3D prostate segmentation of ultrasound images combining longitudinal image registration and machine learning

Xiaofeng Yang, Baowei Fei

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

32 Citations (Scopus)

Abstract

We developed a three-dimensional (3D) segmentation method for transrectal ultrasound (TRUS) images, which is based on longitudinal image registration and machine learning. Using longitudinal images of each individual patient, we register previously acquired images to the new images of the same subject. Three orthogonal Gabor filter banks were used to extract texture features from each registered image. Patient-specific Gabor features from the registered images are used to train kernel support vector machines (KSVMs) and then to segment the newly acquired prostate image. The segmentation method was tested in TRUS data from five patients. The average surface distance between our and manual segmentation is 1.18 ± 0.31 mm, indicating that our automatic segmentation method based on longitudinal image registration is feasible for segmenting the prostate in TRUS images.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2012
Subtitle of host publicationImage-Guided Procedures, Robotic Interventions, and Modeling
DOIs
StatePublished - May 1 2012
Externally publishedYes
EventMedical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling - San Diego, CA, United States
Duration: Feb 5 2012Feb 7 2012

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume8316
ISSN (Print)1605-7422

Other

OtherMedical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling
CountryUnited States
CitySan Diego, CA
Period2/5/122/7/12

Fingerprint

machine learning
Image registration
learning
Learning systems
Prostate
Ultrasonics
Gabor filters
Filter banks
Support vector machines
Textures
Machine Learning
registers
textures

Keywords

  • image registration
  • image segmentation
  • machine learning
  • prostate cancer
  • support vector machine (SVM)
  • Transrectal ultrasound (TRUS)

ASJC Scopus subject areas

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

Cite this

Yang, X., & Fei, B. (2012). 3D prostate segmentation of ultrasound images combining longitudinal image registration and machine learning. In Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling [83162O] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 8316). https://doi.org/10.1117/12.912188

3D prostate segmentation of ultrasound images combining longitudinal image registration and machine learning. / Yang, Xiaofeng; Fei, Baowei.

Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling. 2012. 83162O (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 8316).

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

Yang, X & Fei, B 2012, 3D prostate segmentation of ultrasound images combining longitudinal image registration and machine learning. in Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling., 83162O, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 8316, Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling, San Diego, CA, United States, 2/5/12. https://doi.org/10.1117/12.912188
Yang X, Fei B. 3D prostate segmentation of ultrasound images combining longitudinal image registration and machine learning. In Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling. 2012. 83162O. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). https://doi.org/10.1117/12.912188
Yang, Xiaofeng ; Fei, Baowei. / 3D prostate segmentation of ultrasound images combining longitudinal image registration and machine learning. Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling. 2012. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).
@inproceedings{3de18e2664b849fb867cd6350ab100be,
title = "3D prostate segmentation of ultrasound images combining longitudinal image registration and machine learning",
abstract = "We developed a three-dimensional (3D) segmentation method for transrectal ultrasound (TRUS) images, which is based on longitudinal image registration and machine learning. Using longitudinal images of each individual patient, we register previously acquired images to the new images of the same subject. Three orthogonal Gabor filter banks were used to extract texture features from each registered image. Patient-specific Gabor features from the registered images are used to train kernel support vector machines (KSVMs) and then to segment the newly acquired prostate image. The segmentation method was tested in TRUS data from five patients. The average surface distance between our and manual segmentation is 1.18 ± 0.31 mm, indicating that our automatic segmentation method based on longitudinal image registration is feasible for segmenting the prostate in TRUS images.",
keywords = "image registration, image segmentation, machine learning, prostate cancer, support vector machine (SVM), Transrectal ultrasound (TRUS)",
author = "Xiaofeng Yang and Baowei Fei",
year = "2012",
month = "5",
day = "1",
doi = "10.1117/12.912188",
language = "English (US)",
isbn = "9780819489654",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
booktitle = "Medical Imaging 2012",

}

TY - GEN

T1 - 3D prostate segmentation of ultrasound images combining longitudinal image registration and machine learning

AU - Yang, Xiaofeng

AU - Fei, Baowei

PY - 2012/5/1

Y1 - 2012/5/1

N2 - We developed a three-dimensional (3D) segmentation method for transrectal ultrasound (TRUS) images, which is based on longitudinal image registration and machine learning. Using longitudinal images of each individual patient, we register previously acquired images to the new images of the same subject. Three orthogonal Gabor filter banks were used to extract texture features from each registered image. Patient-specific Gabor features from the registered images are used to train kernel support vector machines (KSVMs) and then to segment the newly acquired prostate image. The segmentation method was tested in TRUS data from five patients. The average surface distance between our and manual segmentation is 1.18 ± 0.31 mm, indicating that our automatic segmentation method based on longitudinal image registration is feasible for segmenting the prostate in TRUS images.

AB - We developed a three-dimensional (3D) segmentation method for transrectal ultrasound (TRUS) images, which is based on longitudinal image registration and machine learning. Using longitudinal images of each individual patient, we register previously acquired images to the new images of the same subject. Three orthogonal Gabor filter banks were used to extract texture features from each registered image. Patient-specific Gabor features from the registered images are used to train kernel support vector machines (KSVMs) and then to segment the newly acquired prostate image. The segmentation method was tested in TRUS data from five patients. The average surface distance between our and manual segmentation is 1.18 ± 0.31 mm, indicating that our automatic segmentation method based on longitudinal image registration is feasible for segmenting the prostate in TRUS images.

KW - image registration

KW - image segmentation

KW - machine learning

KW - prostate cancer

KW - support vector machine (SVM)

KW - Transrectal ultrasound (TRUS)

UR - http://www.scopus.com/inward/record.url?scp=84860236456&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860236456&partnerID=8YFLogxK

U2 - 10.1117/12.912188

DO - 10.1117/12.912188

M3 - Conference contribution

C2 - 24027622

AN - SCOPUS:84860236456

SN - 9780819489654

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

BT - Medical Imaging 2012

ER -