A multimodality segmentation framework for automatic target delineation in head and neck radiotherapy

Jinzhong Yang, Beth M. Beadle, Adam S. Garden, David Schwartz, Michalis Aristophanous

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Purpose: To develop an automatic segmentation algorithm integrating imaging information from computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging (MRI) to delineate target volume in head and neck cancer radiotherapy. Methods: Eleven patients with unresectable disease at the tonsil or base of tongue who underwent MRI, CT, and PET/CT within two months before the start of radiotherapy or chemoradiotherapy were recruited for the study. For each patient, PET/CT and T1-weighted contrast MRI scans were first registered to the planning CT using deformable and rigid registration, respectively, to resample the PET and magnetic resonance (MR) images to the planning CT space. A binary mask was manually defined to identify the tumor area. The resampled PET and MR images, the planning CT image, and the binary mask were fed into the automatic segmentation algorithm for target delineation. The algorithm was based on a multichannel Gaussian mixture model and solved using an expectation-maximization algorithm with Markov random fields. To evaluate the algorithm, we compared the multichannel autosegmentation with an autosegmentation method using only PET images. The physician-defined gross tumor volume (GTV) was used as the -ground truth- for quantitative evaluation. Results: The median multichannel segmented GTV of the primary tumor was 15.7 cm3 (range, 6.6-44.3 cm3), while the PET segmented GTV was 10.2 cm3 (range, 2.8-45.1 cm3). The median physician-defined GTV was 22.1 cm3 (range, 4.2-38.4 cm3). The median difference between the multichannel segmented and physician-defined GTVs was -10.7%, not showing a statistically significant difference (p-value = 0.43). However, the median difference between the PET segmented and physician-defined GTVs was -19.2%, showing a statistically significant difference (value =0.0037). The median Dice similarity coefficient between the multichannel segmented and physician-defined GTVs was 0.75 (range, 0.55-0.84), and the median sensitivity and positive predictive value between them were 0.76 and 0.81, respectively. Conclusions: The authors developed an automated multimodality segmentation algorithm for tumor volume delineation and validated this algorithm for head and neck cancer radiotherapy. The multichannel segmented GTV agreed well with the physician-defined GTV. The authors expect that their algorithm will improve the accuracy and consistency in target definition for radiotherapy.

Original languageEnglish (US)
Pages (from-to)5310-5320
Number of pages11
JournalMedical physics
Volume42
Issue number9
DOIs
StatePublished - Sep 1 2015

Fingerprint

Tumor Burden
Neck
Radiotherapy
Head
Positron-Emission Tomography
Physicians
Tomography
Magnetic Resonance Imaging
Head and Neck Neoplasms
Masks
Magnetic Resonance Spectroscopy
Palatine Tonsil
Chemoradiotherapy
Tongue
Neoplasms
Positron Emission Tomography Computed Tomography

All Science Journal Classification (ASJC) codes

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

A multimodality segmentation framework for automatic target delineation in head and neck radiotherapy. / Yang, Jinzhong; Beadle, Beth M.; Garden, Adam S.; Schwartz, David; Aristophanous, Michalis.

In: Medical physics, Vol. 42, No. 9, 01.09.2015, p. 5310-5320.

Research output: Contribution to journalArticle

Yang, Jinzhong ; Beadle, Beth M. ; Garden, Adam S. ; Schwartz, David ; Aristophanous, Michalis. / A multimodality segmentation framework for automatic target delineation in head and neck radiotherapy. In: Medical physics. 2015 ; Vol. 42, No. 9. pp. 5310-5320.
@article{5fc5a9c9b0a6403d9ef5d997ac4c57eb,
title = "A multimodality segmentation framework for automatic target delineation in head and neck radiotherapy",
abstract = "Purpose: To develop an automatic segmentation algorithm integrating imaging information from computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging (MRI) to delineate target volume in head and neck cancer radiotherapy. Methods: Eleven patients with unresectable disease at the tonsil or base of tongue who underwent MRI, CT, and PET/CT within two months before the start of radiotherapy or chemoradiotherapy were recruited for the study. For each patient, PET/CT and T1-weighted contrast MRI scans were first registered to the planning CT using deformable and rigid registration, respectively, to resample the PET and magnetic resonance (MR) images to the planning CT space. A binary mask was manually defined to identify the tumor area. The resampled PET and MR images, the planning CT image, and the binary mask were fed into the automatic segmentation algorithm for target delineation. The algorithm was based on a multichannel Gaussian mixture model and solved using an expectation-maximization algorithm with Markov random fields. To evaluate the algorithm, we compared the multichannel autosegmentation with an autosegmentation method using only PET images. The physician-defined gross tumor volume (GTV) was used as the -ground truth- for quantitative evaluation. Results: The median multichannel segmented GTV of the primary tumor was 15.7 cm3 (range, 6.6-44.3 cm3), while the PET segmented GTV was 10.2 cm3 (range, 2.8-45.1 cm3). The median physician-defined GTV was 22.1 cm3 (range, 4.2-38.4 cm3). The median difference between the multichannel segmented and physician-defined GTVs was -10.7{\%}, not showing a statistically significant difference (p-value = 0.43). However, the median difference between the PET segmented and physician-defined GTVs was -19.2{\%}, showing a statistically significant difference (value =0.0037). The median Dice similarity coefficient between the multichannel segmented and physician-defined GTVs was 0.75 (range, 0.55-0.84), and the median sensitivity and positive predictive value between them were 0.76 and 0.81, respectively. Conclusions: The authors developed an automated multimodality segmentation algorithm for tumor volume delineation and validated this algorithm for head and neck cancer radiotherapy. The multichannel segmented GTV agreed well with the physician-defined GTV. The authors expect that their algorithm will improve the accuracy and consistency in target definition for radiotherapy.",
author = "Jinzhong Yang and Beadle, {Beth M.} and Garden, {Adam S.} and David Schwartz and Michalis Aristophanous",
year = "2015",
month = "9",
day = "1",
doi = "10.1118/1.4928485",
language = "English (US)",
volume = "42",
pages = "5310--5320",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "9",

