Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Synthetic computed tomography data allows for accurate absorbed dose calculations in a magnetic resonance imaging only workflow for head and neck radiotherapy

Palmér, Emilia ; Karlsson, Anna LU ; Nordström, Fredrik LU ; Petruson, Karin ; Siversson, Carl LU ; Ljungberg, Maria and Sohlin, Maja (2021) In Physics and imaging in radiation oncology 17. p.36-42
Abstract

Background and purpose: Few studies on magnetic resonance imaging (MRI) only head and neck radiation treatment planning exist, and none using a generally available software. The aim of this study was to evaluate the accuracy of absorbed dose for head and neck synthetic computed tomography data (sCT) generated by a commercial convolutional neural network-based algorithm. Materials and methods: For 44 head and neck cancer patients, sCT were generated and the geometry was validated against computed tomography data (CT). The clinical CT based treatment plan was transferred to the sCT and recalculated without re-optimization, and differences in relative absorbed dose were determined for dose-volume-histogram (DVH) parameters and the 3D... (More)

Background and purpose: Few studies on magnetic resonance imaging (MRI) only head and neck radiation treatment planning exist, and none using a generally available software. The aim of this study was to evaluate the accuracy of absorbed dose for head and neck synthetic computed tomography data (sCT) generated by a commercial convolutional neural network-based algorithm. Materials and methods: For 44 head and neck cancer patients, sCT were generated and the geometry was validated against computed tomography data (CT). The clinical CT based treatment plan was transferred to the sCT and recalculated without re-optimization, and differences in relative absorbed dose were determined for dose-volume-histogram (DVH) parameters and the 3D volume. Results: For overall body, the results of the geometric validation were (Mean ± 1sd): Mean error −5 ± 10 HU, mean absolute error 67 ± 14 HU, Dice similarity coefficient 0.98 ± 0.05, and Hausdorff distance difference 4.2 ± 1.7 mm. Water equivalent depth difference for region Th1-C7, mid mandible and mid nose were −0.3 ± 3.4, 1.1 ± 2.0 and 0.7 ± 3.8 mm respectively. The maximum mean deviation in absorbed dose for all DVH parameters was 0.30% (0.12 Gy). The absorbed doses were considered equivalent (p-value < 0.001) and the mean 3D gamma passing rate was 99.4 (range: 95.7–99.9%). Conclusions: The convolutional neural network-based algorithm generates sCT which allows for accurate absorbed dose calculations for MRI-only head and neck radiation treatment planning. The sCT allows for statistically equivalent absorbed dose calculations compared to CT based radiotherapy.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Convolutional neural network, Head and neck, MRI only, Pseudo CT, Synthetic CT, Treatment planning
in
Physics and imaging in radiation oncology
volume
17
pages
7 pages
publisher
Elsevier
external identifiers
  • pmid:33898776
  • scopus:85099193644
ISSN
2405-6316
DOI
10.1016/j.phro.2020.12.007
language
English
LU publication?
yes
id
6e8c912c-a04f-47c5-a347-629c675a13ae
date added to LUP
2021-01-25 12:51:39
date last changed
2024-04-18 01:51:37
@article{6e8c912c-a04f-47c5-a347-629c675a13ae,
  abstract     = {{<p>Background and purpose: Few studies on magnetic resonance imaging (MRI) only head and neck radiation treatment planning exist, and none using a generally available software. The aim of this study was to evaluate the accuracy of absorbed dose for head and neck synthetic computed tomography data (sCT) generated by a commercial convolutional neural network-based algorithm. Materials and methods: For 44 head and neck cancer patients, sCT were generated and the geometry was validated against computed tomography data (CT). The clinical CT based treatment plan was transferred to the sCT and recalculated without re-optimization, and differences in relative absorbed dose were determined for dose-volume-histogram (DVH) parameters and the 3D volume. Results: For overall body, the results of the geometric validation were (Mean ± 1sd): Mean error −5 ± 10 HU, mean absolute error 67 ± 14 HU, Dice similarity coefficient 0.98 ± 0.05, and Hausdorff distance difference 4.2 ± 1.7 mm. Water equivalent depth difference for region Th1-C7, mid mandible and mid nose were −0.3 ± 3.4, 1.1 ± 2.0 and 0.7 ± 3.8 mm respectively. The maximum mean deviation in absorbed dose for all DVH parameters was 0.30% (0.12 Gy). The absorbed doses were considered equivalent (p-value &lt; 0.001) and the mean 3D gamma passing rate was 99.4 (range: 95.7–99.9%). Conclusions: The convolutional neural network-based algorithm generates sCT which allows for accurate absorbed dose calculations for MRI-only head and neck radiation treatment planning. The sCT allows for statistically equivalent absorbed dose calculations compared to CT based radiotherapy.</p>}},
  author       = {{Palmér, Emilia and Karlsson, Anna and Nordström, Fredrik and Petruson, Karin and Siversson, Carl and Ljungberg, Maria and Sohlin, Maja}},
  issn         = {{2405-6316}},
  keywords     = {{Convolutional neural network; Head and neck; MRI only; Pseudo CT; Synthetic CT; Treatment planning}},
  language     = {{eng}},
  pages        = {{36--42}},
  publisher    = {{Elsevier}},
  series       = {{Physics and imaging in radiation oncology}},
  title        = {{Synthetic computed tomography data allows for accurate absorbed dose calculations in a magnetic resonance imaging only workflow for head and neck radiotherapy}},
  url          = {{http://dx.doi.org/10.1016/j.phro.2020.12.007}},
  doi          = {{10.1016/j.phro.2020.12.007}},
  volume       = {{17}},
  year         = {{2021}},
}