Fully Automatic Lung Segmentation in Thoracic CT Images using K-means Thresholding

DOI

10.2991/978-94-6463-602-4_9How to use a DOI?

Keywords

Lung Segmentation; Computer-Aided Detection; K-means Thresholding; CT Image

Abstract

Lung segmentation can be considered as one of the most important steps in the Computer-Aided Diagnosis (CAD) system for lung cancer at its early stage. Accurate lung segmentation can significantly enhance the efficiency of the CAD system by removing unnecessary parts from the input image. It also helps to reduce challenges in detecting juxta-pleural nodules that show higher malignancy than the other nodule types. This paper uses advanced image processing techniques to present a fully automatic algorithm for lung segmentation of thoracic CT images. The proposed method uses K-means thresholding and various morphological operations to handle juxta-pleural nodules. Closing operation was used for hole filling which preserves objects’ shape, size, and connectivity. Opening operation is applied to smooth object boundaries. The proposed method is tested on forty-two subjects with juxta-pleural nodules (approximately 7,672 CT images) taken from the publicly available dataset LIDC-IDRI. The proposed method demonstrates exceptional performance with a pixel accuracy of 97.28% and a segmentation accuracy of 97.64% based on Jaccard’s index.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Muhammad Basit Khan
AU  - Furqan Shaukat
AU  - Muhammad Abdullah
AU  - Junaid Mir
AU  - Gulistan Raja
PY  - 2024
DA  - 2024/12/24
TI  - Fully Automatic Lung Segmentation in Thoracic CT Images using K-means Thresholding
BT  - Proceedings of the 4th International Conference on Key Enabling Technologies (KEYTECH 2024)
PB  - Atlantis Press
SP  - 62
EP  - 68
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-602-4_9
DO  - 10.2991/978-94-6463-602-4_9
ID  - Khan2024
ER  -