Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Deep learning for segmentation of 49 selected bones in CT scans : First step in automated PET/CT-based 3D quantification of skeletal metastases

Lindgren Belal, Sarah LU orcid ; Sadik, May ; Kaboteh, Reza ; Enqvist, Olof ; Ulén, Johannes LU ; Poulsen, Mads H. ; Simonsen, Jane ; Høilund-Carlsen, Poul F. ; Edenbrandt, Lars LU and Trägårdh, Elin LU orcid (2019) In European Journal of Radiology 113. p.89-95
Abstract


Purpose: The aim of this study was to develop a deep learning-based method for segmentation of bones in CT scans and test its accuracy compared to manual delineation, as a first step in the creation of an automated PET/CT-based method for quantifying skeletal tumour burden. Methods: Convolutional neural networks (CNNs) were trained to segment 49 bones using manual segmentations from 100 CT scans. After training, the CNN-based segmentation method was tested on 46 patients with prostate cancer, who had undergone
18
F-choline-PET/CT and ... (More)


Purpose: The aim of this study was to develop a deep learning-based method for segmentation of bones in CT scans and test its accuracy compared to manual delineation, as a first step in the creation of an automated PET/CT-based method for quantifying skeletal tumour burden. Methods: Convolutional neural networks (CNNs) were trained to segment 49 bones using manual segmentations from 100 CT scans. After training, the CNN-based segmentation method was tested on 46 patients with prostate cancer, who had undergone
18
F-choline-PET/CT and
18
F-NaF PET/CT less than three weeks apart. Bone volumes were calculated from the segmentations. The network's performance was compared with manual segmentations of five bones made by an experienced physician. Accuracy of the spatial overlap between automated CNN-based and manual segmentations of these five bones was assessed using the Sørensen-Dice index (SDI). Reproducibility was evaluated applying the Bland-Altman method. Results: The median (SD) volumes of the five selected bones were by CNN and manual segmentation: Th7 41 (3.8) and 36 (5.1), L3 76 (13) and 75 (9.2), sacrum 284 (40) and 283 (26), 7th rib 33 (3.9) and 31 (4.8), sternum 80 (11) and 72 (9.2), respectively. Median SDIs were 0.86 (Th7), 0.85 (L3), 0.88 (sacrum), 0.84 (7th rib) and 0.83 (sternum). The intraobserver volume difference was less with CNN-based than manual approach: Th7 2% and 14%, L3 7% and 8%, sacrum 1% and 3%, 7th rib 1% and 6%, sternum 3% and 5%, respectively. The average volume difference measured as ratio volume difference/mean volume between the two CNN-based segmentations was 5–6% for the vertebral column and ribs and ≤3% for other bones. Conclusion: The new deep learning-based method for automated segmentation of bones in CT scans provided highly accurate bone volumes in a fast and automated way and, thus, appears to be a valuable first step in the development of a clinical useful processing procedure providing reliable skeletal segmentation as a key part of quantification of skeletal metastases.

(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
Artificial intelligence, Bone, Metastases, PET/CT, Prostate cancer
in
European Journal of Radiology
volume
113
pages
7 pages
publisher
Elsevier
external identifiers
  • scopus:85061371969
  • pmid:30927965
ISSN
0720-048X
DOI
10.1016/j.ejrad.2019.01.028
language
English
LU publication?
yes
id
55dd416c-ae8e-4248-90e3-8c6fd35cf04c
date added to LUP
2019-02-18 13:54:34
date last changed
2024-11-26 23:22:08
@article{55dd416c-ae8e-4248-90e3-8c6fd35cf04c,
  abstract     = {{<p><br>
                                                         Purpose: The aim of this study was to develop a deep learning-based method for segmentation of bones in CT scans and test its accuracy compared to manual delineation, as a first step in the creation of an automated PET/CT-based method for quantifying skeletal tumour burden. Methods: Convolutional neural networks (CNNs) were trained to segment 49 bones using manual segmentations from 100 CT scans. After training, the CNN-based segmentation method was tested on 46 patients with prostate cancer, who had undergone                              <br>
                            <sup>18</sup><br>
                                                         F-choline-PET/CT and                              <br>
                            <sup>18</sup><br>
                                                         F-NaF PET/CT less than three weeks apart. Bone volumes were calculated from the segmentations. The network's performance was compared with manual segmentations of five bones made by an experienced physician. Accuracy of the spatial overlap between automated CNN-based and manual segmentations of these five bones was assessed using the Sørensen-Dice index (SDI). Reproducibility was evaluated applying the Bland-Altman method. Results: The median (SD) volumes of the five selected bones were by CNN and manual segmentation: Th7 41 (3.8) and 36 (5.1), L3 76 (13) and 75 (9.2), sacrum 284 (40) and 283 (26), 7th rib 33 (3.9) and 31 (4.8), sternum 80 (11) and 72 (9.2), respectively. Median SDIs were 0.86 (Th7), 0.85 (L3), 0.88 (sacrum), 0.84 (7th rib) and 0.83 (sternum). The intraobserver volume difference was less with CNN-based than manual approach: Th7 2% and 14%, L3 7% and 8%, sacrum 1% and 3%, 7th rib 1% and 6%, sternum 3% and 5%, respectively. The average volume difference measured as ratio volume difference/mean volume between the two CNN-based segmentations was 5–6% for the vertebral column and ribs and ≤3% for other bones. Conclusion: The new deep learning-based method for automated segmentation of bones in CT scans provided highly accurate bone volumes in a fast and automated way and, thus, appears to be a valuable first step in the development of a clinical useful processing procedure providing reliable skeletal segmentation as a key part of quantification of skeletal metastases.                         <br>
                        </p>}},
  author       = {{Lindgren Belal, Sarah and Sadik, May and Kaboteh, Reza and Enqvist, Olof and Ulén, Johannes and Poulsen, Mads H. and Simonsen, Jane and Høilund-Carlsen, Poul F. and Edenbrandt, Lars and Trägårdh, Elin}},
  issn         = {{0720-048X}},
  keywords     = {{Artificial intelligence; Bone; Metastases; PET/CT; Prostate cancer}},
  language     = {{eng}},
  pages        = {{89--95}},
  publisher    = {{Elsevier}},
  series       = {{European Journal of Radiology}},
  title        = {{Deep learning for segmentation of 49 selected bones in CT scans : First step in automated PET/CT-based 3D quantification of skeletal metastases}},
  url          = {{http://dx.doi.org/10.1016/j.ejrad.2019.01.028}},
  doi          = {{10.1016/j.ejrad.2019.01.028}},
  volume       = {{113}},
  year         = {{2019}},
}