The Potential of Transformer-Based Models for Automated Lung Cancer Detection from CT Scans

Öz

Lung cancer is the most common type of cancer worldwide and the leading cause of cancer-related deaths. Early diagnosis and treatment can significantly increase the survival rate of this disease. Radiological methods used in the diagnosis of lung cancer, especially Computed Tomography (CT) imaging, allow tumors to be detected more precisely. However, manual analysis of these images is time consuming and error prone due to human factors. In this study, we compared the potential of three different transformer-based state-of-the-art models (ViT, DeiT and Swin Transformer) for automatic lung cancer detection. We collected 690 CT images including small cell lung cancer (SCLC), non-small cell lung cancer (NSCLC) and normal findings from a local hospital. Each image was carefully reviewed and labeled by our expert radiologist, and these labeled images were used to train the models. The ViT, DeiT and Swin Transformer models achieved accuracy rates of 91.3%, 84.1% and 80.4% respectively on the test samples. This study shows that the use of transformer-based models for lung cancer classification is promising in overcoming the difficulties in manual analysis.

Anahtar Kelimeler

• Lung cancer
• Transformer models
• CT imaging
• Diagnosis
• Deep learning

Proje Numarası

TEKF.24.42.

Teşekkür

This study was supported by Scientific Research Projects Unit of Firat University (FUBAP) under the Grant Number TEKF.24.42. The authors thank to FUBAP for their supports

Kaynakça

1. [1] A. Leiter, R. R. Veluswamy, and J. P. Wisnivesky, “The global burden of lung cancer: current status and future trends,” Nat. Rev. Clin. Oncol., vol. 20, no. 9, pp. 624–639, Sep. 2023, doi: 10.1038/s41571-023-00798-3.
2. [2] F. Bray, J. Ferlay, I. Soerjomataram, R. L. Siegel, L. A. Torre, and A. Jemal, “Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries,” CA. Cancer J. Clin., vol. 68, no. 6, pp. 394–424, Nov. 2018, doi: 10.3322/caac.21492.
3. [3] Y. Fang et al., “Burden of lung cancer along with attributable risk factors in China from 1990 to 2019, and projections until 2030,” J. Cancer Res. Clin. Oncol., vol. 149, no. 7, pp. 3209–3218, Jul. 2023, doi: 10.1007/s00432-022-04217-5.
4. [4] M. Kriegsmann et al., “Deep Learning for the Classification of Small-Cell and Non-Small-Cell Lung Cancer,” Cancers (Basel)., vol. 12, no. 6, p. 1604, Jun. 2020, doi: 10.3390/cancers12061604.
5. [5] L. E. L. Hendriks et al., “Non-small-cell lung cancer,” Nat. Rev. Dis. Prim., vol. 10, no. 1, p. 71, Sep. 2024, doi: 10.1038/s41572-024-00551-9.
6. [6] R. L. Siegel, K. D. Miller, H. E. Fuchs, and A. Jemal, “Cancer statistics, 2022,” CA. Cancer J. Clin., vol. 72, no. 1, pp. 7–33, Jan. 2022, doi: 10.3322/caac.21708.
7. [7] S. J. Adams, E. Stone, D. R. Baldwin, R. Vliegenthart, P. Lee, and F. J. Fintelmann, “Lung cancer screening,” Lancet, vol. 401, no. 10374, pp. 390–408, Feb. 2023, doi: 10.1016/S0140-6736(22)01694-4.
8. [8] R. Nooreldeen and H. Bach, “Current and Future Development in Lung Cancer Diagnosis,” Int. J. Mol. Sci., vol. 22, no. 16, p. 8661, Aug. 2021, doi: 10.3390/ijms22168661.

