Computer-assisted framework for machine-learning-based delineation of GTV regions on datasets of planning CT and PET/CT images

Koujiro Ikushima, Hidetaka Arimura, Ze Jin, Hidetake Yabuuchi, Jumpei Kuwazuru, Yoshiyuki Shioyama, Tomonari Sasaki, Hiroshi Honda, Masayuki Sasaki

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

We have proposed a computer-assisted framework for machine-learning-based delineation of gross tumor volumes (GTVs) following an optimum contour selection (OCS) method. The key idea of the proposed framework was to feed image features around GTV contours (determined based on the knowledge of radiation oncologists) into a machine-learning classifier during the training step, after which the classifier produces the 'degree of GTV' for each voxel in the testing step. Initial GTV regions were extracted using a support vector machine (SVM) that learned the image features inside and outside each tumor region (determined by radiation oncologists). The leave-one-out-by-patient test was employed for training and testing the steps of the proposed framework. The final GTV regions were determined using the OCS method that can be used to select a global optimum object contour based on multiple active delineations with a LSM around the GTV. The efficacy of the proposed framework was evaluated in 14 lung cancer cases [solid: 6, ground-glass opacity (GGO): 4, mixed GGO: 4] using the 3D Dice similarity coefficient (DSC), which denotes the degree of region similarity between the GTVs contoured by radiation oncologists and those determined using the proposed framework. The proposed framework achieved an average DSC of 0.777 for 14 cases, whereas the OCS-based framework produced an average DSC of 0.507. The average DSCs for GGO and mixed GGO were 0.763 and 0.701, respectively, obtained by the proposed framework. The proposed framework can be employed as a tool to assist radiation oncologists in delineating various GTV regions.

Original languageEnglish
Pages (from-to)123-134
Number of pages12
JournalJournal of radiation research
Volume58
Issue number1
DOIs
Publication statusPublished - Jan 1 2017

Fingerprint

machine learning
delineation
Tumor Burden
planning
tumors
opacity
Glass
glass
radiation
classifiers
education
coefficients
Positron Emission Tomography Computed Tomography
Datasets
Machine Learning
Lung Neoplasms
lungs
cancer
Radiation Oncologists

All Science Journal Classification (ASJC) codes

  • Radiation
  • Radiology Nuclear Medicine and imaging
  • Health, Toxicology and Mutagenesis

Cite this

@article{1691caf7fead4460b10910cc48dfce7e,
title = "Computer-assisted framework for machine-learning-based delineation of GTV regions on datasets of planning CT and PET/CT images",
abstract = "We have proposed a computer-assisted framework for machine-learning-based delineation of gross tumor volumes (GTVs) following an optimum contour selection (OCS) method. The key idea of the proposed framework was to feed image features around GTV contours (determined based on the knowledge of radiation oncologists) into a machine-learning classifier during the training step, after which the classifier produces the 'degree of GTV' for each voxel in the testing step. Initial GTV regions were extracted using a support vector machine (SVM) that learned the image features inside and outside each tumor region (determined by radiation oncologists). The leave-one-out-by-patient test was employed for training and testing the steps of the proposed framework. The final GTV regions were determined using the OCS method that can be used to select a global optimum object contour based on multiple active delineations with a LSM around the GTV. The efficacy of the proposed framework was evaluated in 14 lung cancer cases [solid: 6, ground-glass opacity (GGO): 4, mixed GGO: 4] using the 3D Dice similarity coefficient (DSC), which denotes the degree of region similarity between the GTVs contoured by radiation oncologists and those determined using the proposed framework. The proposed framework achieved an average DSC of 0.777 for 14 cases, whereas the OCS-based framework produced an average DSC of 0.507. The average DSCs for GGO and mixed GGO were 0.763 and 0.701, respectively, obtained by the proposed framework. The proposed framework can be employed as a tool to assist radiation oncologists in delineating various GTV regions.",
author = "Koujiro Ikushima and Hidetaka Arimura and Ze Jin and Hidetake Yabuuchi and Jumpei Kuwazuru and Yoshiyuki Shioyama and Tomonari Sasaki and Hiroshi Honda and Masayuki Sasaki",
year = "2017",
month = "1",
day = "1",
doi = "10.1093/jrr/rrw082",
language = "English",
volume = "58",
pages = "123--134",
journal = "Journal of Radiation Research",
issn = "0449-3060",
publisher = "Japan Radiation Research Society",
number = "1",

