Home // GLOBAL HEALTH 2024, The Thirteenth International Conference on Global Health Challenges // View article
Authors:
Simin Mirzaei
Hamid Reza Tohidypour
Shahriar Mirabbasi
Panos Nasiopoulos
Keywords: Cone Beam Computed Tomography (CBCT) images; segmentation; denoising.
Abstract:
Segmenting Cone Beam Computed Tomography (CBCT) images is challenging due to high noise levels, artifacts, and limited resolution. This paper evaluates the effectiveness of various segmentation methods on both noisy and denoised CBCT images. We examine the performance of state-of-the-art segmentation techniques on CBCT scans processed with three efficient denoising methods tailored for low-dose CBCT images. Our findings indicate that introducing a denoising step before segmentation significantly enhances the segmentation quality of CBCT images. Additionally, the 3D Slicer approach demonstrates the most robust segmentation performance for both noisy and denoised CBCT images. Among the denoising techniques, Chang et al.’s method proves to be the most effective, yielding promising results across all evaluated segmentation methods.
Pages: 5 to 9
Copyright: Copyright (c) IARIA, 2024
Publication date: September 29, 2024
Published in: conference
ISSN: 2308-4553
ISBN: 978-1-68558-189-3
Location: Venice, Italy
Dates: from September 29, 2024 to October 3, 2024