Robust lung segmentation in Chest X-ray images using modified U-Net with deeper network and residual blocks

Lung diseases remain a leading cause of mortality worldwide, as evidenced by statistics from the World Health Organization (WHO). The limited availability of radiologists to interpret Chest X-ray (CXR) images for diagnosing common lung conditions poses a significant challenge, often resulting in del...

Full description

Saved in:
Bibliographic Details
Main Authors: Wiley Tam, Paul Babyn, Javad Alirezaie
Format: Article
Language:English
Published: Elsevier 2025-01-01
Series:Computer Methods and Programs in Biomedicine Update
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2666990025000369
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849392832255623168
author Wiley Tam
Paul Babyn
Javad Alirezaie
author_facet Wiley Tam
Paul Babyn
Javad Alirezaie
author_sort Wiley Tam
collection DOAJ
description Lung diseases remain a leading cause of mortality worldwide, as evidenced by statistics from the World Health Organization (WHO). The limited availability of radiologists to interpret Chest X-ray (CXR) images for diagnosing common lung conditions poses a significant challenge, often resulting in delayed diagnosis and treatment. In response, Computer-Aided Diagnostic (CAD) tools can be used to potentially streamline and expedite the diagnostic process. Recently, deep learning techniques have gained prominence in the automated analysis of CXR images, particularly in segmenting lung regions as a critical preliminary step. This study aims to develop and evaluate a lung segmentation model based on a modified U-Net architecture. The architecture leverages techniques such as transfer learning with DenseNet201 as a feature extractor alongside dilated convolutions and residual blocks. An ablation study was conducted to evaluate these architectural components, along with additional elements like augmented data, alternative backbones, and attention mechanisms. Numerous and extensive experiments were performed on two publicly available datasets, the Montgomery County (MC) and Shenzhen Hospital (SH) datasets, to validate the efficacy of these techniques on segmentation performance. Outperforming other state-of-the-art methods on the MC dataset, the proposed model achieved a Jaccard Index (IoU) of 97.77 and a Dice Similarity Coefficient (DSC) of 98.87. These results represent a significant improvement over the baseline U-Net, with gains of 3.37% and 1.75% in IoU and DSC, respectively. These findings highlight the importance of architectural enhancements in deep learning-based lung segmentation models, contributing to more efficient, accurate, and reliable CAD systems for lung disease assessment.
format Article
id doaj-art-602e567cb4c54e3fae6cc462aa281a09
institution Kabale University
issn 2666-9900
language English
publishDate 2025-01-01
publisher Elsevier
record_format Article
series Computer Methods and Programs in Biomedicine Update
spelling doaj-art-602e567cb4c54e3fae6cc462aa281a092025-08-20T03:40:41ZengElsevierComputer Methods and Programs in Biomedicine Update2666-99002025-01-01810021110.1016/j.cmpbup.2025.100211Robust lung segmentation in Chest X-ray images using modified U-Net with deeper network and residual blocksWiley Tam0Paul Babyn1Javad Alirezaie2Department of Electrical, Computer and Biomedical Eng., Toronto Metropolitan University, Toronto, M5B 2K3, ON, CanadaDepartment of Medical Imaging, University of Saskatchewan, Saskatoon, S7N 0W8, SK, CanadaDepartment of Electrical, Computer and Biomedical Eng., Toronto Metropolitan University, Toronto, M5B 2K3, ON, Canada; Corresponding author.Lung diseases remain a leading cause of mortality worldwide, as evidenced by statistics from the World Health Organization (WHO). The limited availability of radiologists to interpret Chest X-ray (CXR) images for diagnosing common lung conditions poses a significant challenge, often resulting in delayed diagnosis and treatment. In response, Computer-Aided Diagnostic (CAD) tools can be used to potentially streamline and expedite the diagnostic process. Recently, deep learning techniques have gained prominence in the automated analysis of CXR images, particularly in segmenting lung regions as a critical preliminary step. This study aims to develop and evaluate a lung segmentation model based on a modified U-Net architecture. The architecture leverages techniques such as transfer learning with DenseNet201 as a feature extractor alongside dilated convolutions and residual blocks. An ablation study was conducted to evaluate these architectural components, along with additional elements like augmented data, alternative backbones, and attention mechanisms. Numerous and extensive experiments were performed on two publicly available datasets, the Montgomery County (MC) and Shenzhen Hospital (SH) datasets, to validate the efficacy of these techniques on segmentation performance. Outperforming other state-of-the-art methods on the MC dataset, the proposed model achieved a Jaccard Index (IoU) of 97.77 and a Dice Similarity Coefficient (DSC) of 98.87. These results represent a significant improvement over the baseline U-Net, with gains of 3.37% and 1.75% in IoU and DSC, respectively. These findings highlight the importance of architectural enhancements in deep learning-based lung segmentation models, contributing to more efficient, accurate, and reliable CAD systems for lung disease assessment.http://www.sciencedirect.com/science/article/pii/S2666990025000369Lung segmentationDeep learningComputer aided diagnosticU-netResidual networks
spellingShingle Wiley Tam
Paul Babyn
Javad Alirezaie
Robust lung segmentation in Chest X-ray images using modified U-Net with deeper network and residual blocks
Computer Methods and Programs in Biomedicine Update
Lung segmentation
Deep learning
Computer aided diagnostic
U-net
Residual networks
title Robust lung segmentation in Chest X-ray images using modified U-Net with deeper network and residual blocks
title_full Robust lung segmentation in Chest X-ray images using modified U-Net with deeper network and residual blocks
title_fullStr Robust lung segmentation in Chest X-ray images using modified U-Net with deeper network and residual blocks
title_full_unstemmed Robust lung segmentation in Chest X-ray images using modified U-Net with deeper network and residual blocks
title_short Robust lung segmentation in Chest X-ray images using modified U-Net with deeper network and residual blocks
title_sort robust lung segmentation in chest x ray images using modified u net with deeper network and residual blocks
topic Lung segmentation
Deep learning
Computer aided diagnostic
U-net
Residual networks
url http://www.sciencedirect.com/science/article/pii/S2666990025000369
work_keys_str_mv AT wileytam robustlungsegmentationinchestxrayimagesusingmodifiedunetwithdeepernetworkandresidualblocks
AT paulbabyn robustlungsegmentationinchestxrayimagesusingmodifiedunetwithdeepernetworkandresidualblocks
AT javadalirezaie robustlungsegmentationinchestxrayimagesusingmodifiedunetwithdeepernetworkandresidualblocks