Feasibility of anatomical feature points for the estimation of prostate locations in the Bayesian delineation frameworks for prostate cancer radiotherapy

Kenta Ninomiya, Hidetaka Arimura, Motoki Sasahara, Yudai Kai, Taka aki Hirose, Ohga Saiji

Research output: Contribution to journalArticle

Abstract

This study aimed to investigate the feasibility of anatomical feature points for the estimation of prostate locations in the Bayesian delineation frameworks for prostate cancer radiotherapy. The relationships between the reference centroids of prostate regions (CPRs) (prostate locations) and anatomical feature points were explored, and the most feasible anatomical feature points were selected based on the smallest location estimation errors of CPRs and the largest Dice’s similarity coefficient (DSC) between the reference and extracted prostates. The reference CPRs were calculated according to reference prostate contours determined by radiation oncologists. Five anatomical feature points were manually determined on a prostate, bladder, and rectum in a sagittal plane of a planning computed tomography image for each case. The CPRs were estimated using three machine learning architectures [artificial neural network, random forest, and support vector machine (SVM)], which learned the relationships between the reference CPRs and anatomical feature points. The CPRs were applied for placing a prostate probabilistic atlas at the coordinates and extracting prostate regions using a Bayesian delineation framework. The average estimation errors without and with SVM using three feature points, which indicated the best performance, were 5.6 ± 3.7 mm and 1.8 ± 1.0 mm, respectively (the smallest error) (p < 0.001). The average DSCs without and with SVM using the three feature points were 0.69 ± 0.13 and 0.82 ± 0.055, respectively (the highest DSC) (p < 0.001). The anatomical feature points may be feasible for the estimation of prostate locations, which can be applied to the general Bayesian delineation frameworks for prostate cancer radiotherapy.

Original languageEnglish
Pages (from-to)434-444
Number of pages11
JournalRadiological physics and technology
Volume11
Issue number4
DOIs
Publication statusPublished - Dec 1 2018

Fingerprint

delineation
centroids
Prostate
radiation therapy
Prostatic Neoplasms
Radiotherapy
cancer
rectum
machine learning
bladder
coefficients
planning
tomography
radiation
Atlases
Rectum

All Science Journal Classification (ASJC) codes

  • Radiation
  • Physical Therapy, Sports Therapy and Rehabilitation
  • Radiology Nuclear Medicine and imaging

Cite this

Feasibility of anatomical feature points for the estimation of prostate locations in the Bayesian delineation frameworks for prostate cancer radiotherapy. / Ninomiya, Kenta; Arimura, Hidetaka; Sasahara, Motoki; Kai, Yudai; Hirose, Taka aki; Saiji, Ohga.

In: Radiological physics and technology, Vol. 11, No. 4, 01.12.2018, p. 434-444.

Research output: Contribution to journalArticle

@article{f287a2a864ef48138f9069fc8548bed0,
title = "Feasibility of anatomical feature points for the estimation of prostate locations in the Bayesian delineation frameworks for prostate cancer radiotherapy",
abstract = "This study aimed to investigate the feasibility of anatomical feature points for the estimation of prostate locations in the Bayesian delineation frameworks for prostate cancer radiotherapy. The relationships between the reference centroids of prostate regions (CPRs) (prostate locations) and anatomical feature points were explored, and the most feasible anatomical feature points were selected based on the smallest location estimation errors of CPRs and the largest Dice’s similarity coefficient (DSC) between the reference and extracted prostates. The reference CPRs were calculated according to reference prostate contours determined by radiation oncologists. Five anatomical feature points were manually determined on a prostate, bladder, and rectum in a sagittal plane of a planning computed tomography image for each case. The CPRs were estimated using three machine learning architectures [artificial neural network, random forest, and support vector machine (SVM)], which learned the relationships between the reference CPRs and anatomical feature points. The CPRs were applied for placing a prostate probabilistic atlas at the coordinates and extracting prostate regions using a Bayesian delineation framework. The average estimation errors without and with SVM using three feature points, which indicated the best performance, were 5.6 ± 3.7 mm and 1.8 ± 1.0 mm, respectively (the smallest error) (p < 0.001). The average DSCs without and with SVM using the three feature points were 0.69 ± 0.13 and 0.82 ± 0.055, respectively (the highest DSC) (p < 0.001). The anatomical feature points may be feasible for the estimation of prostate locations, which can be applied to the general Bayesian delineation frameworks for prostate cancer radiotherapy.",
author = "Kenta Ninomiya and Hidetaka Arimura and Motoki Sasahara and Yudai Kai and Hirose, {Taka aki} and Ohga Saiji",
year = "2018",
month = "12",
day = "1",
doi = "10.1007/s12194-018-0481-2",
language = "English",
volume = "11",
pages = "434--444",
journal = "Radiological Physics and Technology",
issn = "1865-0333",
publisher = "Springer Japan",
number = "4",

