A deep convolutional neural network-based automatic delineation strategy for multiple brain metastases stereotactic radiosurgery

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients’ data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.

Original languageEnglish (US)
Article numbere0185844
JournalPLoS One
Volume12
Issue number10
DOIs
StatePublished - Oct 1 2017

Fingerprint

Radiosurgery
metastasis
neural networks
Brain
Brain Neoplasms
Neoplasm Metastasis
Neural networks
brain
Tumors
neoplasms
Image segmentation
Benchmarking
Workflow
ROC Curve
planning
Neoplasms
Planning
Magnetic Resonance Imaging
Learning
Magnetic resonance

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

@article{299582ad0b804ae68e51433169a292b7,
title = "A deep convolutional neural network-based automatic delineation strategy for multiple brain metastases stereotactic radiosurgery",
abstract = "Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients’ data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.",
author = "Yan Liu and Strahinja Stojadinovic and Brian Hrycushko and Zabi Wardak and Steven Lau and Weiguo Lu and Yulong Yan and Jiang, {Steve B.} and Xin Zhen and Robert Timmerman and Lucien Nedzi and Xuejun Gu",
year = "2017",
month = "10",
day = "1",
doi = "10.1371/journal.pone.0185844",
language = "English (US)",
volume = "12",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "10",

}

TY - JOUR

T1 - A deep convolutional neural network-based automatic delineation strategy for multiple brain metastases stereotactic radiosurgery

AU - Liu, Yan

AU - Stojadinovic, Strahinja

AU - Hrycushko, Brian

AU - Wardak, Zabi

AU - Lau, Steven

AU - Lu, Weiguo

AU - Yan, Yulong

AU - Jiang, Steve B.

AU - Zhen, Xin

AU - Timmerman, Robert

AU - Nedzi, Lucien

AU - Gu, Xuejun

PY - 2017/10/1

Y1 - 2017/10/1

N2 - Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients’ data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.

AB - Accurate and automatic brain metastases target delineation is a key step for efficient and effective stereotactic radiosurgery (SRS) treatment planning. In this work, we developed a deep learning convolutional neural network (CNN) algorithm for segmenting brain metastases on contrast-enhanced T1-weighted magnetic resonance imaging (MRI) datasets. We integrated the CNN-based algorithm into an automatic brain metastases segmentation workflow and validated on both Multimodal Brain Tumor Image Segmentation challenge (BRATS) data and clinical patients’ data. Validation on BRATS data yielded average DICE coefficients (DCs) of 0.75±0.07 in the tumor core and 0.81±0.04 in the enhancing tumor, which outperformed most techniques in the 2015 BRATS challenge. Segmentation results of patient cases showed an average of DCs 0.67±0.03 and achieved an area under the receiver operating characteristic curve of 0.98±0.01. The developed automatic segmentation strategy surpasses current benchmark levels and offers a promising tool for SRS treatment planning for multiple brain metastases.

UR - http://www.scopus.com/inward/record.url?scp=85030713901&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030713901&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0185844

DO - 10.1371/journal.pone.0185844

M3 - Article

C2 - 28985229

AN - SCOPUS:85030713901

VL - 12

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 10

M1 - e0185844

ER -