}

TY - JOUR

T1 - A multimodality segmentation framework for automatic target delineation in head and neck radiotherapy

AU - Yang, Jinzhong

AU - Beadle, Beth M.

AU - Garden, Adam S.

AU - Schwartz, David

AU - Aristophanous, Michalis

PY - 2015/9/1

Y1 - 2015/9/1

N2 - Purpose: To develop an automatic segmentation algorithm integrating imaging information from computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging (MRI) to delineate target volume in head and neck cancer radiotherapy. Methods: Eleven patients with unresectable disease at the tonsil or base of tongue who underwent MRI, CT, and PET/CT within two months before the start of radiotherapy or chemoradiotherapy were recruited for the study. For each patient, PET/CT and T1-weighted contrast MRI scans were first registered to the planning CT using deformable and rigid registration, respectively, to resample the PET and magnetic resonance (MR) images to the planning CT space. A binary mask was manually defined to identify the tumor area. The resampled PET and MR images, the planning CT image, and the binary mask were fed into the automatic segmentation algorithm for target delineation. The algorithm was based on a multichannel Gaussian mixture model and solved using an expectation-maximization algorithm with Markov random fields. To evaluate the algorithm, we compared the multichannel autosegmentation with an autosegmentation method using only PET images. The physician-defined gross tumor volume (GTV) was used as the -ground truth- for quantitative evaluation. Results: The median multichannel segmented GTV of the primary tumor was 15.7 cm3 (range, 6.6-44.3 cm3), while the PET segmented GTV was 10.2 cm3 (range, 2.8-45.1 cm3). The median physician-defined GTV was 22.1 cm3 (range, 4.2-38.4 cm3). The median difference between the multichannel segmented and physician-defined GTVs was -10.7%, not showing a statistically significant difference (p-value = 0.43). However, the median difference between the PET segmented and physician-defined GTVs was -19.2%, showing a statistically significant difference (value =0.0037). The median Dice similarity coefficient between the multichannel segmented and physician-defined GTVs was 0.75 (range, 0.55-0.84), and the median sensitivity and positive predictive value between them were 0.76 and 0.81, respectively. Conclusions: The authors developed an automated multimodality segmentation algorithm for tumor volume delineation and validated this algorithm for head and neck cancer radiotherapy. The multichannel segmented GTV agreed well with the physician-defined GTV. The authors expect that their algorithm will improve the accuracy and consistency in target definition for radiotherapy.

AB - Purpose: To develop an automatic segmentation algorithm integrating imaging information from computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging (MRI) to delineate target volume in head and neck cancer radiotherapy. Methods: Eleven patients with unresectable disease at the tonsil or base of tongue who underwent MRI, CT, and PET/CT within two months before the start of radiotherapy or chemoradiotherapy were recruited for the study. For each patient, PET/CT and T1-weighted contrast MRI scans were first registered to the planning CT using deformable and rigid registration, respectively, to resample the PET and magnetic resonance (MR) images to the planning CT space. A binary mask was manually defined to identify the tumor area. The resampled PET and MR images, the planning CT image, and the binary mask were fed into the automatic segmentation algorithm for target delineation. The algorithm was based on a multichannel Gaussian mixture model and solved using an expectation-maximization algorithm with Markov random fields. To evaluate the algorithm, we compared the multichannel autosegmentation with an autosegmentation method using only PET images. The physician-defined gross tumor volume (GTV) was used as the -ground truth- for quantitative evaluation. Results: The median multichannel segmented GTV of the primary tumor was 15.7 cm3 (range, 6.6-44.3 cm3), while the PET segmented GTV was 10.2 cm3 (range, 2.8-45.1 cm3). The median physician-defined GTV was 22.1 cm3 (range, 4.2-38.4 cm3). The median difference between the multichannel segmented and physician-defined GTVs was -10.7%, not showing a statistically significant difference (p-value = 0.43). However, the median difference between the PET segmented and physician-defined GTVs was -19.2%, showing a statistically significant difference (value =0.0037). The median Dice similarity coefficient between the multichannel segmented and physician-defined GTVs was 0.75 (range, 0.55-0.84), and the median sensitivity and positive predictive value between them were 0.76 and 0.81, respectively. Conclusions: The authors developed an automated multimodality segmentation algorithm for tumor volume delineation and validated this algorithm for head and neck cancer radiotherapy. The multichannel segmented GTV agreed well with the physician-defined GTV. The authors expect that their algorithm will improve the accuracy and consistency in target definition for radiotherapy.

UR - http://www.scopus.com/inward/record.url?scp=84939839777&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84939839777&partnerID=8YFLogxK

U2 - 10.1118/1.4928485

DO - 10.1118/1.4928485

M3 - Article

VL - 42

SP - 5310

EP - 5320

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 9

ER -