![Fig. 2.](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-f0002.gif)
![Fig. 3.](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-f0003.gif)
![Fig. 4.](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-f0004.gif)
![Fig. 5.](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-f0005.gif)
![formula image](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-ig0006.jpg)
![formula image](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-ig0007.jpg)
![formula image](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-ig0008.jpg)
![formula image](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-ig0009.jpg)
![formula image](https://cdn.statically.io/img/www.ncbi.nlm.nih.gov/pmc/articles/instance/7393217/bin/nihms-1595878-ig0010.jpg)
Update of
-
Relational Modeling for Robust and Efficient Pulmonary Lobe Segmentation in CT Scans.ArXiv [Preprint]. 2020 May 12:arXiv:2004.07443v4. ArXiv. 2020. Update in: IEEE Trans Med Imaging. 2020 Aug;39(8):2664-2675. doi: 10.1109/TMI.2020.2995108. PMID: 32550251 Free PMC article. Updated. Preprint.
Similar articles
-
Detection of coronavirus disease from X-ray images using deep learning and transfer learning algorithms.J Xray Sci Technol. 2020;28(5):841-850. doi: 10.3233/XST-200720. J Xray Sci Technol. 2020. PMID: 32804113 Free PMC article.
-
A Noise-Robust Framework for Automatic Segmentation of COVID-19 Pneumonia Lesions From CT Images.IEEE Trans Med Imaging. 2020 Aug;39(8):2653-2663. doi: 10.1109/TMI.2020.3000314. IEEE Trans Med Imaging. 2020. PMID: 32730215 Free PMC article.
-
Chest computed tomography findings of COVID-19 pneumonia: pictorial essay with literature review.Jpn J Radiol. 2020 Nov;38(11):1012-1019. doi: 10.1007/s11604-020-01010-7. Epub 2020 Jun 25. Jpn J Radiol. 2020. PMID: 32588277 Free PMC article. Review.
-
Relational Modeling for Robust and Efficient Pulmonary Lobe Segmentation in CT Scans.ArXiv [Preprint]. 2020 May 12:arXiv:2004.07443v4. ArXiv. 2020. Update in: IEEE Trans Med Imaging. 2020 Aug;39(8):2664-2675. doi: 10.1109/TMI.2020.2995108. PMID: 32550251 Free PMC article. Updated. Preprint.
-
Ultrasound in COVID-19: a timeline of ultrasound findings in relation to CT.Clin Radiol. 2020 Jul;75(7):553-554. doi: 10.1016/j.crad.2020.04.003. Epub 2020 Apr 18. Clin Radiol. 2020. PMID: 32331781 Free PMC article. Review. No abstract available.
Cited by
-
Micro-CT and deep learning: Modern techniques and applications in insect morphology and neuroscience.Front Insect Sci. 2023 Jan 23;3:1016277. doi: 10.3389/finsc.2023.1016277. eCollection 2023. Front Insect Sci. 2023. PMID: 38469492 Free PMC article. Review.
-
Estimating lung function from computed tomography at the patient and lobe level using machine learning.Med Phys. 2024 Apr;51(4):2834-2845. doi: 10.1002/mp.16915. Epub 2024 Feb 8. Med Phys. 2024. PMID: 38329315
-
Lightweight deep learning model incorporating an attention mechanism and feature fusion for automatic classification of gastric lesions in gastroscopic images.Biomed Opt Express. 2023 Aug 11;14(9):4677-4695. doi: 10.1364/BOE.487456. eCollection 2023 Sep 1. Biomed Opt Express. 2023. PMID: 37791283 Free PMC article.
-
A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises.Proc IEEE Inst Electr Electron Eng. 2021 May;109(5):820-838. doi: 10.1109/JPROC.2021.3054390. Epub 2021 Feb 26. Proc IEEE Inst Electr Electron Eng. 2021. PMID: 37786449 Free PMC article.
-
Quantification of pulmonary involvement in COVID-19 pneumonia: an upgrade of the LungQuant software for lung CT segmentation.Eur Phys J Plus. 2023;138(4):326. doi: 10.1140/epjp/s13360-023-03896-4. Epub 2023 Apr 11. Eur Phys J Plus. 2023. PMID: 37064789 Free PMC article.
References
-
- Prokop M, van Everdingen W, van Rees Vellinga T, Quarles van Ufford J, Stöger L, Beenen L, Geurts B, Gietema H, Krdzalic J, Schaefer-Prokop C, van Ginneken B, Brink M, and The “COVID-19 Standardized Reporting” Working Group of the Dutch Radiological Society, “CO-RADS – a categorical CT assessment scheme for patients with suspected COVID-19: definition and evaluation,” Radiology, p. 201473 , 2020. [Online]. Available: https://pubs.rsna.org/doi/abs/10.1148/radiol.2020201473 - DOI - PMC - PubMed
-
- Kubo M, Niki N, Nakagawa S, Eguchi K, Kaneko M, Moriyama N, Omatsu H, Kakinuma R, and Yamaguchi N, “Extraction algorithm of pulmonary fissures from thin-section CT images based on linear feature detector method,” IEEE Trans Nucl Sci, vol. 46, no. 6, pp. 2128–2133, 1999.
-
- Kubo M, Niki N, Eguchi K, Kaneko M, Kusumoto M, Moriyama N, Omatsu H, Kakinuma R, Nishiyama H, Mori K et al., “Extraction of pulmonary fissures from thin-section CT images using calculation of surface-curvatures and morphology filters,” in ICIP, vol. 2, 2000, pp. 637–640.