Abstract
Purpose
The proposed work aims to develop an algorithm to precisely segment the lung parenchyma in thoracic CT scans. To achieve this goal, the proposed technique utilized a combination of deep learning and traditional image processing algorithms. The initial step utilized a trained convolutional neural network (CNN) to generate preliminary lung masks, followed by the proposed post-processing algorithm for lung boundary correction.
Methods
First, the proposed method trained an improved 2D U-Net CNN model with Inception-ResNet-v2 as its backbone. The model was trained on 32 CT scans from two different sources: one from the VESSEL12 grand challenge and the other from AIIMS Delhi. Further, the model’s performance was evaluated on a test dataset of 16 CT scans with juxta-pleural nodules obtained from AIIMS Delhi and the LUNA16 challenge. The model’s performance was assessed using evaluation metrics such as average volumetric dice coefficient (DSCavg), average IoU score (IoUavg), and average F1 score (F1avg). Finally, the proposed post-processing algorithm was implemented to eliminate false positives from the model's prediction and to include juxta-pleural nodules in the final lung masks.
Results
The trained model reported a DSCavg of 0.9791 ± 0.008, IoUavg of 0.9624 ± 0.007, and F1avg of 0.9792 ± 0.004 on the test dataset. Applying the post-processing algorithm to the predicted lung masks obtained a DSCavg of 0.9713 ± 0.007, IoUavg of 0.9486 ± 0.007, and F1avg of 0.9701 ± 0.008. The post-processing algorithm successfully included juxta-pleural nodules in the final lung mask.
Conclusions
Using a CNN model, the proposed method for lung parenchyma segmentation produced precise segmentation results. Furthermore, the post-processing algorithm addressed false positives and negatives in the model’s predictions. Overall, the proposed approach demonstrated promising results for lung parenchyma segmentation. The method has the potential to be valuable in the advancement of computer-aided diagnosis (CAD) systems for automatic nodule detection.
This is a preview of subscription content, log in via an institution to check access.
References
Lung Cancer Statistics | How Common is Lung Cancer? https://www.cancer.org/cancer/lung-cancer/about/key-statistics.html. Accessed 26 Sep 2022
Liu Y, Wang H, Li Q, McGettigan MJ, Balagurunathan Y, Garcia AL, Thompson ZJ, Heine JJ, Ye Z, Gillies RJ, Schabath MB (2017) Radiologic Features of Small Pulmonary Nodules and Lung Cancer Risk in the National Lung Screening Trial: A Nested Case-Control Study. https://doi.org/10.1148/radiol2017161458 286:298–306.
Larici AR, Farchione A, Franchi P, Ciliberto M, Cicchetti G, Calandriello L, del Ciello A, Bonomo L (2017) Lung nodules: size still matters. Eur Respir Rev. https://doi.org/10.1183/16000617.0025-2017
Hansell DM, Bankier AA, MacMahon H, McLoud TC, Müller NL, Remy J (2008) Fleischner Society: Glossary of terms for thoracic imaging. Radiology 246:697–722
Kostis WJ, Reeves AP, Yankelevitz DF, Henschke CI (2003) Three-Dimensional Segmentation and Growth-Rate Estimation of Small Pulmonary Nodules in Helical CT Images. IEEE Trans Med Imaging 22:1259–1274
Team TNLSTR (2013) Results of Initial Low-Dose Computed Tomographic Screening for Lung Cancer. N Engl J Med 368:1980
Halder A, Dey D, Sadhu AK (2020) Lung Nodule Detection from Feature Engineering to Deep Learning in Thoracic CT Images: a Comprehensive Review. J Digit Imaging 33:655–677
da Silva Sousa JRF, Silva AC, de Paiva AC, Nunes RA (2010) Methodology for automatic detection of lung nodules in computerized tomography images. Comput Methods Programs Biomed 98:1–14
Nishio M, Sugiyama O, Yakami M, Ueno S, Kubo T, Kuroda T, Togashi K (2018) Computer-aided diagnosis of lung nodule classification between benign nodule, primary lung cancer, and metastatic lung cancer at different image size using deep convolutional neural network with transfer learning. PLoS ONE 13:e0200721
Mahersia H, Zaroug M, Gabralla L (2015) Lung Cancer Detection on CT Scan Images: A Review on the Analysis Techniques. Int J Adv Res Artif Intell. https://doi.org/10.14569/IJARAI.2015.040406
Hu S, Hoffman EA, Reinhardt JM (2001) Automatic lung segmentation for accurate quantitation of volumetric X-ray CT images. IEEE Trans Med Imaging 20:490–498
Leader JK, Zheng B, Rogers RM, Sciurba FC, Perez A, Chapman BE, Patel S, Fuhrman CR, Gur D (2003) Automated lung segmentation in X-ray computed tomography: development and evaluation of a heuristic threshold-based scheme. Acad Radiol 10:1224–1236
Nurfauzi R, Nugroho HA, Ardiyanto I (2017) Lung detection using Adaptive Border correction. Proceeding - 2017 3rd Int Conf Sci Technol ICST 2017 57–60.
Li XY, Xu F, Hu X, Peng SH, Do NH, Zhao JM (2017) Self-adapting threshold of pulmonary parenchyma. Proc - 2016 9th Int Congr Image Signal Process Biomed Eng Informatics. CISP-BMEI 2016:1429–1434
Wei Y, Shen G, Li JJ (2013) A fully automatic method for lung parenchyma segmentation and repairing. J Digit Imaging 26:483–495
El-Regaily SA, Salem MAM, Aziz MHA, Roushdy MI (2017) Lung nodule segmentation and detection in computed tomography. 2017 IEEE 8th Int Conf Intell Comput Inf Syst ICICIS 2017 2018-January:72–78.
