Su Diao, Yajie Wan, Danyi Huang, Shijia Huang, Touseef Sadiq, Mohammad Shahbaz Khan, Lal Hussain, Badr S. Alkahtani, Tehseen Mazhar.
Abstract
Lung cancer remains a leading cause of cancer-related deaths worldwide, with low survival rates often attributed to late-stage diagnosis. To address this critical health challenge, researchers have developed computer-aided diagnosis (CAD) systems that rely on feature extraction from medical images.
Introduction
Lung cancer remains the most commonly diagnosed cancer and the leading cause of cancer death globally, particularly among men. According to recent estimates of 2024, approximately 2.5 million cases of lung cancer occur worldwide each year, accounting for roughly one in eight cancers.
Materials and methods
The dataset for this study was obtained from a publicly accessible web-based repository maintained by LCA the sole national non-DICOM format images from 76 patients utilized in our previous studies in [46–48]. Of these, 568 images belong to SCLC subjects, while 377 images represent NSCLC subjects from CT images
Results and discussions
This study aimed to enhance lung cancer detection through a dual approach. The first approach involved extracting hand-crafted features, including GLCM to capture distinct image characteristics. These features were subsequently fed into robust machine learning algorithms. The second approach explored the potential of deep learning by employing LSTM algorithms.
Conclusions
Lung cancer remains a major global health crisis, characterized by exceptionally low survival rates. As the most prevalent and deadliest cancer worldwide, its incidence has surged dramatically.
Citation: Diao S, Wan Y, Huang D, Huang S, Sadiq T, Khan MS, et al. (2025) Optimizing Bi-LSTM networks for improved lung cancer detection accuracy. PLoS ONE 20(2): e0316136. https://doi.org/10.1371/journal.pone.0316136
Editor: Jawad Rasheed, Istanbul Sabahattin Zaim University: Istanbul Sabahattin Zaim Universitesi, TÜRKIYE
Received: September 30, 2024; Accepted: December 5, 2024; Published: February 24, 2025.
Copyright: © 2025 Diao et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The dataset for this study was obtained from a publicly accessible web-based repository maintained by LCA the sole national non-DICOM format images from 76 patients selected image are accessible at https://www.cancerimagingarchive.net/collection/nsclc-radiogenomics/. The dataset is utilized and published in our previous studies [46–48].
Funding: The author B.A. extend his appreciation to research Supporting Project number (RSPD2025R526), King Saud University, Riyadh, Saudi Arabia.
Competing interests: The authors have declared that no competing interests exist.
