AI-Driven Deep Learning for Lung Segmentation in CT Imaging
DOI:
https://doi.org/10.21015/vtse.v14i1.2263Abstract
With the remarkable growth in the field of artificial intelligence; fully automated diagnostic systems are no longer a dream. Recent advancements in medical imaging have paved an early and accurate pulmonary complications. This study presents an updated approach of lung nodule segmentation from CT (computed tomography) scans using an improved UNet architecture. The performance of model is enhanced with ResNet-34 and EfficientNet backbone. After segmenting, feature extraction and subsequent classification is performed to classify the type of chest cancer. A publicly available dataset comprising 1,000 CT scan images available on Kaggle is used. The core of our segmentation model is built upon the U-Net architecture, renowned for its efficacy in biomedical image segmentation tasks. Implemented using Python and TensorFlow within the Google Colab environment, the model trains to accurately classify 3 types of chest cancers. The performance is evaluated by metrics such as the Intersection over Union (IoU) to assess segmentation quality. To further enhance the model's robustness and generalizability, UNet architecture integrated with hybrid ResNet-34 and EfficientNet-B4 backbone encoder which achieves a dice score of 0.66 with better performance on small nodules i.e., 0.58. The hybrid backbone ensures efficient feature extraction, contributing to improved segmentation results. The proposed model employed multiphase optimization which further enhances its potential for AI-driven solutions in medical diagnostic. The automated segmentation and classification can aid clinicians in the diagnosis and detection of pulmonary abnormalities, including COVID-19.
References
O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional networks for biomedical image segmentation,” in Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2015, pp. 234–241.
Ö. Çiçek, A. Abdulkadir, S. S. Lienkamp, T. Brox, and O. Ronneberger, “3D U-Net: Learning dense volumetric segmentation from sparse annotation,” in Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2016, pp. 424–432.
G. Litjens et al., “A survey on deep learning in medical image analysis,” Medical Image Analysis, vol. 42, pp. 60–88, 2017.
A. A. A. Setio et al., “Pulmonary nodule detection in CT images: False positive reduction using multi-view convolutional networks,” IEEE Transactions on Medical Imaging, vol. 35, no. 5, pp. 1160–1169, 2016.
G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 4700–4708.
M. Tan and Q. Le, “EfficientNet: Rethinking model scaling for convolutional neural networks,” in Proceedings of the International Conference on Machine Learning (ICML), 2019, pp. 6105–6114.
F. Milletari, N. Navab, and S. A. Ahmadi, “V-Net: Fully convolutional neural networks for volumetric medical image segmentation,” in Proceedings of the International Conference on 3D Vision (3DV), 2016, pp. 565–571.
S. G. Armato et al., “The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): A completed reference database of lung nodules on CT scans,” Medical Physics, vol. 38, no. 2, pp. 915–931, 2011.
J. Chen et al., “TransUNet: Transformers make strong encoders for medical image segmentation,” arXiv preprint arXiv:2102.04306, 2021.
H. Zhao, J. Shi, X. Qi, X. Wang, and J. Jia, “Pyramid scene parsing network,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 2881–2890.
F. Isensee, P. F. Jaeger, S. A. A. Kohl, J. Petersen, and K. H. Maier-Hein, “nnU-Net: A self-adapting framework for U-Net-based medical image segmentation,” Nature Methods, vol. 18, no. 2, pp. 203–211, 2021.
A. Mansoor et al., “Segmentation and image analysis of abnormal lungs at CT: Current approaches, challenges, and future trends,” RadioGraphics, vol. 34, no. 2, pp. 341–361, 2014.
P. Rajpurkar et al., “CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning,” arXiv preprint arXiv:1711.05225, 2017.
Roboflow, “Roboflow: Organize, annotate, and preprocess images for computer vision,” 2023. [Online]. Available: https://roboflow.com
Y. Xu et al., “Lung nodule classification using deep feature fusion in chest CT images,” Computers in Biology and Medicine, vol. 122, p. 103819, 2020.
T. Awan and K. B. Khan, “Analysis of underfitting and overfitting in U-Net semantic segmentation for lung nodule identification from X-ray radiographs,” in Proceedings of the IEEE International Conference on Emerging Trends in Engineering, Sciences and Technology (ICES&T), 2023, pp. 1–5.
T. Awan, K. B. Khan, and A. Mannan, “A compact CNN model for automated detection of COVID-19 using thorax X-ray images,” Journal of Intelligent & Fuzzy Systems, vol. 44, no. 5, pp. 7887–7907, 2023.
T. Awan and K. B. Khan, “Investigating the impact of novel XRayGAN in feature extraction for thoracic disease detection in chest radiographs: Lung cancer,” Signal, Image and Video Processing, vol. 18, no. 5, pp. 3957–3972, 2024.
T. Awan et al., “AI-powered lung cancer detection from CT imaging,” VFAST Transactions on Software Engineering, vol. 12, no. 2, pp. 241–249, 2024.
T. Awan et al., “Advanced EEG-based brain monitoring for early detection of harmful neural activities using EfficientNet architecture,” Biomedical Signal Processing and Control, vol. 112, p. 108360, 2026.
T. Awan et al., “Generating synthetic data in biomedical imaging by designing GANs,” VFAST Transactions on Software Engineering, vol. 12, no. 3, pp. 44–54, 2024.
L. Yan, D. Liu, Q. Xiang, Y. Luo, T. Wang, D. Wu, H. Chen, Y. Zhang, and Q. Li, “PSPNet-based automatic segmentation network model for prostate magnetic resonance imaging,” Comput. Methods Programs Biomed., vol. 207, p. 106211, 2021.
P. Sabitha, R. A. Canessane, M. S. P. Minu, V. Gowri, and M. S. A. Vigil, “An improved deep network model to isolate lung nodules from histopathological images using an orchestrated and shifted window vision transformer,” Traitement du Signal, vol. 41, pp. 2081–2091, 2024.
H. Polat, “A modified DeepLabV3+ based semantic segmentation of chest computed tomography images for COVID-19 lung infections,” Int. J. Imaging Syst. Technol., vol. 32, no. 5, pp. 1481–1495, 2022.
Downloads
Published
How to Cite
Issue
Section
License
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (CC-By) that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
This work is licensed under a Creative Commons Attribution License CC BY
