Improving Image Segmentation for Extrapulmonary Tuberculosis Using PET/CT
Author Information
Author(s): Avazpour Iman, Saripan M Iqbal, Nordin Abdul Jalil, Abdullah Raja Syamsul Azmir Raja
Primary Institution: Universiti Putra Malaysia
Hypothesis
Can the hot spot values in PET imaging improve the segmentation of CT images for better diagnosis of extrapulmonary tuberculosis?
Conclusion
The proposed methods yield fairly good performance in terms of over- and under-segmentation area, with the sliding windows approach showing the best accuracy.
Supporting Evidence
- The proposed segmentation methods were evaluated based on under-segmentation and over-segmentation percentages.
- SRG with local averaging and variance yielded the best results for over-segmentation.
- The modified SRG with local averaging and variance showed the best performance in terms of time complexity.
Takeaway
This study shows a new way to help doctors see and understand images of tuberculosis better by using two types of scans together.
Methodology
The study used a modified seeded region growing technique to segment images from PET/CT scans of patients with extrapulmonary tuberculosis.
Limitations
The study primarily focused on a small sample size and may not generalize to all cases of extrapulmonary tuberculosis.
Participant Demographics
Patients with extrapulmonary tuberculosis were included in the study.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website