Automated method for extraction of lung tumors using a machine learning classifier with knowledge of radiation oncologists on data sets of planning CT and FDG-PET/CT images

Hidetaka Arimura, Ze Jin, Yoshiyuki Shioyama, Katsumasa Nakamura, Taiki Magome, Masayuki Sasaki

Research output: Contribution to journalArticle

Abstract

We have developed an automated method for extraction of lung tumors using a machine learning classifier with knowledge of radiation oncologists on data sets of treatment planning computed tomography (CT) and 18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET)/CT images. First, the PET images were registered with the treatment planning CT images through the diagnostic CT images of PET/CT. Second, six voxel-based features including voxel values and magnitudes of image gradient vectors were derived from each voxel in the planning CT and PET /CT image data sets. Finally, lung tumors were extracted by using a support vector machine (SVM), which learned 6 voxel-based features inside and outside each true tumor region determined by radiation oncologists. The results showed that the average DSCs for 3 and 6 features for three cases were 0.744 and 0.899, and thus the SVM may need 6 features to learn the distinguishable characteristics. The proposed method may be useful for assisting treatment planners in delineation of the tumor region.

Fingerprint

Positron emission tomography
Tomography
Learning systems
Tumors
Classifiers
Radiation
Planning
Lung
Neoplasms
Fluorodeoxyglucose F18
Support vector machines
Positron-Emission Tomography
Therapeutics
Radiation Oncologists
Positron Emission Tomography Computed Tomography
Datasets
Machine Learning
Support Vector Machine

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Biomedical Engineering
  • Computer Vision and Pattern Recognition
  • Health Informatics

Cite this

@article{9522376a041e4217b9061cc4c3e8a586,
title = "Automated method for extraction of lung tumors using a machine learning classifier with knowledge of radiation oncologists on data sets of planning CT and FDG-PET/CT images",
abstract = "We have developed an automated method for extraction of lung tumors using a machine learning classifier with knowledge of radiation oncologists on data sets of treatment planning computed tomography (CT) and 18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET)/CT images. First, the PET images were registered with the treatment planning CT images through the diagnostic CT images of PET/CT. Second, six voxel-based features including voxel values and magnitudes of image gradient vectors were derived from each voxel in the planning CT and PET /CT image data sets. Finally, lung tumors were extracted by using a support vector machine (SVM), which learned 6 voxel-based features inside and outside each true tumor region determined by radiation oncologists. The results showed that the average DSCs for 3 and 6 features for three cases were 0.744 and 0.899, and thus the SVM may need 6 features to learn the distinguishable characteristics. The proposed method may be useful for assisting treatment planners in delineation of the tumor region.",
author = "Hidetaka Arimura and Ze Jin and Yoshiyuki Shioyama and Katsumasa Nakamura and Taiki Magome and Masayuki Sasaki",
year = "2013",
month = "1",
day = "1",
doi = "10.1109/EMBC.2013.6610168",
language = "English",
volume = "2013",
pages = "2988--2991",
journal = "Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings",
issn = "1557-170X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Automated method for extraction of lung tumors using a machine learning classifier with knowledge of radiation oncologists on data sets of planning CT and FDG-PET/CT images

AU - Arimura, Hidetaka

AU - Jin, Ze

AU - Shioyama, Yoshiyuki

AU - Nakamura, Katsumasa

AU - Magome, Taiki

AU - Sasaki, Masayuki

PY - 2013/1/1

Y1 - 2013/1/1

N2 - We have developed an automated method for extraction of lung tumors using a machine learning classifier with knowledge of radiation oncologists on data sets of treatment planning computed tomography (CT) and 18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET)/CT images. First, the PET images were registered with the treatment planning CT images through the diagnostic CT images of PET/CT. Second, six voxel-based features including voxel values and magnitudes of image gradient vectors were derived from each voxel in the planning CT and PET /CT image data sets. Finally, lung tumors were extracted by using a support vector machine (SVM), which learned 6 voxel-based features inside and outside each true tumor region determined by radiation oncologists. The results showed that the average DSCs for 3 and 6 features for three cases were 0.744 and 0.899, and thus the SVM may need 6 features to learn the distinguishable characteristics. The proposed method may be useful for assisting treatment planners in delineation of the tumor region.

AB - We have developed an automated method for extraction of lung tumors using a machine learning classifier with knowledge of radiation oncologists on data sets of treatment planning computed tomography (CT) and 18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET)/CT images. First, the PET images were registered with the treatment planning CT images through the diagnostic CT images of PET/CT. Second, six voxel-based features including voxel values and magnitudes of image gradient vectors were derived from each voxel in the planning CT and PET /CT image data sets. Finally, lung tumors were extracted by using a support vector machine (SVM), which learned 6 voxel-based features inside and outside each true tumor region determined by radiation oncologists. The results showed that the average DSCs for 3 and 6 features for three cases were 0.744 and 0.899, and thus the SVM may need 6 features to learn the distinguishable characteristics. The proposed method may be useful for assisting treatment planners in delineation of the tumor region.

UR - http://www.scopus.com/inward/record.url?scp=84931069421&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84931069421&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2013.6610168

DO - 10.1109/EMBC.2013.6610168

M3 - Article

C2 - 24110355

VL - 2013

SP - 2988

EP - 2991

JO - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

JF - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

SN - 1557-170X

ER -