A generative model for segmentation of tumor and organs-at-risk for radiation therapy planning of glioblastoma patients

Mikael Agn, Ian Law, Per Munck Af Rosenschöld, Koen Van Leemput

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

206 Downloads (Pure)

Abstract

We present a fully automated generative method for simultaneous brain tumor and organs-at-risk segmentation in multi-modal magnetic resonance images. The method combines an existing whole-brain segmentation technique with a spatial tumor prior, which uses convolutional restricted Boltzmann machines to model tumor shape. The method is not tuned to any specific imaging protocol and can simultaneously segment the gross tumor volume, peritumoral edema and healthy tissue structures relevant for radiotherapy planning. We validate the method on a manually delineated clinical data set of glioblastoma patients by comparing segmentations of gross tumor volume, brainstem and hippocampus. The preliminary results demonstrate the feasibility of the method.
Original languageEnglish
Title of host publicationSPIE Medical Imaging 2016: Image Processing
Number of pages9
Volume9784
PublisherSPIE - International Society for Optical Engineering
Publication date2016
Article number97841D
DOIs
Publication statusPublished - 2016
EventSPIE Medical Imaging 2016: Image Processing - San Diego, United States
Duration: 27 Feb 20163 Mar 2016
Conference number: 9784

Conference

ConferenceSPIE Medical Imaging 2016
Number9784
CountryUnited States
CitySan Diego
Period27/02/201603/03/2016
SeriesProceedings of SPIE, the International Society for Optical Engineering
Volume9784
ISSN0277-786X

Bibliographical note

Copyright 2016 Society of Photo Optical Instrumentation Engineers. One print or electronic copy may be made for personal use only. Systematic electronic or print reproduction and distribution, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper are prohibited.

Cite this

Agn, M., Law, I., Munck Af Rosenschöld, P., & Van Leemput, K. (2016). A generative model for segmentation of tumor and organs-at-risk for radiation therapy planning of glioblastoma patients. In SPIE Medical Imaging 2016: Image Processing (Vol. 9784). [97841D] SPIE - International Society for Optical Engineering. Proceedings of SPIE, the International Society for Optical Engineering, Vol.. 9784 https://doi.org/10.1117/12.2216814
Agn, Mikael ; Law, Ian ; Munck Af Rosenschöld, Per ; Van Leemput, Koen. / A generative model for segmentation of tumor and organs-at-risk for radiation therapy planning of glioblastoma patients. SPIE Medical Imaging 2016: Image Processing. Vol. 9784 SPIE - International Society for Optical Engineering, 2016. (Proceedings of SPIE, the International Society for Optical Engineering, Vol. 9784).
@inproceedings{2e7fb8116c1043c193f5e50a437b5280,
title = "A generative model for segmentation of tumor and organs-at-risk for radiation therapy planning of glioblastoma patients",
abstract = "We present a fully automated generative method for simultaneous brain tumor and organs-at-risk segmentation in multi-modal magnetic resonance images. The method combines an existing whole-brain segmentation technique with a spatial tumor prior, which uses convolutional restricted Boltzmann machines to model tumor shape. The method is not tuned to any specific imaging protocol and can simultaneously segment the gross tumor volume, peritumoral edema and healthy tissue structures relevant for radiotherapy planning. We validate the method on a manually delineated clinical data set of glioblastoma patients by comparing segmentations of gross tumor volume, brainstem and hippocampus. The preliminary results demonstrate the feasibility of the method.",
author = "Mikael Agn and Ian Law and {Munck Af Rosensch{\"o}ld}, Per and {Van Leemput}, Koen",
note = "Copyright 2016 Society of Photo Optical Instrumentation Engineers. One print or electronic copy may be made for personal use only. Systematic electronic or print reproduction and distribution, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper are prohibited.",
year = "2016",
doi = "10.1117/12.2216814",
language = "English",
volume = "9784",
booktitle = "SPIE Medical Imaging 2016: Image Processing",
publisher = "SPIE - International Society for Optical Engineering",

}

Agn, M, Law, I, Munck Af Rosenschöld, P & Van Leemput, K 2016, A generative model for segmentation of tumor and organs-at-risk for radiation therapy planning of glioblastoma patients. in SPIE Medical Imaging 2016: Image Processing. vol. 9784, 97841D, SPIE - International Society for Optical Engineering, Proceedings of SPIE, the International Society for Optical Engineering, vol. 9784, SPIE Medical Imaging 2016, San Diego, United States, 27/02/2016. https://doi.org/10.1117/12.2216814

A generative model for segmentation of tumor and organs-at-risk for radiation therapy planning of glioblastoma patients. / Agn, Mikael; Law, Ian; Munck Af Rosenschöld, Per; Van Leemput, Koen.

SPIE Medical Imaging 2016: Image Processing. Vol. 9784 SPIE - International Society for Optical Engineering, 2016. 97841D (Proceedings of SPIE, the International Society for Optical Engineering, Vol. 9784).

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

TY - GEN

T1 - A generative model for segmentation of tumor and organs-at-risk for radiation therapy planning of glioblastoma patients

AU - Agn, Mikael

AU - Law, Ian

AU - Munck Af Rosenschöld, Per

AU - Van Leemput, Koen

N1 - Copyright 2016 Society of Photo Optical Instrumentation Engineers. One print or electronic copy may be made for personal use only. Systematic electronic or print reproduction and distribution, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper are prohibited.

PY - 2016

Y1 - 2016

N2 - We present a fully automated generative method for simultaneous brain tumor and organs-at-risk segmentation in multi-modal magnetic resonance images. The method combines an existing whole-brain segmentation technique with a spatial tumor prior, which uses convolutional restricted Boltzmann machines to model tumor shape. The method is not tuned to any specific imaging protocol and can simultaneously segment the gross tumor volume, peritumoral edema and healthy tissue structures relevant for radiotherapy planning. We validate the method on a manually delineated clinical data set of glioblastoma patients by comparing segmentations of gross tumor volume, brainstem and hippocampus. The preliminary results demonstrate the feasibility of the method.

AB - We present a fully automated generative method for simultaneous brain tumor and organs-at-risk segmentation in multi-modal magnetic resonance images. The method combines an existing whole-brain segmentation technique with a spatial tumor prior, which uses convolutional restricted Boltzmann machines to model tumor shape. The method is not tuned to any specific imaging protocol and can simultaneously segment the gross tumor volume, peritumoral edema and healthy tissue structures relevant for radiotherapy planning. We validate the method on a manually delineated clinical data set of glioblastoma patients by comparing segmentations of gross tumor volume, brainstem and hippocampus. The preliminary results demonstrate the feasibility of the method.

U2 - 10.1117/12.2216814

DO - 10.1117/12.2216814

M3 - Article in proceedings

VL - 9784

BT - SPIE Medical Imaging 2016: Image Processing

PB - SPIE - International Society for Optical Engineering

ER -

Agn M, Law I, Munck Af Rosenschöld P, Van Leemput K. A generative model for segmentation of tumor and organs-at-risk for radiation therapy planning of glioblastoma patients. In SPIE Medical Imaging 2016: Image Processing. Vol. 9784. SPIE - International Society for Optical Engineering. 2016. 97841D. (Proceedings of SPIE, the International Society for Optical Engineering, Vol. 9784). https://doi.org/10.1117/12.2216814