Unsupervised Lung Tumor Segmentation in CT Images Using K-Means and Hierarchical Clustering: A Comparative Analysis Toward Early Detection

References

[1] M. P. Hosseini, T. Lu, and M. Karg, “Deep learning for lung cancer segmentation in medical images: A systematic review,” IEEE Rev. Biomed. Eng., vol. 14, pp. 482–494, 2021.

[2] J. Li, X. Wang, and Y. Zhou, “A comparative analysis of clustering-based segmentation techniques for lung nodule detection,” Comput. Med. Imaging Graph., vol. 99, p. 102142, 2022.

[3] X. Chen, Y. Wang, and H. Liu, “A comparative study on clustering techniques for medical image segmentation,” J. Med. Imaging Res., vol. 12, no. 4, pp. 217–230, 2023.

[4] A. Kumar and S. Rathore, “Lung cancer detection using GMM and K-means hybrid clustering,” Int. J. Healthc. Inform., vol. 45, no. 2, pp. 112–121, 2023.

[5] A. Dutta, V. Singh, and N. Dey, “Medical image segmentation using unsupervised learning: A review,” Biomed. Signal Process. Control, vol. 84, p. 104974, 2024.

[6] R. Thakur, N. Chauhan, and S. Kaur, “An efficient clustering-based method for lung CT image segmentation,” Health Technol., vol. 12, pp. 1201–1212, 2022.

[7] T. Zhang, H. Liu, and Q. Wang, “Segmentation of lung tumors from CT images using spatial FCM with Gaussian kernels,” Multimed. Tools Appl., vol. 81, pp. 23789–23810, 2022.

[8] L. Wei and X. Wu, “Improved spectral clustering for medical image segmentation,” Signal Process. Image Commun., vol. 110, p. 116983, 2023.

[9] S. Mehta and A. Sharma, “Lung cancer segmentation using K-means and morphological filtering,” ICT Express, vol. 9, no. 1, pp. 25–33, 2023.

[10] P. Roy, D. Ghosh, and K. Ghosh, “Unsupervised segmentation of lung tumors using hybrid clustering and wavelet transforms,” Comput. Biol. Med., vol. 152, p. 106350, 2023.

[11] B. Singh and A. Choudhury, “Review of lung cancer classification techniques using ML and DL,” Health Inf. Sci. Syst., vol. 12, p. 20, 2024.

[12] M. Al-Dhief and H. Ismail, “Hybrid unsupervised approaches for lung tumor segmentation in noisy CT images,” Sensors, vol. 22, no. 15, p. 5777, 2022.

[13] V. Prasad and N. Agarwal, “Comparative study of medical clustering techniques for tumor identification,” J. Comput. Sci. Technol., vol. 40, no. 2, pp. 203–217, 2023.

[14] C. Kim and E. Park, “K-means optimized by genetic algorithm for lung CT tumor detection,” Expert Syst. Appl., vol. 206, p. 118133, 2022.

[15] H. Zhang, L. Chen, Review on unsupervised learning in medical imaging, J. Healthc. Eng., 2021, 6642871 (2021). https://doi.org/10.1155/2021/6642871.

@article{1711184,
      author = {Paavai J, Naveen A, Diviya K},
      title = {Unsupervised Lung Tumor Segmentation in CT Images Using K-Means and Hierarchical Clustering: A Comparative Analysis Toward Early Detection},
      journal = {Iconic Research And Engineering Journals},
      year = {2025},
      volume = {9},
      number = {4},
      pages = {442-448},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1711184.pdf},
      abstract = {Background: Lung cancer is among a top cause of cancer-related fatalities worldwide, and timely, accurate detection is essential to improving patient survival rates. Examining CT scans manually takes significant time and is susceptible to observer variation, creating the need for automated detection systems. Methods Used: This study presents an unsupervised segmentation framework for lung tumours in CT images using K-means and Hierarchical Clustering. The process involves grayscale conversion, noise filtering, contrast enhancement, clustering-based segmentation, and morphological post-processing. Results Achieved: Performance indicators such as the Dice Similarity Coefficient (DSC) and the Jaccard Index, tumor area, perimeter, and circularity show that hierarchical clustering provides more accurate and morphologically consistent results, while K-means is computationally faster but less precise. Concluding Remarks: The findings support the use of unsupervised clustering as an effective annotation-free approach for tumor segmentation and pave the way for future research into hybrid and deep learning-based models that can enhance segmentation accuracy and clinical applicability.},
      month = {October},
      doi = {https://doi.org/10.64388/IREV9I4-1711184-9501}
  }