Automatic 3D segmentation of the kidney in MR images using wavelet feature extraction and probability shape model

Hamed Akbari, Baowei Fei

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

8 Citations (Scopus)

Abstract

Numerical estimation of the size of the kidney is useful in evaluating conditions of the kidney, especially, when serial MR imaging is performed to evaluate the kidney function. This paper presents a new method for automatic segmentation of the kidney in three-dimensional (3D) MR images, by extracting texture features and statistical matching of geometrical shape of the kidney. A set of Wavelet-based support vector machines (W-SVMs) is trained on the MR images. The W-SVMs capture texture priors of MRI for classification of the kidney and non-kidney tissues in different zones around the kidney boundary. In the segmentation procedure, these W-SVMs are trained to tentatively label each voxel around the kidney model as a kidney or non-kidney voxel by texture matching. A probability kidney model is created using 10 segmented MRI data. The model is initially localized based on the intensity profiles in three directions. The weight functions are defined for each labeled voxel for each Wavelet-based, intensity-based, and model-based label. Consequently, each voxel has three labels and three weights for the Wavelet feature, intensity, and probability model. Using a 3D edge detection method, the model is re-localized and the segmented kidney is modified based on a region growing method in the model region. The probability model is re-localized based on the results and this loop continues until the segmentation converges. Experimental results with mouse MRI data show the good performance of the proposed method in segmenting the kidney in MR images.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2012
Subtitle of host publicationImage Processing
DOIs
StatePublished - May 14 2012
Externally publishedYes
EventMedical Imaging 2012: Image Processing - San Diego, CA, United States
Duration: Feb 6 2012Feb 9 2012

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume8314
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2012: Image Processing
CountryUnited States
CitySan Diego, CA
Period2/6/122/9/12

Fingerprint

kidneys
pattern recognition
Feature extraction
Kidney
Magnetic resonance imaging
Support vector machines
Labels
Textures
textures
Edge detection
Weights and Measures
edge detection
Tissue
Imaging techniques
mice

Keywords

  • Image segmentation
  • Kidney
  • MRI
  • Polycystic kidney disease
  • Probability Shape Model
  • Support Vector Machines
  • Wavelet

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Akbari, H., & Fei, B. (2012). Automatic 3D segmentation of the kidney in MR images using wavelet feature extraction and probability shape model. In Medical Imaging 2012: Image Processing [83143D] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 8314). https://doi.org/10.1117/12.912028

Automatic 3D segmentation of the kidney in MR images using wavelet feature extraction and probability shape model. / Akbari, Hamed; Fei, Baowei.

Medical Imaging 2012: Image Processing. 2012. 83143D (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 8314).

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

Akbari, H & Fei, B 2012, Automatic 3D segmentation of the kidney in MR images using wavelet feature extraction and probability shape model. in Medical Imaging 2012: Image Processing., 83143D, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 8314, Medical Imaging 2012: Image Processing, San Diego, CA, United States, 2/6/12. https://doi.org/10.1117/12.912028
Akbari H, Fei B. Automatic 3D segmentation of the kidney in MR images using wavelet feature extraction and probability shape model. In Medical Imaging 2012: Image Processing. 2012. 83143D. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). https://doi.org/10.1117/12.912028
Akbari, Hamed ; Fei, Baowei. / Automatic 3D segmentation of the kidney in MR images using wavelet feature extraction and probability shape model. Medical Imaging 2012: Image Processing. 2012. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).
@inproceedings{69d72e172c7445ef932c9df3f2360ad5,
title = "Automatic 3D segmentation of the kidney in MR images using wavelet feature extraction and probability shape model",
abstract = "Numerical estimation of the size of the kidney is useful in evaluating conditions of the kidney, especially, when serial MR imaging is performed to evaluate the kidney function. This paper presents a new method for automatic segmentation of the kidney in three-dimensional (3D) MR images, by extracting texture features and statistical matching of geometrical shape of the kidney. A set of Wavelet-based support vector machines (W-SVMs) is trained on the MR images. The W-SVMs capture texture priors of MRI for classification of the kidney and non-kidney tissues in different zones around the kidney boundary. In the segmentation procedure, these W-SVMs are trained to tentatively label each voxel around the kidney model as a kidney or non-kidney voxel by texture matching. A probability kidney model is created using 10 segmented MRI data. The model is initially localized based on the intensity profiles in three directions. The weight functions are defined for each labeled voxel for each Wavelet-based, intensity-based, and model-based label. Consequently, each voxel has three labels and three weights for the Wavelet feature, intensity, and probability model. Using a 3D edge detection method, the model is re-localized and the segmented kidney is modified based on a region growing method in the model region. The probability model is re-localized based on the results and this loop continues until the segmentation converges. Experimental results with mouse MRI data show the good performance of the proposed method in segmenting the kidney in MR images.",
keywords = "Image segmentation, Kidney, MRI, Polycystic kidney disease, Probability Shape Model, Support Vector Machines, Wavelet",
author = "Hamed Akbari and Baowei Fei",
year = "2012",
month = "5",
day = "14",
doi = "10.1117/12.912028",
language = "English (US)",
isbn = "9780819489630",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
booktitle = "Medical Imaging 2012",

