Automated segmentation of dental CBCT image with prior-guided sequential random forests

Li Wang, Yaozong Gao, Feng Shi, Gang Li, Ken Chung Chen, Zhen Tang, James J. Xia, Dinggang Shen

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Purpose: Cone-beam computed tomography (CBCT) is an increasingly utilized imaging modality for the diagnosis and treatment planning of the patients with craniomaxillofacial (CMF) deformities. Accurate segmentation of CBCT image is an essential step to generate 3D models for the diagnosis and treatment planning of the patients with CMF deformities. However, due to the image artifacts caused by beam hardening, imaging noise, inhomogeneity, truncation, and maximal intercuspation, it is difficult to segment the CBCT. Methods: In this paper, the authors present a new automatic segmentation method to address these problems. Specifically, the authors first employ a majority voting method to estimate the initial segmentation probability maps of both mandible and maxilla based on multiple aligned expert-segmented CBCT images. These probability maps provide an important prior guidance for CBCT segmentation. The authors then extract both the appearance features from CBCTs and the context features from the initial probability maps to train the first-layer of random forest classifier that can select discriminative features for segmentation. Based on the first-layer of trained classifier, the probability maps are updated, which will be employed to further train the next layer of random forest classifier. By iteratively training the subsequent random forest classifier using both the original CBCT features and the updated segmentation probability maps, a sequence of classifiers can be derived for accurate segmentation of CBCT images. Results: Segmentation results on CBCTs of 30 subjects were both quantitatively and qualitatively validated based on manually labeled ground truth. The average Dice ratios of mandible and maxilla by the authors' method were 0.94 and 0.91, respectively, which are significantly better than the state-of-the-art method based on sparse representation (p-value <0.001). Conclusions: The authors have developed and validated a novel fully automated method for CBCT segmentation.

Original languageEnglish
Pages (from-to)336-346
Number of pages11
JournalMedical Physics
Volume43
Issue number1
DOIs
Publication statusPublished - 2016 Jan 1

Fingerprint

Cone-Beam Computed Tomography
Tooth
Maxilla
Mandible
Forests
Politics
Artifacts
Noise
Therapeutics

Keywords

  • atlas based segmentation
  • CBCT
  • craniomaxillofacial deformities
  • maximal intercuspation
  • prior knowledge
  • random forest

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

Automated segmentation of dental CBCT image with prior-guided sequential random forests. / Wang, Li; Gao, Yaozong; Shi, Feng; Li, Gang; Chen, Ken Chung; Tang, Zhen; Xia, James J.; Shen, Dinggang.

In: Medical Physics, Vol. 43, No. 1, 01.01.2016, p. 336-346.

Research output: Contribution to journalArticle

Wang, L, Gao, Y, Shi, F, Li, G, Chen, KC, Tang, Z, Xia, JJ & Shen, D 2016, 'Automated segmentation of dental CBCT image with prior-guided sequential random forests', Medical Physics, vol. 43, no. 1, pp. 336-346. https://doi.org/10.1118/1.4938267
Wang, Li ; Gao, Yaozong ; Shi, Feng ; Li, Gang ; Chen, Ken Chung ; Tang, Zhen ; Xia, James J. ; Shen, Dinggang. / Automated segmentation of dental CBCT image with prior-guided sequential random forests. In: Medical Physics. 2016 ; Vol. 43, No. 1. pp. 336-346.
@article{c4e3eb21a51e46549b2d661613896751,
title = "Automated segmentation of dental CBCT image with prior-guided sequential random forests",
abstract = "Purpose: Cone-beam computed tomography (CBCT) is an increasingly utilized imaging modality for the diagnosis and treatment planning of the patients with craniomaxillofacial (CMF) deformities. Accurate segmentation of CBCT image is an essential step to generate 3D models for the diagnosis and treatment planning of the patients with CMF deformities. However, due to the image artifacts caused by beam hardening, imaging noise, inhomogeneity, truncation, and maximal intercuspation, it is difficult to segment the CBCT. Methods: In this paper, the authors present a new automatic segmentation method to address these problems. Specifically, the authors first employ a majority voting method to estimate the initial segmentation probability maps of both mandible and maxilla based on multiple aligned expert-segmented CBCT images. These probability maps provide an important prior guidance for CBCT segmentation. The authors then extract both the appearance features from CBCTs and the context features from the initial probability maps to train the first-layer of random forest classifier that can select discriminative features for segmentation. Based on the first-layer of trained classifier, the probability maps are updated, which will be employed to further train the next layer of random forest classifier. By iteratively training the subsequent random forest classifier using both the original CBCT features and the updated segmentation probability maps, a sequence of classifiers can be derived for accurate segmentation of CBCT images. Results: Segmentation results on CBCTs of 30 subjects were both quantitatively and qualitatively validated based on manually labeled ground truth. The average Dice ratios of mandible and maxilla by the authors' method were 0.94 and 0.91, respectively, which are significantly better than the state-of-the-art method based on sparse representation (p-value <0.001). Conclusions: The authors have developed and validated a novel fully automated method for CBCT segmentation.",
keywords = "atlas based segmentation, CBCT, craniomaxillofacial deformities, maximal intercuspation, prior knowledge, random forest",
author = "Li Wang and Yaozong Gao and Feng Shi and Gang Li and Chen, {Ken Chung} and Zhen Tang and Xia, {James J.} and Dinggang Shen",
year = "2016",
month = "1",
day = "1",
doi = "10.1118/1.4938267",
language = "English",
volume = "43",
pages = "336--346",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "1",

