🤖 AI Summary
To address the critical shortage of radiologists and limited diagnostic capacity for early lung cancer screening in resource-constrained settings, this paper proposes a lightweight, end-to-end deep learning framework comprising a three-stage cascade: 3D Res-U-Net for lung segmentation, YOLOv5 for nodule detection, and Vision Transformer for benign–malignant classification. Specifically optimized for low-compute environments and minimal clinical expertise requirements, the framework achieves state-of-the-art performance on the LUNA16 benchmark: a Dice coefficient of 98.82% for lung segmentation, an mAP@0.5 of 0.76 for nodule detection (with markedly reduced false positives), and a classification accuracy of 93.57%—surpassing current SOTA by 1.21%. To our knowledge, this is the first fully automated CT analysis pipeline tailored for primary healthcare settings, delivering both high diagnostic accuracy and practical deployability. The framework significantly enhances accessibility and reliability of early lung cancer screening in underserved regions.
📝 Abstract
Lung cancer is a leading cause of cancer-related deaths worldwide, and early detection is crucial for improving patient outcomes. Nevertheless, early diagnosis of cancer is a major challenge, particularly in low-resource settings where access to medical resources and trained radiologists is limited. The objective of this study is to propose an automated end-to-end deep learning-based framework for the early detection and classification of lung nodules, specifically for low-resource settings. The proposed framework consists of three stages: lung segmentation using a modified 3D U-Net named 3D Res-U-Net, nodule detection using YOLO-v5, and classification with a Vision Transformer-based architecture. We evaluated the proposed framework on a publicly available dataset, LUNA16. The proposed framework's performance was measured using the respective domain's evaluation matrices. The proposed framework achieved a 98.82% lung segmentation dice score while detecting the lung nodule with 0.76 mAP@50 from the segmented lung, at a low false-positive rate. The performance of both networks of the proposed framework was compared with other studies and found to outperform them regarding segmentation and detection accuracy. Additionally, our proposed Vision transformer network obtained an accuracy of 93.57%, which is 1.21% higher than the state-of-the-art networks. Our proposed end-to-end deep learning-based framework can effectively segment lungs, and detect and classify lung nodules, specifically in low-resource settings with limited access to radiologists. The proposed framework outperforms existing studies regarding all the respective evaluation metrics. The proposed framework can potentially improve the accuracy and efficiency of lung cancer screening in low-resource settings, ultimately leading to better patient outcomes.
