Semi-automatic segmentation of prostate by directional search for edge boundaries

Juha M. Kortelainen, Kari J. Antila, Alain Schmitt, Charles Mougenot, Gösta J. Ehnholm, Rajiv Chopra

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

Abstract

Semi-automatic segmentation of the prostate boundary is presented for the pre-operational images of the MRIguided ultrasonic thermal therapy of the prostate cancer. The specific deformable surface method is based on firstly fitting an ellipsoid on the given manual landmark points, then modifying the shape of the initialization surface mesh by masking out the regions of the separately segmented bladder and rectum, and finally adapting the surface mesh by searching image for the edge boundaries in the direction of the surface normal. The suggested segmentation method combines information from two types of pre-operational MR-images showing different contrast for the tissue structure. Dice similarity coefficient (DSC) between the semi-automatic segmentation and the manual reference was on average 0.89 for a group of N=5 patients having the MRI guided ultrasound thermal treatment. The robustness of the surface fitting method was tested by simulating 30 randomized initialization sets of the landmark points for each patient, and the resulting standard deviation of DSC was 0.01.
Original languageEnglish
Title of host publication22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014
Subtitle of host publicationCommunication Papers Proceedings
EditorsVaclav Skala
PublisherVaclav Skala Union Agency
Pages285-292
Number of pages8
ISBN (Print)978-808694371-8
Publication statusPublished - 1 Jan 2015
MoE publication typeNot Eligible
Event22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014 - Plzen, Czech Republic
Duration: 2 Jun 20145 Jun 2014

Conference

Conference22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014
Abbreviated titleWSCG 2014
CountryCzech Republic
CityPlzen
Period2/06/145/06/14

Fingerprint

landmarks
mesh
rectum
bladder
coefficients
masking
ellipsoids
standard deviation
therapy
ultrasonics
cancer

Keywords

  • Deformable Surface
  • MRI segmentation
  • Prostate

Cite this

Kortelainen, J. M., Antila, K. J., Schmitt, A., Mougenot, C., Ehnholm, G. J., & Chopra, R. (2015). Semi-automatic segmentation of prostate by directional search for edge boundaries. In V. Skala (Ed.), 22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014: Communication Papers Proceedings (pp. 285-292). Vaclav Skala Union Agency.
Kortelainen, Juha M. ; Antila, Kari J. ; Schmitt, Alain ; Mougenot, Charles ; Ehnholm, Gösta J. ; Chopra, Rajiv. / Semi-automatic segmentation of prostate by directional search for edge boundaries. 22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014: Communication Papers Proceedings. editor / Vaclav Skala. Vaclav Skala Union Agency, 2015. pp. 285-292
@inproceedings{72f102191aaa4754ad6e3ada4fd9a3be,
title = "Semi-automatic segmentation of prostate by directional search for edge boundaries",
abstract = "Semi-automatic segmentation of the prostate boundary is presented for the pre-operational images of the MRIguided ultrasonic thermal therapy of the prostate cancer. The specific deformable surface method is based on firstly fitting an ellipsoid on the given manual landmark points, then modifying the shape of the initialization surface mesh by masking out the regions of the separately segmented bladder and rectum, and finally adapting the surface mesh by searching image for the edge boundaries in the direction of the surface normal. The suggested segmentation method combines information from two types of pre-operational MR-images showing different contrast for the tissue structure. Dice similarity coefficient (DSC) between the semi-automatic segmentation and the manual reference was on average 0.89 for a group of N=5 patients having the MRI guided ultrasound thermal treatment. The robustness of the surface fitting method was tested by simulating 30 randomized initialization sets of the landmark points for each patient, and the resulting standard deviation of DSC was 0.01.",
keywords = "Deformable Surface, MRI segmentation, Prostate",
author = "Kortelainen, {Juha M.} and Antila, {Kari J.} and Alain Schmitt and Charles Mougenot and Ehnholm, {G{\"o}sta J.} and Rajiv Chopra",
year = "2015",
month = "1",
day = "1",
language = "English",
isbn = "978-808694371-8",
pages = "285--292",
editor = "Vaclav Skala",
booktitle = "22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014",
publisher = "Vaclav Skala Union Agency",
address = "Czech Republic",

}

Kortelainen, JM, Antila, KJ, Schmitt, A, Mougenot, C, Ehnholm, GJ & Chopra, R 2015, Semi-automatic segmentation of prostate by directional search for edge boundaries. in V Skala (ed.), 22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014: Communication Papers Proceedings. Vaclav Skala Union Agency, pp. 285-292, 22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014, Plzen, Czech Republic, 2/06/14.

Semi-automatic segmentation of prostate by directional search for edge boundaries. / Kortelainen, Juha M.; Antila, Kari J.; Schmitt, Alain; Mougenot, Charles; Ehnholm, Gösta J.; Chopra, Rajiv.

22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014: Communication Papers Proceedings. ed. / Vaclav Skala. Vaclav Skala Union Agency, 2015. p. 285-292.

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - Semi-automatic segmentation of prostate by directional search for edge boundaries

AU - Kortelainen, Juha M.

AU - Antila, Kari J.

AU - Schmitt, Alain

AU - Mougenot, Charles

AU - Ehnholm, Gösta J.

AU - Chopra, Rajiv

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Semi-automatic segmentation of the prostate boundary is presented for the pre-operational images of the MRIguided ultrasonic thermal therapy of the prostate cancer. The specific deformable surface method is based on firstly fitting an ellipsoid on the given manual landmark points, then modifying the shape of the initialization surface mesh by masking out the regions of the separately segmented bladder and rectum, and finally adapting the surface mesh by searching image for the edge boundaries in the direction of the surface normal. The suggested segmentation method combines information from two types of pre-operational MR-images showing different contrast for the tissue structure. Dice similarity coefficient (DSC) between the semi-automatic segmentation and the manual reference was on average 0.89 for a group of N=5 patients having the MRI guided ultrasound thermal treatment. The robustness of the surface fitting method was tested by simulating 30 randomized initialization sets of the landmark points for each patient, and the resulting standard deviation of DSC was 0.01.

AB - Semi-automatic segmentation of the prostate boundary is presented for the pre-operational images of the MRIguided ultrasonic thermal therapy of the prostate cancer. The specific deformable surface method is based on firstly fitting an ellipsoid on the given manual landmark points, then modifying the shape of the initialization surface mesh by masking out the regions of the separately segmented bladder and rectum, and finally adapting the surface mesh by searching image for the edge boundaries in the direction of the surface normal. The suggested segmentation method combines information from two types of pre-operational MR-images showing different contrast for the tissue structure. Dice similarity coefficient (DSC) between the semi-automatic segmentation and the manual reference was on average 0.89 for a group of N=5 patients having the MRI guided ultrasound thermal treatment. The robustness of the surface fitting method was tested by simulating 30 randomized initialization sets of the landmark points for each patient, and the resulting standard deviation of DSC was 0.01.

KW - Deformable Surface

KW - MRI segmentation

KW - Prostate

UR - http://www.scopus.com/inward/record.url?scp=84957879170&partnerID=8YFLogxK

M3 - Conference article in proceedings

SN - 978-808694371-8

SP - 285

EP - 292

BT - 22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014

A2 - Skala, Vaclav

PB - Vaclav Skala Union Agency

ER -

Kortelainen JM, Antila KJ, Schmitt A, Mougenot C, Ehnholm GJ, Chopra R. Semi-automatic segmentation of prostate by directional search for edge boundaries. In Skala V, editor, 22nd International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, WSCG 2014: Communication Papers Proceedings. Vaclav Skala Union Agency. 2015. p. 285-292