Segmentation of Pulmonary Cavity in Lung CT Scan for Tuberculosis Disease

Authors

  • Zhuoyi Tan Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Serdang, 43400, Malaysia
  • Hizmawati Madzin Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Serdang, 43400, Malaysia
  • Fatimah Khalid Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Serdang, 43400, Malaysia
  • Ng Seng Beng Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Serdang, 43400, Malaysia

DOI:

https://doi.org/10.37934/araset.33.2.98106

Keywords:

Pulmonary tuberculosis, semantic segmentation, Deep Learning, object detection

Abstract

The complexity of pulmonary tuberculosis (TB) lung cavity lesion features significantly increase the cost of semantic segmentation and labelling.  However, the high cost of semantic segmentation has limited the development of TB automatic recognition to some extent.  To address this issue, we developed an algorithm that automatically generates a semantic segmentation mask of TB from the TB target detection boundary box. Pulmonologists only need to identify and label the location of TB, and the algorithm can automatically generate the semantic segmentation mask of TB lesions in the labelled area.  The algorithm, first, calculates the optimal threshold for separating the lesion from the background region. Then, based on this threshold, the lesion tissue within the bounding box is extracted and forms a mask that can be used for semantic segmentation tasks.  Finally, we use the generated TB semantic segmentation mask to train Unet and Vnet models to verify the effectiveness of the algorithm. The experimental results demonstrate that Unet and Vnet achieve mean Dice coefficients of 0.612 and 0.637, respectively, in identifying TB lesion tissue.

Downloads

Download data is not yet available.

Author Biographies

Zhuoyi Tan, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Serdang, 43400, Malaysia

gs63891@student.upm.edu.my

Hizmawati Madzin, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Serdang, 43400, Malaysia

hizmawati@upm.edu.my

Fatimah Khalid, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Serdang, 43400, Malaysia

fatimahk@upm.edu.my

Ng Seng Beng, Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Serdang, 43400, Malaysia

ngsengbeng@upm.edu.my

Downloads

Published

2023-11-01

How to Cite

Zhuoyi Tan, Hizmawati Madzin, Fatimah Khalid, & Ng Seng Beng. (2023). Segmentation of Pulmonary Cavity in Lung CT Scan for Tuberculosis Disease. Journal of Advanced Research in Applied Sciences and Engineering Technology, 33(2), 98–106. https://doi.org/10.37934/araset.33.2.98106

Issue

Section

Articles