Research on Performance Metrics and Augmentation Methods in Lung Nodule Classification
Source abstract
Abstract
Lung nodule classification is crucial for the diagnosis and treatment of lung diseases. However, selecting appropriate metrics to evaluate classifier performance is challenging, due to the prevalence of negative samples over positive ones, resulting in imbalanced datasets. This imbalance often necessitates the augmentation of positive samples to train powerful models effectively. Furthermore, specific medical tasks require tailored augmentation methods, the effectiveness of which merits further exploration based on task objectives. This study conducted a detailed analysis of commonly used metrics in lung nodule detection, examining their characteristics and selecting suitable metrics based on this analysis and our experimental findings. The selected metrics were then applied to assessing different combinations of image augmentation techniques for nodule classification. Ultimately, the most effective metric was identified, leading to the determination of the most advantageous augmentation method combinations.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Lung Cancer Diagnosis and Treatment · Radiomics and Machine Learning in Medical Imaging · COVID-19 diagnosis using AI
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Industrial & Information Systems Engineering, Jeonbuk National University, 567, Baekje-daero, Deokjin-gu, Jeonju-si 54896, Jellabuk-do, Republic of Korea
Department of Management of Technology, Jeonbuk National University, 567, Baekje-daero, Deokjin-gu, Jeonju-si 54896, Jellabuk-do, Republic of Korea
Department of Computer and Information Technology, Purdue University, West Lafayette, IN 47907, USA
Funding
MOTIE funding program
References (30)
Show source-provided references
1. The global burden of lung cancer: Current status and future trends10.1038/s41571-023-00798-3 ↗
2. Missed lung cancer: When, where, and why?10.5152/dir.2016.16187 ↗
3. Parameter-free similarity-aware attention module for medical image classification and segmentation10.1109/TETCI.2022.3199733 ↗
4. Recent advances and clinical applications of deep learning in medical image analysis10.1016/j.media.2022.102444 ↗
5. Dsal: Deeply supervised active learning from strong and weak labelers for biomedical image segmentation10.1109/JBHI.2021.3052320 ↗
6. Kim, B., Kim, J., Lee, J.G., Kim, D.H., Park, S.H., and Ye, J.C. (2019, January 13–17). Unsupervised deformable image registration using cycle-consistent CNN. Proceedings of the Medical Image Computing and Computer Assisted Intervention–MICCAI 2019: 22nd International Conference, Shenzhen, China. Proceedings, Part VI 22.10.1007/978-3-030-32226-7_19 ↗
7. DNetUnet: A semi-supervised CNN of medical image segmentation for super-computing AI service10.1007/s11227-020-03407-7 ↗
8. Lerousseau, M., Vakalopoulou, M., Classe, M., Adam, J., Battistella, E., Carré, A., Estienne, T., Henry, T., Deutsch, E., and Paragios, N. (2020, January 4–8). Weakly supervised multiple instance learning histopathological tumor segmentation. Proceedings of the Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru. Proceedings, Part V 23.10.1007/978-3-030-59722-1_45 ↗
9. Li, Y., Chen, J., Xie, X., Ma, K., and Zheng, Y. (2020, January 4–8). Self-loop uncertainty: A novel pseudo-label for semi-supervised medical image segmentation. Proceedings of the Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru. Proceedings, Part I 23.10.1007/978-3-030-59710-8_60 ↗
10. Survey on deep learning for pulmonary medical imaging10.1007/s11684-019-0726-4 ↗
11. Deep learning and medical image processing for coronavirus pandemic: A survey10.1016/j.scs.2020.102589 ↗
12. The state of the art of deep learning models in medical science and their challenges10.1007/s00530-020-00694-1 ↗
13. An open IoHT-based deep learning framework for online medical image recognition10.1109/JSAC.2020.3020598 ↗
14. A comparison review of transfer learning and self-supervised learning: Definitions, applications, advantages and limitations10.1016/j.eswa.2023.122807 ↗
15. Masud, M., Sikder, N., Nahid, A.A., Bairagi, A.K., and AlZain, M.A. (2021). A machine learning approach to diagnosing lung and colon cancer using a deep learning-based classification framework. Sensors, 21.10.3390/s21030748 ↗
16. Radiomics and artificial intelligence in lung cancer screening10.21037/tlcr-20-708 ↗
17. Distant domain transfer learning for medical imaging10.1109/JBHI.2021.3051470 ↗
18. Liao, W., Xiong, H., Wang, Q., Mo, Y., Li, X., Liu, Y., Chen, Z., Huang, S., and Dou, D. (2022, January 18–22). Muscle: Multi-task self-supervised continual learning to pre-train deep models for x-ray images of multiple body parts. Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, Singapore.10.1007/978-3-031-16452-1_15 ↗
19. Lipton, Z.C., Elkan, C., and Narayanaswamy, B. (2014). Thresholding classifiers to maximize F1 score. arXiv.10.1007/978-3-662-44851-9_15 ↗
20. Chicco, D., and Jurman, G. (2020). The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC Genom., 21.10.1186/s12864-019-6413-7 ↗
21. The impact of class imbalance in classification performance metrics based on the binary confusion matrix10.1016/j.patcog.2019.02.023 ↗
22. A review of medical image data augmentation techniques for deep learning applications10.1111/1754-9485.13261 ↗
23. Fang, T. (2018, January 18–20). A novel computer-aided lung cancer detection method based on transfer learning from GoogLeNet and median intensity projections. Proceedings of the 2018 IEEE International Conference on Computer and Communication Engineering Technology (CCET), Beijing, China.10.1109/CCET.2018.8542189 ↗
24. Gsaxner, C., Roth, P.M., Wallner, J., and Egger, J. (2019). Exploit fully automatic low-level segmented PET data for training high-level deep learning algorithms for the corresponding CT data. PLoS ONE, 14.10.1371/journal.pone.0212550 ↗
25. Stevens, E., Antiga, L., and Viehmann, T. (2020). Deep Learning with PyTorch, Manning Publications.
26. Boughorbel, S., Jarray, F., and El-Anbari, M. (2017). Optimal classifier for imbalanced data using Matthews Correlation Coefficient metric. PLoS ONE, 12.10.1371/journal.pone.0177678 ↗
27. Computer-aided diagnosis of pulmonary nodules on CT scans: Improvement of classification performance with nodule surface features10.1118/1.3140589 ↗
28. Radiological image traits predictive of cancer status in pulmonary nodules10.1158/1078-0432.CCR-15-3102 ↗
29. Automatic detection of subsolid pulmonary nodules in thoracic computed tomography images10.1016/j.media.2013.12.001 ↗
30. Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: The LUNA16 challenge10.1016/j.media.2017.06.015 ↗