Multiscale segmentation of the skull in MR images for MRI-based attenuation correction of combined MR/PET

Xiaofeng Yang, Baowei Fei

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Background and objective Combined magnetic resonance/positron emission tomography (MR/PET) is a relatively new, hybrid imaging modality. MR-based attenuation correction often requires segmentation of the bone on MR images. In this study, we present an automatic segmentation method for the skull on MR images for attenuation correction in brain MR/PET applications. Materials and methods Our method transforms T1-weighted MR images to the Radon domain and then detects the features of the skull image. In the Radon domain we use a bilateral filter to construct a multiscale image series. For the repeated convolution we increase the spatial smoothing in each scale and make the width of the spatial and range Gaussian function doubled in each scale. Two filters with different kernels along the vertical direction are applied along the scales from the coarse to fine levels. The results from a coarse scale give a mask for the next fine scale and supervise the segmentation in the next fine scale. The use of the multiscale bilateral filtering scheme is to improve the robustness of the method for noise MR images. After combining the two filtered sinograms, the reciprocal binary sinogram of the skull is obtained for the reconstruction of the skull image. Results This method has been tested with brain phantom data, simulated brain data, and real MRI data. For real MRI data the Dice overlap ratios are 92.2% ±1.9% between our segmentation and manual segmentation. Conclusions The multiscale segmentation method is robust and accurate and can be used for MRI-based attenuation correction in combined MR/PET.

Original languageEnglish (US)
Pages (from-to)1037-1045
Number of pages9
JournalJournal of the American Medical Informatics Association
Volume20
Issue number6
DOIs
StatePublished - Oct 29 2013
Externally publishedYes

Fingerprint

Skull
Positron-Emission Tomography
Magnetic Resonance Spectroscopy
Radon
Brain
Multimodal Imaging
Computer-Assisted Image Processing
Masks
Noise
Bone and Bones

ASJC Scopus subject areas

  • Health Informatics

Cite this

Multiscale segmentation of the skull in MR images for MRI-based attenuation correction of combined MR/PET. / Yang, Xiaofeng; Fei, Baowei.

In: Journal of the American Medical Informatics Association, Vol. 20, No. 6, 29.10.2013, p. 1037-1045.

Research output: Contribution to journalArticle

@article{c6395708e09e4fdcafa4f00600ffb9f0,
title = "Multiscale segmentation of the skull in MR images for MRI-based attenuation correction of combined MR/PET",
abstract = "Background and objective Combined magnetic resonance/positron emission tomography (MR/PET) is a relatively new, hybrid imaging modality. MR-based attenuation correction often requires segmentation of the bone on MR images. In this study, we present an automatic segmentation method for the skull on MR images for attenuation correction in brain MR/PET applications. Materials and methods Our method transforms T1-weighted MR images to the Radon domain and then detects the features of the skull image. In the Radon domain we use a bilateral filter to construct a multiscale image series. For the repeated convolution we increase the spatial smoothing in each scale and make the width of the spatial and range Gaussian function doubled in each scale. Two filters with different kernels along the vertical direction are applied along the scales from the coarse to fine levels. The results from a coarse scale give a mask for the next fine scale and supervise the segmentation in the next fine scale. The use of the multiscale bilateral filtering scheme is to improve the robustness of the method for noise MR images. After combining the two filtered sinograms, the reciprocal binary sinogram of the skull is obtained for the reconstruction of the skull image. Results This method has been tested with brain phantom data, simulated brain data, and real MRI data. For real MRI data the Dice overlap ratios are 92.2{\%} ±1.9{\%} between our segmentation and manual segmentation. Conclusions The multiscale segmentation method is robust and accurate and can be used for MRI-based attenuation correction in combined MR/PET.",
author = "Xiaofeng Yang and Baowei Fei",
year = "2013",
month = "10",
day = "29",
doi = "10.1136/amiajnl-2012-001544",
language = "English (US)",
volume = "20",
pages = "1037--1045",
journal = "Journal of the American Medical Informatics Association : JAMIA",
issn = "1067-5027",
publisher = "Oxford University Press",
number = "6",

}

TY - JOUR

T1 - Multiscale segmentation of the skull in MR images for MRI-based attenuation correction of combined MR/PET

AU - Yang, Xiaofeng

AU - Fei, Baowei

PY - 2013/10/29

Y1 - 2013/10/29

N2 - Background and objective Combined magnetic resonance/positron emission tomography (MR/PET) is a relatively new, hybrid imaging modality. MR-based attenuation correction often requires segmentation of the bone on MR images. In this study, we present an automatic segmentation method for the skull on MR images for attenuation correction in brain MR/PET applications. Materials and methods Our method transforms T1-weighted MR images to the Radon domain and then detects the features of the skull image. In the Radon domain we use a bilateral filter to construct a multiscale image series. For the repeated convolution we increase the spatial smoothing in each scale and make the width of the spatial and range Gaussian function doubled in each scale. Two filters with different kernels along the vertical direction are applied along the scales from the coarse to fine levels. The results from a coarse scale give a mask for the next fine scale and supervise the segmentation in the next fine scale. The use of the multiscale bilateral filtering scheme is to improve the robustness of the method for noise MR images. After combining the two filtered sinograms, the reciprocal binary sinogram of the skull is obtained for the reconstruction of the skull image. Results This method has been tested with brain phantom data, simulated brain data, and real MRI data. For real MRI data the Dice overlap ratios are 92.2% ±1.9% between our segmentation and manual segmentation. Conclusions The multiscale segmentation method is robust and accurate and can be used for MRI-based attenuation correction in combined MR/PET.

AB - Background and objective Combined magnetic resonance/positron emission tomography (MR/PET) is a relatively new, hybrid imaging modality. MR-based attenuation correction often requires segmentation of the bone on MR images. In this study, we present an automatic segmentation method for the skull on MR images for attenuation correction in brain MR/PET applications. Materials and methods Our method transforms T1-weighted MR images to the Radon domain and then detects the features of the skull image. In the Radon domain we use a bilateral filter to construct a multiscale image series. For the repeated convolution we increase the spatial smoothing in each scale and make the width of the spatial and range Gaussian function doubled in each scale. Two filters with different kernels along the vertical direction are applied along the scales from the coarse to fine levels. The results from a coarse scale give a mask for the next fine scale and supervise the segmentation in the next fine scale. The use of the multiscale bilateral filtering scheme is to improve the robustness of the method for noise MR images. After combining the two filtered sinograms, the reciprocal binary sinogram of the skull is obtained for the reconstruction of the skull image. Results This method has been tested with brain phantom data, simulated brain data, and real MRI data. For real MRI data the Dice overlap ratios are 92.2% ±1.9% between our segmentation and manual segmentation. Conclusions The multiscale segmentation method is robust and accurate and can be used for MRI-based attenuation correction in combined MR/PET.

UR - http://www.scopus.com/inward/record.url?scp=84886296480&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84886296480&partnerID=8YFLogxK

U2 - 10.1136/amiajnl-2012-001544

DO - 10.1136/amiajnl-2012-001544

M3 - Article

C2 - 23761683

AN - SCOPUS:84886296480

VL - 20

SP - 1037

EP - 1045

JO - Journal of the American Medical Informatics Association : JAMIA

JF - Journal of the American Medical Informatics Association : JAMIA

SN - 1067-5027

IS - 6

ER -