9. [9] J. Kim and K. H. Kim, “Role of chest radiographs in early lung cancer detection,” Transl. Lung Cancer Res., vol. 9, no. 3, pp. 522–531, Jun. 2020, doi: 10.21037/tlcr.2020.04.02.
10. [10] B. Philip et al., “Current investigative modalities for detecting and staging lung cancers: a comprehensive summary,” Indian J. Thorac. Cardiovasc. Surg., vol. 39, no. 1, pp. 42–52, Jan. 2023, doi: 10.1007/s12055-022-01430-2.
11. [11] H. L. Lancaster, M. A. Heuvelmans, and M. Oudkerk, “Low‐dose computed tomography lung cancer screening: Clinical evidence and implementation research,” J. Intern. Med., vol. 292, no. 1, pp. 68–80, Jul. 2022, doi: 10.1111/joim.13480.
12. [12] M. A. Thanoon, M. A. Zulkifley, M. A. A. Mohd Zainuri, and S. R. Abdani, “A Review of Deep Learning Techniques for Lung Cancer Screening and Diagnosis Based on CT Images,” Diagnostics, vol. 13, no. 16, p. 2617, Aug. 2023, doi: 10.3390/diagnostics13162617.
13. [13] E. W. Zhang et al., “Characteristics and Outcomes of Lung Cancers Detected on Low-Dose Lung Cancer Screening CT,” Cancer Epidemiol. Biomarkers Prev., vol. 30, no. 8, pp. 1472–1479, Aug. 2021, doi: 10.1158/1055-9965.EPI-20-1847.
14. [14] K.-L. Huang, S.-Y. Wang, W.-C. Lu, Y.-H. Chang, J. Su, and Y.-T. Lu, “Effects of low-dose computed tomography on lung cancer screening: a systematic review, meta-analysis, and trial sequential analysis,” BMC Pulm. Med., vol. 19, no. 1, p. 126, Dec. 2019, doi: 10.1186/s12890-019-0883-x.
15. [15] E. F. Patz, E. Greco, C. Gatsonis, P. Pinsky, B. S. Kramer, and D. R. Aberle, “Lung cancer incidence and mortality in National Lung Screening Trial participants who underwent low-dose CT prevalence screening: a retrospective cohort analysis of a randomised, multicentre, diagnostic screening trial,” Lancet Oncol., vol. 17, no. 5, pp. 590–599, May 2016, doi: 10.1016/S1470-2045(15)00621-X.
16. [16] S. J. van Riel et al., “Observer variability for Lung-RADS categorisation of lung cancer screening CTs: impact on patient management,” Eur. Radiol., vol. 29, no. 2, pp. 924–931, Feb. 2019, doi: 10.1007/s00330-018-5599-4.
17. [17] S. H. Hosseini, R. Monsefi, and S. Shadroo, “Deep learning applications for lung cancer diagnosis: A systematic review,” Multimed. Tools Appl., vol. 83, no. 5, pp. 14305–14335, Jul. 2023, doi: 10.1007/s11042-023-16046-w.
18. [18] R. Javed, T. Abbas, A. H. Khan, A. Daud, A. Bukhari, and R. Alharbey, “Deep learning for lungs cancer detection: a review,” Artif. Intell. Rev., vol. 57, no. 8, p. 197, Jul. 2024, doi: 10.1007/s10462-024-10807-1.
19. [19] J. V. Naga Ramesh, R. Agarwal, P. Deekshita, S. A. Elahi, S. H. Surya Bindu, and J. S. Pavani, “Application of Several Transfer Learning Approach for Early Classification of Lung Cancer,” EAI Endorsed Trans. Pervasive Heal. Technol., vol. 10, Mar. 2024, doi: 10.4108/eetpht.10.5434.
20. [20] H. Li, Q. Song, D. Gui, M. Wang, X. Min, and A. Li, “Reconstruction-Assisted Feature Encoding Network for Histologic Subtype Classification of Non-Small Cell Lung Cancer,” IEEE J. Biomed. Heal. Informatics, vol. 26, no. 9, pp. 4563–4574, Sep. 2022, doi: 10.1109/JBHI.2022.3192010.
21. [21] S. Pang, Y. Zhang, M. Ding, X. Wang, and X. Xie, “A Deep Model for Lung Cancer Type Identification by Densely Connected Convolutional Networks and Adaptive Boosting,” IEEE Access, vol. 8, pp. 4799–4805, 2020, doi: 10.1109/ACCESS.2019.2962862.
22. [22] Y. Han et al., “Histologic subtype classification of non-small cell lung cancer using PET/CT images,” Eur. J. Nucl. Med. Mol. Imaging, vol. 48, no. 2, pp. 350–360, Feb. 2021, doi: 10.1007/s00259-020-04771-5.
23. [23] T. L. Chaunzwa et al., “Deep learning classification of lung cancer histology using CT images,” Sci. Rep., vol. 11, no. 1, p. 5471, Mar. 2021, doi: 10.1038/s41598-021-84630-x.
24. [24] A. Fanizzi et al., “Comparison between vision transformers and convolutional neural networks to predict non-small lung cancer recurrence,” Sci. Rep., vol. 13, no. 1, p. 20605, 2023, doi: 10.1038/s41598-023-48004-9
25. [25] C. Venkatesh et al., “A hybrid model for lung cancer prediction using patch processing and deeplearning on CT images,” Multimedia Tools and Applications, vol. 83, no. 15, p. 43931-43952, 2024, doi: 10.1007/s11042-024-19208-6
26. [26] R. Azad et al., “Advances in medical image analysis with vision Transformers: A comprehensive review,” Med. Image Anal., vol. 91, p. 103000, Jan. 2024, doi: 10.1016/j.media.2023.103000.
27. [27] A. Parvaiz, M. A. Khalid, R. Zafar, H. Ameer, M. Ali, and M. M. Fraz, “Vision Transformers in medical computer vision—A contemplative retrospection,” Eng. Appl. Artif. Intell., vol. 122, p. 106126, Jun. 2023, doi: 10.1016/j.engappai.2023.106126.
28. [28] A. M. Hafiz, S. A. Parah, and R. U. A. Bhat, “Attention mechanisms and deep learning for machine vision: A survey of the state of the art,” Jun. 2021.
29. [29] A. Dosovitskiy et al., “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” Oct. 2020.
30. [30] O. Katar and O. Yildirim, “An Explainable Vision Transformer Model Based White Blood Cells Classification and Localization,” Diagnostics, vol. 13, no. 14, p. 2459, Jul. 2023, doi: 10.3390/diagnostics13142459.
31. [31] H. Touvron, M. Cord, M. Douze, F. Massa, A. Sablayrolles, and H. Jegou, “Training data-efficient image transformers & distillation through attention,” in Proceedings of the 38th International Conference on Machine Learning, M. Meila and T. Zhang, Eds., in Proceedings of Machine Learning Research, vol. 139. PMLR, Oct. 2021, pp. 10347–10357. [Online]. Available: https://proceedings.mlr.press/v139/touvron21a.html
32. [32] G. Habib, T. J. Saleem, and B. Lall, “Knowledge Distillation in Vision Transformers: A Critical Review,” Feb. 2023.
33. [33] Z. Liu et al., “Swin transformer: Hierarchical vision transformer using shifted windows,” in Proceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 10012–10022.
34. [34] J. Huang et al., “Swin transformer for fast MRI,” Neurocomputing, vol. 493, pp. 281–304, Jul. 2022, doi: 10.1016/j.neucom.2022.04.051.