}

TY - JOUR

T1 - Automated segmentation of dental CBCT image with prior-guided sequential random forests

AU - Wang, Li

AU - Gao, Yaozong

AU - Shi, Feng

AU - Li, Gang

AU - Chen, Ken Chung

AU - Tang, Zhen

AU - Xia, James J.

AU - Shen, Dinggang

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Purpose: Cone-beam computed tomography (CBCT) is an increasingly utilized imaging modality for the diagnosis and treatment planning of the patients with craniomaxillofacial (CMF) deformities. Accurate segmentation of CBCT image is an essential step to generate 3D models for the diagnosis and treatment planning of the patients with CMF deformities. However, due to the image artifacts caused by beam hardening, imaging noise, inhomogeneity, truncation, and maximal intercuspation, it is difficult to segment the CBCT. Methods: In this paper, the authors present a new automatic segmentation method to address these problems. Specifically, the authors first employ a majority voting method to estimate the initial segmentation probability maps of both mandible and maxilla based on multiple aligned expert-segmented CBCT images. These probability maps provide an important prior guidance for CBCT segmentation. The authors then extract both the appearance features from CBCTs and the context features from the initial probability maps to train the first-layer of random forest classifier that can select discriminative features for segmentation. Based on the first-layer of trained classifier, the probability maps are updated, which will be employed to further train the next layer of random forest classifier. By iteratively training the subsequent random forest classifier using both the original CBCT features and the updated segmentation probability maps, a sequence of classifiers can be derived for accurate segmentation of CBCT images. Results: Segmentation results on CBCTs of 30 subjects were both quantitatively and qualitatively validated based on manually labeled ground truth. The average Dice ratios of mandible and maxilla by the authors' method were 0.94 and 0.91, respectively, which are significantly better than the state-of-the-art method based on sparse representation (p-value <0.001). Conclusions: The authors have developed and validated a novel fully automated method for CBCT segmentation.

AB - Purpose: Cone-beam computed tomography (CBCT) is an increasingly utilized imaging modality for the diagnosis and treatment planning of the patients with craniomaxillofacial (CMF) deformities. Accurate segmentation of CBCT image is an essential step to generate 3D models for the diagnosis and treatment planning of the patients with CMF deformities. However, due to the image artifacts caused by beam hardening, imaging noise, inhomogeneity, truncation, and maximal intercuspation, it is difficult to segment the CBCT. Methods: In this paper, the authors present a new automatic segmentation method to address these problems. Specifically, the authors first employ a majority voting method to estimate the initial segmentation probability maps of both mandible and maxilla based on multiple aligned expert-segmented CBCT images. These probability maps provide an important prior guidance for CBCT segmentation. The authors then extract both the appearance features from CBCTs and the context features from the initial probability maps to train the first-layer of random forest classifier that can select discriminative features for segmentation. Based on the first-layer of trained classifier, the probability maps are updated, which will be employed to further train the next layer of random forest classifier. By iteratively training the subsequent random forest classifier using both the original CBCT features and the updated segmentation probability maps, a sequence of classifiers can be derived for accurate segmentation of CBCT images. Results: Segmentation results on CBCTs of 30 subjects were both quantitatively and qualitatively validated based on manually labeled ground truth. The average Dice ratios of mandible and maxilla by the authors' method were 0.94 and 0.91, respectively, which are significantly better than the state-of-the-art method based on sparse representation (p-value <0.001). Conclusions: The authors have developed and validated a novel fully automated method for CBCT segmentation.

KW - atlas based segmentation

KW - CBCT

KW - craniomaxillofacial deformities

KW - maximal intercuspation

KW - prior knowledge

KW - random forest

UR - http://www.scopus.com/inward/record.url?scp=84953857928&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84953857928&partnerID=8YFLogxK

U2 - 10.1118/1.4938267

DO - 10.1118/1.4938267

M3 - Article

VL - 43

SP - 336

EP - 346

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 1

ER -