}

TY - GEN

T1 - Automatic 3D segmentation of the kidney in MR images using wavelet feature extraction and probability shape model

AU - Akbari, Hamed

AU - Fei, Baowei

PY - 2012/5/14

Y1 - 2012/5/14

N2 - Numerical estimation of the size of the kidney is useful in evaluating conditions of the kidney, especially, when serial MR imaging is performed to evaluate the kidney function. This paper presents a new method for automatic segmentation of the kidney in three-dimensional (3D) MR images, by extracting texture features and statistical matching of geometrical shape of the kidney. A set of Wavelet-based support vector machines (W-SVMs) is trained on the MR images. The W-SVMs capture texture priors of MRI for classification of the kidney and non-kidney tissues in different zones around the kidney boundary. In the segmentation procedure, these W-SVMs are trained to tentatively label each voxel around the kidney model as a kidney or non-kidney voxel by texture matching. A probability kidney model is created using 10 segmented MRI data. The model is initially localized based on the intensity profiles in three directions. The weight functions are defined for each labeled voxel for each Wavelet-based, intensity-based, and model-based label. Consequently, each voxel has three labels and three weights for the Wavelet feature, intensity, and probability model. Using a 3D edge detection method, the model is re-localized and the segmented kidney is modified based on a region growing method in the model region. The probability model is re-localized based on the results and this loop continues until the segmentation converges. Experimental results with mouse MRI data show the good performance of the proposed method in segmenting the kidney in MR images.

AB - Numerical estimation of the size of the kidney is useful in evaluating conditions of the kidney, especially, when serial MR imaging is performed to evaluate the kidney function. This paper presents a new method for automatic segmentation of the kidney in three-dimensional (3D) MR images, by extracting texture features and statistical matching of geometrical shape of the kidney. A set of Wavelet-based support vector machines (W-SVMs) is trained on the MR images. The W-SVMs capture texture priors of MRI for classification of the kidney and non-kidney tissues in different zones around the kidney boundary. In the segmentation procedure, these W-SVMs are trained to tentatively label each voxel around the kidney model as a kidney or non-kidney voxel by texture matching. A probability kidney model is created using 10 segmented MRI data. The model is initially localized based on the intensity profiles in three directions. The weight functions are defined for each labeled voxel for each Wavelet-based, intensity-based, and model-based label. Consequently, each voxel has three labels and three weights for the Wavelet feature, intensity, and probability model. Using a 3D edge detection method, the model is re-localized and the segmented kidney is modified based on a region growing method in the model region. The probability model is re-localized based on the results and this loop continues until the segmentation converges. Experimental results with mouse MRI data show the good performance of the proposed method in segmenting the kidney in MR images.

KW - Image segmentation

KW - Kidney

KW - MRI

KW - Polycystic kidney disease

KW - Probability Shape Model

KW - Support Vector Machines

KW - Wavelet

UR - http://www.scopus.com/inward/record.url?scp=84860730759&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860730759&partnerID=8YFLogxK

U2 - 10.1117/12.912028

DO - 10.1117/12.912028

M3 - Conference contribution

C2 - 24027620

AN - SCOPUS:84860730759

SN - 9780819489630

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2012

ER -