Lung Segmentation With Lightweight Convolutional Attention Residual U-Net

Authors

Meftahul Jannat, Rajshahi University of Engineering and TechnologyFollow
Shaikh Afnan Birahim, Royal University of DhakaFollow
Mohammad Asif Hasan, Rajshahi University of Engineering and TechnologyFollow
Tonmoy Roy, Utah State UniversityFollow
Lubna Sultana, International Islamic University ChittagongFollow
Hasan Sarker, Rajshahi University of Engineering and TechnologyFollow
Samia Fairuz, Rajshahi University of Engineering and TechnologyFollow
Hanaa A. Abdallah, Princess Nourah bint Abdulrahman UniversityFollow

Document Type

Article

Journal/Book Title/Conference

Diagnostics

Author ORCID Identifier

Meftahul Jannat https://orcid.org/0009-0004-4380-9114

Shaikh Afnan Birahim https://orcid.org/0000-0003-4358-2939

Mohammad Asif Hasan https://orcid.org/0009-0003-6447-2541

Hanaa A. Abdallah https://orcid.org/0000-0003-0307-1384

Volume

15

Issue

7

Publisher

MDPI AG

Publication Date

3-27-2025

Journal Article Version

Version of Record

First Page

1

Last Page

25

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 License.

Abstract

Background: Examining chest radiograph images (CXR) is an intricate and time-consuming process, sometimes requiring the identification of many anomalies at the same time. Lung segmentation is key to overcoming this challenge through different deep learning (DL) techniques. Many researchers are working to improve the performance and efficiency of lung segmentation models. This article presents a DL-based approach to accurately identify the lung mask region in CXR images to assist radiologists in recognizing early signs of high-risk lung diseases. Methods: This paper proposes a novel technique, Lightweight Residual U-Net, combining the strengths of the convolutional block attention module (CBAM), the Atrous Spatial Pyramid Pooling (ASPP) block, and the attention module, which consists of only 3.24 million trainable parameters. Furthermore, the proposed model has been trained using both the RELU and LeakyReLU activation functions, with LeakyReLU yielding superior performance. The study indicates that the Dice loss function is more effective in achieving better results. Results: The proposed model is evaluated on three benchmark datasets: JSRT, SZ, and MC, achieving a Dice score of 98.72%, 97.49%, and 99.08%, respectively, outperforming the state-of-the-art models. Conclusions: Using the capabilities of DL and cutting-edge attention processes, the proposed model improves current efforts to enhance lung segmentation for the early identification of many serious lung diseases.

Recommended Citation

Jannat, M.; Birahim, S.A.; Hasan, M.A.; Roy, T.; Sultana, L.; Sarker, H.; Fairuz, S.; Abdallah, H.A. Lung Segmentation with Lightweight Convolutional Attention Residual U-Net. Diagnostics 2025, 15, 854. https://doi.org/10.3390/diagnostics15070854