Sun S, Li W, Kang Y (2016) Lung nodule detection based on GA and SVM. Proc - 2015 8th Int Conf Biomed Eng Informatics. BMEI 2015:96–100
Armato SG, Sensakovic WF (2004) Automated lung segmentation for thoracic CT impact on computer-aided diagnosis. Acad Radiol 11:1011–1021
Zhang W, Wang X, Li X, Chen J (2018) 3D skeletonization feature based computer-aided detection system for pulmonary nodules in CT datasets. Comput Biol Med 92:64–72
Li Z, Hoffman EA, Reinhardt JM (2006) Atlas-driven lung lobe segmentation in volumetric X-ray CT images. IEEE Trans Med Imaging 25:1–16
Zhang L, Reinhardt JM 3D Pulmonary CT Image Registration with a Standard Lung Atlas. Proc. SPIE 3978, Medical Imaging 2000: Physiology and Function from Multidimensional Images. doi: https://doi.org/10.1117/12383441
Chen X, Udupa JK, Bagci U, Zhuge Y, Yao J (2012) Medical image segmentation by combining graph cuts and oriented active appearance models. IEEE Trans Image Process 21:2035–2046
Wels M, Carneiro G, Aplas A, Huber M, Hornegger J, Comaniciu D (2008) A discriminative model-constrained graph cuts approach to fully automated pediatric brain tumor segmentation in 3-D MRI. Lect Notes Comput Sci (including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) 5241 LNCS:67–75.
Tsai YP, Ko CH, Hung YP, Shih ZC (2007) Background removal of multiview images by learning shape priors. IEEE Trans Image Process 16:2607–2616
Benzakoun J, Bommart S, Coste J, Chassagnon G, Lederlin M, Boussouar S, Revel MP (2016) Computer-aided diagnosis (CAD) of subsolid nodules: Evaluation of a commercial CAD system. Eur J Radiol 85:1728–1734
Tan J, Jing L, Huo Y, Li L, Akin O, Tian Y (2021) LGAN: Lung segmentation in CT scans using generative adversarial network. Comput Med Imaging Graph 87:1–24
Singadkar G, Mahajan A, Thakur M, Talbar S (2021) Automatic lung segmentation for the inclusion of juxtapleural nodules and pulmonary vessels using curvature based border correction. J King Saud Univ - Comput Inf Sci 33:975–987
Shariaty F, Orooji M, Mousavi M, Baranov M, Velichko E (2020) Automatic lung segmentation in computed tomography images using active shape model. Proc 2020 IEEE Int Conf Electr Eng Photonics. EExPolytech 2020:156–159
Chung H, Ko H, Jeon SJ, Yoon KH, Lee J (2018) Automatic Lung Segmentation with Juxta-Pleural Nodule Identification Using Active Contour Model and Bayesian Approach. IEEE J Transl Eng Heal Med. https://doi.org/10.1109/JTEHM.2018.2837901
Zhou S, Cheng Y, Tamura S (2014) Automated lung segmentation and smoothing techniques for inclusion of juxtapleural nodules and pulmonary vessels on chest CT images. Biomed Signal Process Control 13:62–70
Shi Z, Ma J, Zhao M, Liu Y, Feng Y, Zhang M, He L, Suzuki K (2016) Many Is Better Than One: An Integration of Multiple Simple Strategies for Accurate Lung Segmentation in CT Images. Biomed Res Int. https://doi.org/10.1155/2016/1480423
Xu M, Qi S, Yue Y, Teng Y, Xu L, Yao Y, Qian W (2019) Segmentation of lung parenchyma in CT images using CNN trained with the clustering algorithm generated dataset 08 Information and Computing Sciences 0801 Artificial Intelligence and Image Processing Robert Koprowski. Biomed Eng Online 18:1–21
Hofmanninger J, Prayer F, Pan J, Röhrich S, Prosch H, Langs G (2020) Automatic lung segmentation in routine imaging is primarily a data diversity problem, not a methodology problem. Eur Radiol Exp. https://doi.org/10.1186/s41747-020-00173-2
Home - Grand Challenge. https://vessel12.grand-challenge.org/. Accessed 27 Sep 2022
Download - Grand Challenge. https://luna16.grand-challenge.org/Download/. Accessed 27 Sep 2021
qubvel/segmentation_models: Segmentation models with pretrained backbones. Keras and TensorFlow Keras. https://github.com/qubvel/segmentation_models. Accessed 7 Jan 2023
Szegedy C, Ioffe S, Vanhoucke V, Alemi AA (2017) Inception-v4, inception-ResNet and the impact of residual connections on learning. 31st AAAI Conf Artif Intell AAAI 2017 4278–4284.
Ronneberger O, Fischer P, Brox T (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. Lect Notes Comput Sci (including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) 9351:234–241
Funding
This research received external funding from Indian Council of Medical Research (ICMR), New Delhi, India (AI-Adhoc/06/2022-AI Cell).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors declare that they have no competing interests.
Ethical approval
Protocol for this retrospective study was approved by the Institute Ethics Committee of All India Institute of Medical Sciences, New Delhi, India.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Rikhari, H., Baidya Kayal, E., Ganguly, S. et al. Fully automatic deep learning-based lung parenchyma segmentation and boundary correction in thoracic CT scans. Int J CARS 19, 261–272 (2024). https://doi.org/10.1007/s11548-023-03010-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11548-023-03010-0