}

TY - JOUR

T1 - Computer-assisted framework for machine-learning-based delineation of GTV regions on datasets of planning CT and PET/CT images

AU - Ikushima, Koujiro

AU - Arimura, Hidetaka

AU - Jin, Ze

AU - Yabuuchi, Hidetake

AU - Kuwazuru, Jumpei

AU - Shioyama, Yoshiyuki

AU - Sasaki, Tomonari

AU - Honda, Hiroshi

AU - Sasaki, Masayuki

PY - 2017/1/1

Y1 - 2017/1/1

N2 - We have proposed a computer-assisted framework for machine-learning-based delineation of gross tumor volumes (GTVs) following an optimum contour selection (OCS) method. The key idea of the proposed framework was to feed image features around GTV contours (determined based on the knowledge of radiation oncologists) into a machine-learning classifier during the training step, after which the classifier produces the 'degree of GTV' for each voxel in the testing step. Initial GTV regions were extracted using a support vector machine (SVM) that learned the image features inside and outside each tumor region (determined by radiation oncologists). The leave-one-out-by-patient test was employed for training and testing the steps of the proposed framework. The final GTV regions were determined using the OCS method that can be used to select a global optimum object contour based on multiple active delineations with a LSM around the GTV. The efficacy of the proposed framework was evaluated in 14 lung cancer cases [solid: 6, ground-glass opacity (GGO): 4, mixed GGO: 4] using the 3D Dice similarity coefficient (DSC), which denotes the degree of region similarity between the GTVs contoured by radiation oncologists and those determined using the proposed framework. The proposed framework achieved an average DSC of 0.777 for 14 cases, whereas the OCS-based framework produced an average DSC of 0.507. The average DSCs for GGO and mixed GGO were 0.763 and 0.701, respectively, obtained by the proposed framework. The proposed framework can be employed as a tool to assist radiation oncologists in delineating various GTV regions.

AB - We have proposed a computer-assisted framework for machine-learning-based delineation of gross tumor volumes (GTVs) following an optimum contour selection (OCS) method. The key idea of the proposed framework was to feed image features around GTV contours (determined based on the knowledge of radiation oncologists) into a machine-learning classifier during the training step, after which the classifier produces the 'degree of GTV' for each voxel in the testing step. Initial GTV regions were extracted using a support vector machine (SVM) that learned the image features inside and outside each tumor region (determined by radiation oncologists). The leave-one-out-by-patient test was employed for training and testing the steps of the proposed framework. The final GTV regions were determined using the OCS method that can be used to select a global optimum object contour based on multiple active delineations with a LSM around the GTV. The efficacy of the proposed framework was evaluated in 14 lung cancer cases [solid: 6, ground-glass opacity (GGO): 4, mixed GGO: 4] using the 3D Dice similarity coefficient (DSC), which denotes the degree of region similarity between the GTVs contoured by radiation oncologists and those determined using the proposed framework. The proposed framework achieved an average DSC of 0.777 for 14 cases, whereas the OCS-based framework produced an average DSC of 0.507. The average DSCs for GGO and mixed GGO were 0.763 and 0.701, respectively, obtained by the proposed framework. The proposed framework can be employed as a tool to assist radiation oncologists in delineating various GTV regions.

UR - http://www.scopus.com/inward/record.url?scp=85014579330&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85014579330&partnerID=8YFLogxK

U2 - 10.1093/jrr/rrw082

DO - 10.1093/jrr/rrw082

M3 - Article

C2 - 27609193

AN - SCOPUS:85014579330

VL - 58

SP - 123

EP - 134

JO - Journal of Radiation Research

JF - Journal of Radiation Research

SN - 0449-3060

IS - 1

ER -