Deep-learning–assisted automatic digitization of applicators in 3D CT image-based high-dose-rate brachytherapy of gynecological cancer

Hyunuk Jung, Yesenia Gonzalez, Chenyang Shen, Peter Klages, Kevin Albuquerque, Xun Jia

Research output: Contribution to journalArticle

Abstract

Purpose: Applicator digitization is one of the most critical steps in 3D high-dose-rate brachytherapy (HDRBT) treatment planning. Motivated by recent advances in deep-learning, we propose a deep-learning–assisted applicator digitization method for 3D CT image-based HDRBT. This study demonstrates its feasibility and potential in gynecological cancer HDRBT. Methods and Materials: Our method consisted of two steps. The first step used a U-net to segment applicator regions. We trained the U-net using two-dimensional CT images with a tandem-and-ovoid (T&O) applicator and corresponding applicator mask images. The second step applied a spectral clustering method and a polynomial curve fitting method to extract applicator central paths. We evaluated the accuracy, efficiency, and robustness of our method in different scenarios including other T&O cases that were not used in training, a T&O case scanned with cone-beam CT, and Y-tandem and cylinder-applicator cases. Results: In test cases with a T&O applicator, average 3D Dice similarity coefficient between automatic and manual segmented applicator regions was 0.93. Average distance between tip positions and average Hausdorff distance between applicator channels determined by our method and manually were 0.64 mm and 0.68 mm, respectively. Although trained only using CT images of T&O cases, our tool can also digitize Y-tandem, cylinder applicator, and T&O applicator scanned in cone-beam CT with error of tip position and Hausdorff distance <1 mm. Computation time was ∼15 s per case. Conclusions: We have developed a deep-learning–assisted applicator digitization tool for 3D CT image–based HDRBT of gynecological cancer. The achieved accuracy, efficiency, and robustness made our tool clinically attractive.

Original languageEnglish (US)
JournalBrachytherapy
DOIs
StatePublished - Jan 1 2019

Fingerprint

Brachytherapy
Neoplasms
Cone-Beam Computed Tomography
Masks
Cluster Analysis
Learning
Efficiency

Keywords

  • Applicator digitization
  • Deep-learning
  • High-dose-rate

ASJC Scopus subject areas

  • Oncology
  • Radiology Nuclear Medicine and imaging

Cite this

Deep-learning–assisted automatic digitization of applicators in 3D CT image-based high-dose-rate brachytherapy of gynecological cancer. / Jung, Hyunuk; Gonzalez, Yesenia; Shen, Chenyang; Klages, Peter; Albuquerque, Kevin; Jia, Xun.

In: Brachytherapy, 01.01.2019.

Research output: Contribution to journalArticle

@article{5998a13e87a84a71bc82291135b399e5,
title = "Deep-learning–assisted automatic digitization of applicators in 3D CT image-based high-dose-rate brachytherapy of gynecological cancer",
abstract = "Purpose: Applicator digitization is one of the most critical steps in 3D high-dose-rate brachytherapy (HDRBT) treatment planning. Motivated by recent advances in deep-learning, we propose a deep-learning–assisted applicator digitization method for 3D CT image-based HDRBT. This study demonstrates its feasibility and potential in gynecological cancer HDRBT. Methods and Materials: Our method consisted of two steps. The first step used a U-net to segment applicator regions. We trained the U-net using two-dimensional CT images with a tandem-and-ovoid (T&O) applicator and corresponding applicator mask images. The second step applied a spectral clustering method and a polynomial curve fitting method to extract applicator central paths. We evaluated the accuracy, efficiency, and robustness of our method in different scenarios including other T&O cases that were not used in training, a T&O case scanned with cone-beam CT, and Y-tandem and cylinder-applicator cases. Results: In test cases with a T&O applicator, average 3D Dice similarity coefficient between automatic and manual segmented applicator regions was 0.93. Average distance between tip positions and average Hausdorff distance between applicator channels determined by our method and manually were 0.64 mm and 0.68 mm, respectively. Although trained only using CT images of T&O cases, our tool can also digitize Y-tandem, cylinder applicator, and T&O applicator scanned in cone-beam CT with error of tip position and Hausdorff distance <1 mm. Computation time was ∼15 s per case. Conclusions: We have developed a deep-learning–assisted applicator digitization tool for 3D CT image–based HDRBT of gynecological cancer. The achieved accuracy, efficiency, and robustness made our tool clinically attractive.",
keywords = "Applicator digitization, Deep-learning, High-dose-rate",
author = "Hyunuk Jung and Yesenia Gonzalez and Chenyang Shen and Peter Klages and Kevin Albuquerque and Xun Jia",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.brachy.2019.06.003",
language = "English (US)",
journal = "Brachytherapy",
issn = "1538-4721",
publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - Deep-learning–assisted automatic digitization of applicators in 3D CT image-based high-dose-rate brachytherapy of gynecological cancer

