Neural Network–Based Multi-Level Classification for Region-Oriented CT Lung Image Analysis Neural Network–Based Classification
International Journal of Medical Toxicology and Forensic Medicine,
Vol. 16 (2026),
29 Dey 2025
,
Page 1-6
https://doi.org/10.22037/ijmtfm.v16.51398
Abstract
Background: Lung cancer remains a leading cause of mortality worldwide, and computed tomography (CT) imaging plays a vital role in its diagnosis, follow-up, and medico-legal evaluation. However, manual interpretation of CT images is time-consuming and often affected by inter-observer variability, particularly in resource-limited settings.
Methods: This study proposes a region-oriented CT lung image analysis framework that integrates Fuzzy Possibilistic C-Means (FPCM) segmentation with a multi-level neural network classifier. After preprocessing, CT images are segmented to delineate lung parenchyma, nodules, and surrounding tissues. Intensity, texture, and shape-based features are then extracted, and a hierarchical neural network is employed to progressively classify lung regions, abnormalities, and suspicious cancer-related patterns.
Results: Experimental evaluation demonstrates that the proposed framework outperforms conventional clustering techniques and single-level classifiers. Improved performance is observed in terms of accuracy, sensitivity, specificity, and Dice similarity coefficient, along with a reduction in false positive detections across diverse CT lung images.
Conclusion: The proposed automated framework provides an accurate and cost-effective solution for CT lung image analysis. Its robustness and interpretability make it suitable for clinical diagnosis as well as medical toxicology and forensic applications, including deployment in resource-constrained environments.
- CT lung imaging, Lung cancer analysis, Region-based segmentation, FPCM, Neural networks, Multi-level classification
How to Cite
References
[1] World Health Organization. Global cancer observatory: cancer today. Lyon, France: International Agency for Research on Cancer; 2021. [DOI: 10.1093/annonc/mdab345]
[2] Siegel RL, Miller KD, Jemal A. Cancer statistics, 2021. CA Cancer J Clin. 2021;71(1):7–33. [DOI: 10.3322/caac.21654]
[3] Gonzalez RC, Woods RE. Digital image processing. 4th ed. New York: Pearson; 2018. [DOI: 10.5555/3217436]
[4] Jain AK. Fundamentals of digital image processing. Englewood Cliffs, NJ: Prentice Hall; 1989. [DOI: 10.1016/c2009-0-22745-3]
[5] Bezdek JC. Pattern recognition with fuzzy objective function algorithms. New York: Springer; 1981. [DOI: 10.1007/978-1-4757-0450-1]
[6] Krishnapuram R, Keller JM. A possibilistic approach to clustering. IEEE Trans Fuzzy Syst. 1993;1(2):98–110. [DOI: 10.1109/91.227387]
[7] Pal NR, Bezdek JC. On cluster validity for the fuzzy c-means model. IEEE Trans Fuzzy Syst. 1995;3(3):370–9. [DOI: 10.1109/91.413225]
[8] Haralick RM, Shanmugam K, Dinstein I. Textural features for image classification. IEEE Trans Syst Man Cybern. 1973;SMC-3(6):610–21. [DOI: 10.1109/TSMC.1973.4309314]
[9] Armato SG 3rd, McLennan G, Bidaut L, McNitt-Gray MF, Meyer CR, Reeves AP, et al. The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): completed collection of thoracic CT scans with reference nodule annotations. Med Phys. 2011;38(2):915–31. [DOI: 10.1118/1.3528204]
[10] Litjens G, Kooi T, Bejnordi BE, Setio AAA, Ciompi F, Ghafoorian M, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60–88. [DOI: 10.1016/j.media.2017.07.005]
[11] Chan HP, Hadjiiski LM, Zhou C. Computer-aided diagnosis of lung cancer in CT. Acad Radiol. 2008;15(5):535–55. [DOI: 10.1016/j.acra.2008.01.002]
[12] El-Baz A, Gimel’farb G, Falk R. A new approach for automatic analysis of 3D low dose CT images for accurate diagnosis of lung cancer. Conf Proc IEEE Eng Med Biol Soc. 2009;2009:6723–6726. [DOI: 10.1109/iembs.2009.5333114]
[13] Murphy K, van Ginneken B, Schilham AM, de Hoop B, Gietema HA, Prokop M. A large-scale evaluation of automatic pulmonary nodule detection in chest CT using local image features and k-nearest-neighbour classification. Med Image Anal. 2009;13(5):757–70. [DOI: 10.1016/j.media.2009.07.001]
[14] Doi K. Computer-aided diagnosis in medical imaging: historical review, current status and future potential. Comput Med Imaging Graph. 2007;31(4–5):198–211. [DOI: 10.1016/j.compmedimag.2007.02.002]
[15] Suzuki K. Overview of deep learning in medical imaging. Radiol Phys Technol. 2017;10(3):257–73. [DOI: 10.1007/s12194-017-0408-5]
[16] Esteva A, Robicquet A, Ramsundar B, Kuleshov V, DePristo M, Chou K, et al. A guide to deep learning in healthcare. Nat Med. 2019;25(1):24–9. [DOI: 10.1038/s41591-018-0316-z]
[17] Xu X, Tian J, Liu J, Zhang X, Zhang Y. Automated lung cancer detection using CT images. Comput Biol Med. 2014;45:81–9. [DOI: 10.1016/j.compbiomed.2013.11.012]
[18] Yang X, Sun X, Zhang Y, Liu J, Wang H. Hybrid machine learning techniques for lung disease classification. Expert Syst Appl. 2018;92:1–14. [DOI: 10.1016/j.eswa.2017.09.030]
[19] Bishop CM. Neural networks for pattern recognition. Oxford: Oxford University Press; 1995. [DOI: 10.1093/acprof:oso/9780198538646.001.0001]
[20] Kumar D, Wong A, Clausi DA. Lung nodule classification using deep feature extraction in X-ray images. Conf Proc IEEE Eng Med Biol Soc. 2015;2015:1333–6. [DOI: 10.1109/embc.2015.7318621]
- Abstract Viewed: 57 times
- PDF Downloaded: 40 times