A recursive ensemble organ segmentation (REOS) framework

Application in brain radiotherapy

Haibin Chen, Weiguo Lu, Mingli Chen, Linghong Zhou, Robert Timmerman, Dan Tu, Lucien A Nedzi, Zabihullah Wardak, Steve B Jiang, Xin Zhen, Xuejun Gu

Research output: Contribution to journalArticle

Abstract

The aim of this work is to develop a novel recursive ensemble OARs segmentation (REOS) framework for accurate organs-at-risk (OARs) automatic segmentation. The REOS recursively segment individual OARs by ensembling images features extracted from an organ localization module and a contour detection module. Both modules are based on a 3D U-Net architecture. The organ localization module is trained for rough segmentation to localize a region of interest (ROI) that encompasses the to-be-delineated OAR, while the contour detection module is trained to segment the OAR within the identified ROI. In this study, the developed REOS framework is applied for brain radiotherapy on segmenting six OARs including the eyes, the brainstem (BS), the optical nerves and the chiasm. Eighty T1-weighted magnetic resonance images (MRI) from 80 brain cancer patients' cases with OARs' gold standard contours were collected for training and testing REOS. On 20 testing cases, the REOS achieve a high segmentation accuracy with Dice similarity coefficient (DSC) mean and standard deviation of 93.9% ± 1.4%, 94.5% ± 2.0%, 90.6% ± 2.7%, on the left and right eyes and the BS, respectively. On small and segmentation-challenging organs, the left and right optical nerves and the chiasm, the REOS achieves DSC of 78.0% ± 10.5%, 82.2% ± 5.9% and 71.1% ± 9.1%. The satisfactory performances demonstrated the effectiveness of the REOS in OARs segmentation.

Original languageEnglish (US)
Article number025015
JournalPhysics in Medicine and Biology
Volume64
Issue number2
DOIs
StatePublished - Jan 11 2019

Fingerprint

Organs at Risk
Radiotherapy
Brain
Brain Stem
Brain Neoplasms
Magnetic Resonance Spectroscopy

Keywords

  • brain radiotherapy
  • deep learning
  • segmentation

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging

Cite this

A recursive ensemble organ segmentation (REOS) framework : Application in brain radiotherapy. / Chen, Haibin; Lu, Weiguo; Chen, Mingli; Zhou, Linghong; Timmerman, Robert; Tu, Dan; Nedzi, Lucien A; Wardak, Zabihullah; Jiang, Steve B; Zhen, Xin; Gu, Xuejun.

In: Physics in Medicine and Biology, Vol. 64, No. 2, 025015, 11.01.2019.

Research output: Contribution to journalArticle

@article{41cdc59ccf6443e1b2adb14593785968,
title = "A recursive ensemble organ segmentation (REOS) framework: Application in brain radiotherapy",
abstract = "The aim of this work is to develop a novel recursive ensemble OARs segmentation (REOS) framework for accurate organs-at-risk (OARs) automatic segmentation. The REOS recursively segment individual OARs by ensembling images features extracted from an organ localization module and a contour detection module. Both modules are based on a 3D U-Net architecture. The organ localization module is trained for rough segmentation to localize a region of interest (ROI) that encompasses the to-be-delineated OAR, while the contour detection module is trained to segment the OAR within the identified ROI. In this study, the developed REOS framework is applied for brain radiotherapy on segmenting six OARs including the eyes, the brainstem (BS), the optical nerves and the chiasm. Eighty T1-weighted magnetic resonance images (MRI) from 80 brain cancer patients' cases with OARs' gold standard contours were collected for training and testing REOS. On 20 testing cases, the REOS achieve a high segmentation accuracy with Dice similarity coefficient (DSC) mean and standard deviation of 93.9{\%} ± 1.4{\%}, 94.5{\%} ± 2.0{\%}, 90.6{\%} ± 2.7{\%}, on the left and right eyes and the BS, respectively. On small and segmentation-challenging organs, the left and right optical nerves and the chiasm, the REOS achieves DSC of 78.0{\%} ± 10.5{\%}, 82.2{\%} ± 5.9{\%} and 71.1{\%} ± 9.1{\%}. The satisfactory performances demonstrated the effectiveness of the REOS in OARs segmentation.",
keywords = "brain radiotherapy, deep learning, segmentation",
author = "Haibin Chen and Weiguo Lu and Mingli Chen and Linghong Zhou and Robert Timmerman and Dan Tu and Nedzi, {Lucien A} and Zabihullah Wardak and Jiang, {Steve B} and Xin Zhen and Xuejun Gu",
year = "2019",
month = "1",
day = "11",
doi = "10.1088/1361-6560/aaf83c",
language = "English (US)",
volume = "64",
journal = "Physics in Medicine and Biology",
issn = "0031-9155",
publisher = "IOP Publishing Ltd.",
number = "2",

