A novel approach based on joint optimization of alignment and statistical surface representation with wavelet transform for CBCT segmentation

Yu Bing Chang, Peng Yuan, Tai Hong Kuo, Zixiang Xiong, Jaime Gateno, James J. Xia, Xiaobo Zhou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Cone-beam computed tomography (CBCT) can provide true 3D information of anatomical structures, with advantages of much thinner slice thickness and significantly lowered effective dose of radiation. However, CBCT images are extremely low contract and noisy. It is very difficult to segment thin bones. It usually takes 4-5 hours to manually segment a set of CBCT data. To this end, we developed a novel approach based on the joint optimization of alignment and statistical surface representation with wavelet transform for segmentation of CBCT images. It included two main steps: customized wavelet base initialization (CWBI) and base invariant wavelet active shape model (BIWASM). We validated our approach with others by comparing the surface deviation between segmented shape to the ground truth. The results showed that our approach outperformed the others in accuracy and computing time.

Original languageEnglish (US)
Title of host publicationComputational Intelligence Methods for Bioinformatics and Biostatistics - 9th International Meeting, CIBB 2012, Revised Selected Papers
Pages48-56
Number of pages9
DOIs
StatePublished - Sep 9 2013
Event9th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2012 - Houston, TX, United States
Duration: Jul 12 2012Jul 14 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7845 LNBI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other9th International Meeting on Computational Intelligence Methods for Bioinformatics and Biostatistics, CIBB 2012
CountryUnited States
CityHouston, TX
Period7/12/127/14/12

Keywords

  • Cone-beam CT
  • Segmentation
  • Statistical Shape Model

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Fingerprint Dive into the research topics of 'A novel approach based on joint optimization of alignment and statistical surface representation with wavelet transform for CBCT segmentation'. Together they form a unique fingerprint.

Cite this