Authors : Venkata Mahesh B.; Ramya B.; Deep Naga Sai G.; Vijayalakshmi V.; B. Sai Kumar

Volume/Issue : Volume 11 - 2026, Issue 6 - June

Google Scholar : https://tinyurl.com/25us8vfs

Scribd : https://tinyurl.com/bdzyzw5t

DOI : https://doi.org/10.38124/ijisrt/26jun1634

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Lung cancer is still the reason people die from cancer around the world. It is responsible, for 18 percent of all deaths that happen because of cancer. Lung cancer is a serious thing that causes a lot of deaths. People are still dying from lung cancer at a high rate .Timely and accurate identification of conditions is crucial for improving the chances of survival in patients.This study introduces a lightweight deep learning framework for detecting lung cancer and classifying multiple classes. It uses the MobileNetV2 architecture combined with Gradient-weighted Class Activation Mapping (Grad-CAM) to improve visual understanding .The proposed system was tested using CT scan images from the Kaggle Mohamed Hany dataset (chest-ctscan-images). This dataset includes 1,500 CT scan images divided into four categories: adenocarcinoma, large cell carcinoma, squamous cell carcinoma, and normal lung tissue. The MobileNetV2 model, which was pre-trained on ImageNet, was adjusted for multi-class classification of these four lung tissue types.

Keywords : Lung Cancer Detection, Deep Learning, Convolutional Neural Networks (CNN), MobileNetV2, CT Scan Images, Medical Image Classification, Transfer Learning, Grad-CAM, Image Segmentation, U-Net, Explainable Artificial Intelligence (XAI) .

References :

1. M. A. Thanoon, M. A. Zulkifley, M. A. A. M. Zainuri, and S. R. Abdani, “Deep learning approaches for lung cancer screening and diagnosis using CT images: A comprehensive review,” Diagnostics, vol. 13, no. 16, pp. 1–27, 2023.
2. A. O. Salau, M. R. Pooja, N. F. Hasani, and S. L. Braide, “Neural network-based risk assessment model for evaluating lung functionality and early asthma prognosis,” Mathematical Modelling of Engineering Problems, vol. 9, no. 4, pp. 1053–1060, 2022.
3. M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L. C. Chen, “MobileNetV2: An efficient architecture using inverted residuals and linear bottleneck layers,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4510–4520, 2018.
4. R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-CAM: Gradient-based localization technique for visual explanations in deep neural networks,” in Proceedings of the IEEE International Conference on Computer Vision, pp. 618–626, 2017.
5. N. Tajbakhsh, L. Jeyaseelan, Q. Li, J. N. Chiang, Z. Wu, and X. Ding, “Handling imperfect datasets: A survey of deep learning techniques for medical image segmentation,” Medical Image Analysis, vol. 63, p. 101693, 2020.
6. R. S. Herbst, D. Morgensztern, and C. Boshoff, “Biological mechanisms and management strategies for non-small cell lung cancer,” Nature, vol. 553, no. 7689, pp. 446–454, 2018.
7. A. Christie et al., “Evaluation of radiologists and computer-aided detection software for lung nodule identification at varying CT dose levels,” European Society of Radiology, 2013.
8. M. Firmino, A. H. Morais, R. M. Mendonc¸a, M. R. Dantas, H. R. Hekis, and R. Valentim, “Review and future directions of computer-aided detection systems for lung cancer using CT imaging,” BioMedical Engineering Online, vol. 13, p. 41, 2014.
9. D. Ardila et al., “End-to-end lung cancer screening using 3D deep learning on low-dose chest CT scans,” Nature Medicine, vol. 25, no. 6, pp. 954–961, 2019.
10. A. Barredo Arrieta et al., “Explainable artificial intelligence: Concepts, taxonomy, opportunities, and challenges for responsible AI,” Information Fusion, vol. 58, pp. 82–115, 2020.
11. Vaid et al., “Explainable AI techniques for clinical risk prediction: Concepts, methodologies, and applications,” Journal of the American Medical Informatics Association, vol. 28, no. 10, pp. 2184–2196, 2021.
12. K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning frame-work for large-scale image recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778, 2016.
13. Z. Riaz, B. Khan, M. S. Islam, S. Khan, and S. Abdullah, “Hybrid neural network architecture for improved lung tumor segmentation from CT scans,” Bioengineering, vol. 10, no. 8, p. 981, 2023.
14. X. Zhang, X. Zhou, M. Lin, and J. Sun, “ShuffleNet: A highly ef-ficient convolutional neural network designed for mobile devices,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 6848–6856, 2018.
15. A. Tripathi, T. Singh, R. R. Nair, and P. Duraisamy, “Enhanced early detection and classification of lung diseases using an improved Mo-bileNetV2 framework,” IEEE Access, 2024.
16. F. M. J. M. Shamrat, S. Azam, A. Karim, R. Islam, Z. Tasnim, P. Ghosh, and F. De Boer, “LungNet22: A fine-tuned deep learning model for multiclass lung disease prediction using X-ray images,” Journal of Personalized Medicine, vol. 12, no. 5, p. 680, 2022.
17. S. K. Lakshmanaprabu, S. N. Mohanty, K. Shankar, N. Arunkumar, and G. Ramirez, “Optimized deep learning approach for lung cancer classification using CT scan images,” Future Generation Computer Systems, vol. 92, pp. 374–382, 2019.
18. M. A. Cifci, “SegChaNet: A novel deep learning model for lung cancer segmentation in CT images,” Applied Bionics and Biomechanics, vol. 2022, 2022.
19. Q. Hu et al., “Lung segmentation from CT scans using mask region-based convolutional neural networks,” Artificial Intelligence in Medicine, vol. 103, 2020.
20. Z. Wu, X. Li, and J. Zuo, “RAD-UNet: Improved deep learning-based semantic segmentation algorithm for lung nodules,” Frontiers in Oncology, vol. 13, 2023.
21. B. Wossene, D. Assefa, A. O. Salau, S. L. Braide, A. Ali, and T. T. Tin, “CT-based lung cancer identification utilizing spatially local-ized integral transform techniques combined with U-Net classification,” Bulletin of Electrical Engineering and Informatics, vol. 15, no. 1, pp. 827–844, 2026.
22. R. Sun, Y. Pang, and W. Li, “Improved Swin Transformer-based algo-rithm for efficient lung cancer image classification and segmentation,” Electronics, vol. 12, no. 4, p. 1024, 2023.
23. S. B. Shuvo and T. B. Mamun, “An automated deep learning framework for detecting and classifying lung nodules in lung cancer diagnosis,” arXiv preprint arXiv:2305.00046, 2023.
24. T. Dao, A. Gu, A. Ratner, V. Smith, C. De Sa, and C. Re´, “Kernel-based theoretical analysis of modern data augmentation techniques,” in Proceedings of the International Conference on Machine Learning, pp. 1528–1537, 2023.
25. E. D. Cubuk, B. Zoph, D. Mane, V. Vasudevan, and Q. V. Le, “Au-toAugment: Learning optimal data augmentation policies directly from data,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 113–123, 2019.
26. F. Zhuang et al., “A comprehensive overview of transfer learning techniques and applications,” Proceedings of the IEEE, vol. 109, no. 1, pp. 43–76, 2021.
27. M. Hammad, M. ElAffendi, A. A. Abd El-Latif, A. A. Ateya, G. Ali, and P. Plawiak, “Explainable artificial intelligence approach for lung cancer detection from CT images using a customized convolutional neural network,” Scientific Reports, vol. 15, pp. 1–15, 2025.