}

TY - JOUR

T1 - A recursive ensemble organ segmentation (REOS) framework

T2 - Application in brain radiotherapy

AU - Chen, Haibin

AU - Lu, Weiguo

AU - Chen, Mingli

AU - Zhou, Linghong

AU - Timmerman, Robert

AU - Tu, Dan

AU - Nedzi, Lucien A

AU - Wardak, Zabihullah

AU - Jiang, Steve B

AU - Zhen, Xin

AU - Gu, Xuejun

PY - 2019/1/11

Y1 - 2019/1/11

N2 - The aim of this work is to develop a novel recursive ensemble OARs segmentation (REOS) framework for accurate organs-at-risk (OARs) automatic segmentation. The REOS recursively segment individual OARs by ensembling images features extracted from an organ localization module and a contour detection module. Both modules are based on a 3D U-Net architecture. The organ localization module is trained for rough segmentation to localize a region of interest (ROI) that encompasses the to-be-delineated OAR, while the contour detection module is trained to segment the OAR within the identified ROI. In this study, the developed REOS framework is applied for brain radiotherapy on segmenting six OARs including the eyes, the brainstem (BS), the optical nerves and the chiasm. Eighty T1-weighted magnetic resonance images (MRI) from 80 brain cancer patients' cases with OARs' gold standard contours were collected for training and testing REOS. On 20 testing cases, the REOS achieve a high segmentation accuracy with Dice similarity coefficient (DSC) mean and standard deviation of 93.9% ± 1.4%, 94.5% ± 2.0%, 90.6% ± 2.7%, on the left and right eyes and the BS, respectively. On small and segmentation-challenging organs, the left and right optical nerves and the chiasm, the REOS achieves DSC of 78.0% ± 10.5%, 82.2% ± 5.9% and 71.1% ± 9.1%. The satisfactory performances demonstrated the effectiveness of the REOS in OARs segmentation.

AB - The aim of this work is to develop a novel recursive ensemble OARs segmentation (REOS) framework for accurate organs-at-risk (OARs) automatic segmentation. The REOS recursively segment individual OARs by ensembling images features extracted from an organ localization module and a contour detection module. Both modules are based on a 3D U-Net architecture. The organ localization module is trained for rough segmentation to localize a region of interest (ROI) that encompasses the to-be-delineated OAR, while the contour detection module is trained to segment the OAR within the identified ROI. In this study, the developed REOS framework is applied for brain radiotherapy on segmenting six OARs including the eyes, the brainstem (BS), the optical nerves and the chiasm. Eighty T1-weighted magnetic resonance images (MRI) from 80 brain cancer patients' cases with OARs' gold standard contours were collected for training and testing REOS. On 20 testing cases, the REOS achieve a high segmentation accuracy with Dice similarity coefficient (DSC) mean and standard deviation of 93.9% ± 1.4%, 94.5% ± 2.0%, 90.6% ± 2.7%, on the left and right eyes and the BS, respectively. On small and segmentation-challenging organs, the left and right optical nerves and the chiasm, the REOS achieves DSC of 78.0% ± 10.5%, 82.2% ± 5.9% and 71.1% ± 9.1%. The satisfactory performances demonstrated the effectiveness of the REOS in OARs segmentation.

KW - brain radiotherapy

KW - deep learning

KW - segmentation

UR - http://www.scopus.com/inward/record.url?scp=85059829737&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059829737&partnerID=8YFLogxK

U2 - 10.1088/1361-6560/aaf83c

DO - 10.1088/1361-6560/aaf83c

M3 - Article

VL - 64

JO - Physics in Medicine and Biology

JF - Physics in Medicine and Biology

SN - 0031-9155

IS - 2

M1 - 025015

ER -