AU - Jung, Hyunuk

AU - Gonzalez, Yesenia

AU - Shen, Chenyang

AU - Klages, Peter

AU - Albuquerque, Kevin

AU - Jia, Xun

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Purpose: Applicator digitization is one of the most critical steps in 3D high-dose-rate brachytherapy (HDRBT) treatment planning. Motivated by recent advances in deep-learning, we propose a deep-learning–assisted applicator digitization method for 3D CT image-based HDRBT. This study demonstrates its feasibility and potential in gynecological cancer HDRBT. Methods and Materials: Our method consisted of two steps. The first step used a U-net to segment applicator regions. We trained the U-net using two-dimensional CT images with a tandem-and-ovoid (T&O) applicator and corresponding applicator mask images. The second step applied a spectral clustering method and a polynomial curve fitting method to extract applicator central paths. We evaluated the accuracy, efficiency, and robustness of our method in different scenarios including other T&O cases that were not used in training, a T&O case scanned with cone-beam CT, and Y-tandem and cylinder-applicator cases. Results: In test cases with a T&O applicator, average 3D Dice similarity coefficient between automatic and manual segmented applicator regions was 0.93. Average distance between tip positions and average Hausdorff distance between applicator channels determined by our method and manually were 0.64 mm and 0.68 mm, respectively. Although trained only using CT images of T&O cases, our tool can also digitize Y-tandem, cylinder applicator, and T&O applicator scanned in cone-beam CT with error of tip position and Hausdorff distance <1 mm. Computation time was ∼15 s per case. Conclusions: We have developed a deep-learning–assisted applicator digitization tool for 3D CT image–based HDRBT of gynecological cancer. The achieved accuracy, efficiency, and robustness made our tool clinically attractive.

AB - Purpose: Applicator digitization is one of the most critical steps in 3D high-dose-rate brachytherapy (HDRBT) treatment planning. Motivated by recent advances in deep-learning, we propose a deep-learning–assisted applicator digitization method for 3D CT image-based HDRBT. This study demonstrates its feasibility and potential in gynecological cancer HDRBT. Methods and Materials: Our method consisted of two steps. The first step used a U-net to segment applicator regions. We trained the U-net using two-dimensional CT images with a tandem-and-ovoid (T&O) applicator and corresponding applicator mask images. The second step applied a spectral clustering method and a polynomial curve fitting method to extract applicator central paths. We evaluated the accuracy, efficiency, and robustness of our method in different scenarios including other T&O cases that were not used in training, a T&O case scanned with cone-beam CT, and Y-tandem and cylinder-applicator cases. Results: In test cases with a T&O applicator, average 3D Dice similarity coefficient between automatic and manual segmented applicator regions was 0.93. Average distance between tip positions and average Hausdorff distance between applicator channels determined by our method and manually were 0.64 mm and 0.68 mm, respectively. Although trained only using CT images of T&O cases, our tool can also digitize Y-tandem, cylinder applicator, and T&O applicator scanned in cone-beam CT with error of tip position and Hausdorff distance <1 mm. Computation time was ∼15 s per case. Conclusions: We have developed a deep-learning–assisted applicator digitization tool for 3D CT image–based HDRBT of gynecological cancer. The achieved accuracy, efficiency, and robustness made our tool clinically attractive.

KW - Applicator digitization

KW - Deep-learning

KW - High-dose-rate

UR - http://www.scopus.com/inward/record.url?scp=85069711440&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069711440&partnerID=8YFLogxK

U2 - 10.1016/j.brachy.2019.06.003

DO - 10.1016/j.brachy.2019.06.003

M3 - Article

JO - Brachytherapy

JF - Brachytherapy

SN - 1538-4721

ER -