}

TY - JOUR

T1 - Feasibility of anatomical feature points for the estimation of prostate locations in the Bayesian delineation frameworks for prostate cancer radiotherapy

AU - Ninomiya, Kenta

AU - Arimura, Hidetaka

AU - Sasahara, Motoki

AU - Kai, Yudai

AU - Hirose, Taka aki

AU - Saiji, Ohga

PY - 2018/12/1

Y1 - 2018/12/1

N2 - This study aimed to investigate the feasibility of anatomical feature points for the estimation of prostate locations in the Bayesian delineation frameworks for prostate cancer radiotherapy. The relationships between the reference centroids of prostate regions (CPRs) (prostate locations) and anatomical feature points were explored, and the most feasible anatomical feature points were selected based on the smallest location estimation errors of CPRs and the largest Dice’s similarity coefficient (DSC) between the reference and extracted prostates. The reference CPRs were calculated according to reference prostate contours determined by radiation oncologists. Five anatomical feature points were manually determined on a prostate, bladder, and rectum in a sagittal plane of a planning computed tomography image for each case. The CPRs were estimated using three machine learning architectures [artificial neural network, random forest, and support vector machine (SVM)], which learned the relationships between the reference CPRs and anatomical feature points. The CPRs were applied for placing a prostate probabilistic atlas at the coordinates and extracting prostate regions using a Bayesian delineation framework. The average estimation errors without and with SVM using three feature points, which indicated the best performance, were 5.6 ± 3.7 mm and 1.8 ± 1.0 mm, respectively (the smallest error) (p < 0.001). The average DSCs without and with SVM using the three feature points were 0.69 ± 0.13 and 0.82 ± 0.055, respectively (the highest DSC) (p < 0.001). The anatomical feature points may be feasible for the estimation of prostate locations, which can be applied to the general Bayesian delineation frameworks for prostate cancer radiotherapy.

AB - This study aimed to investigate the feasibility of anatomical feature points for the estimation of prostate locations in the Bayesian delineation frameworks for prostate cancer radiotherapy. The relationships between the reference centroids of prostate regions (CPRs) (prostate locations) and anatomical feature points were explored, and the most feasible anatomical feature points were selected based on the smallest location estimation errors of CPRs and the largest Dice’s similarity coefficient (DSC) between the reference and extracted prostates. The reference CPRs were calculated according to reference prostate contours determined by radiation oncologists. Five anatomical feature points were manually determined on a prostate, bladder, and rectum in a sagittal plane of a planning computed tomography image for each case. The CPRs were estimated using three machine learning architectures [artificial neural network, random forest, and support vector machine (SVM)], which learned the relationships between the reference CPRs and anatomical feature points. The CPRs were applied for placing a prostate probabilistic atlas at the coordinates and extracting prostate regions using a Bayesian delineation framework. The average estimation errors without and with SVM using three feature points, which indicated the best performance, were 5.6 ± 3.7 mm and 1.8 ± 1.0 mm, respectively (the smallest error) (p < 0.001). The average DSCs without and with SVM using the three feature points were 0.69 ± 0.13 and 0.82 ± 0.055, respectively (the highest DSC) (p < 0.001). The anatomical feature points may be feasible for the estimation of prostate locations, which can be applied to the general Bayesian delineation frameworks for prostate cancer radiotherapy.

UR - http://www.scopus.com/inward/record.url?scp=85054296488&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054296488&partnerID=8YFLogxK

U2 - 10.1007/s12194-018-0481-2

DO - 10.1007/s12194-018-0481-2

M3 - Article

C2 - 30267211

AN - SCOPUS:85054296488

VL - 11

SP - 434

EP - 444

JO - Radiological Physics and Technology

JF - Radiological Physics and Technology

SN - 1865-0333

IS - 4

ER -