Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Deep learning-based evaluation of normal bone marrow activity in 18F-NaF PET/CT in patients with prostate cancer

Lindgren Belal, Sarah LU orcid ; Enqvist, Olof LU ; Ulén, Johannes ; Edenbrandt, Lars and Trägårdh, Elin LU (2020) European Congress of Radiology (ECR) In Insights into Imaging 11(Suppl. 1). p.349-350
Abstract
Purpose: Bone marrow is the primary site of skeletal metastases in prostate cancer. 18F-sodium fluoride (NaF) can be used to detect malignant activity, but also identifies irrelevant degenerative cortical uptake. Normal radiotracer activity in solely the marrow has yet to be described and could be a first step towards automated tumor burden calculation as SUV thresholds. We aimed to investigate normal activity of 18F-NaF in whole bone and bone marrow in patients with localized prostate cancer.

Methods and materials: 18F-NaF PET/CT scans from 87 patients with high-risk prostate cancer from two centers were retrospectively analyzed. All patients had a recent negative or inconclusive... (More)
Purpose: Bone marrow is the primary site of skeletal metastases in prostate cancer. 18F-sodium fluoride (NaF) can be used to detect malignant activity, but also identifies irrelevant degenerative cortical uptake. Normal radiotracer activity in solely the marrow has yet to be described and could be a first step towards automated tumor burden calculation as SUV thresholds. We aimed to investigate normal activity of 18F-NaF in whole bone and bone marrow in patients with localized prostate cancer.

Methods and materials: 18F-NaF PET/CT scans from 87 patients with high-risk prostate cancer from two centers were retrospectively analyzed. All patients had a recent negative or inconclusive bone scan. In the first center, PET scan was acquired 1-1.5 hours after i.v. injection of 4 MBq/kg 18F-NaF on an integrated PET/CT system (Gemini TF, Philips Medical Systems) (53/87). In the second center, scanning was performed 1 hour after i.v. injection of 3 MBq/kg 18F-NaF on an integrated PET/CT system (Discovery ST, GE Healthcare) (34/87). CT scans were obtained in immediate connection to the PET scan. Automated segmentations of vertebrae, pelvis, femora, humeri and sternum were performed in the CT scans using a deep learning-based method. Bone <7 mm from skeletal surfaces was removed to isolate the marrow. SUV was measured within the remaining area in the PET scan.

Results: SUVmax and SUVmean in the whole bone and bone marrow of the different regions were presented.

Conclusion: We present a deep-learning approach for evaluation of normal radiotracer activity in whole bone and bone marrow. Knowledge about radiotracer uptake in the normal bone prior to cancerous involvement is a necessary first step for subsequent tumor assessment and could be of value in the implementation of future tracers. (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
in
Insights into Imaging
volume
11
issue
Suppl. 1
pages
2 pages
publisher
Springer
conference name
European Congress of Radiology (ECR)
conference location
Vienna, Austria
conference dates
2020-07-15 - 2020-07-19
ISSN
1869-4101
DOI
10.1186/s13244-020-00851-0
language
English
LU publication?
yes
id
9e335575-8b64-4346-a04b-a03a6db6f79e
date added to LUP
2022-12-14 14:30:11
date last changed
2023-11-07 10:16:10
@misc{9e335575-8b64-4346-a04b-a03a6db6f79e,
  abstract     = {{<b>Purpose:</b> Bone marrow is the primary site of skeletal metastases in prostate cancer. <sup>18</sup>F-sodium fluoride (NaF) can be used to detect malignant activity, but also identifies irrelevant degenerative cortical uptake. Normal radiotracer activity in solely the marrow has yet to be described and could be a first step towards automated tumor burden calculation as SUV thresholds. We aimed to investigate normal activity of <sup>18</sup>F-NaF in whole bone and bone marrow in patients with localized prostate cancer.<br/><br/><b>Methods and materials: </b><sup>18</sup>F-NaF PET/CT scans from 87 patients with high-risk prostate cancer from two centers were retrospectively analyzed. All patients had a recent negative or inconclusive bone scan. In the first center, PET scan was acquired 1-1.5 hours after i.v. injection of 4 MBq/kg <sup>18</sup>F-NaF on an integrated PET/CT system (Gemini TF, Philips Medical Systems) (53/87). In the second center, scanning was performed 1 hour after i.v. injection of 3 MBq/kg <sup>18</sup>F-NaF on an integrated PET/CT system (Discovery ST, GE Healthcare) (34/87). CT scans were obtained in immediate connection to the PET scan. Automated segmentations of vertebrae, pelvis, femora, humeri and sternum were performed in the CT scans using a deep learning-based method. Bone &lt;7 mm from skeletal surfaces was removed to isolate the marrow. SUV was measured within the remaining area in the PET scan.<br/><br/><b>Results:</b> SUV<sub>max</sub> and SUV<sub>mean</sub> in the whole bone and bone marrow of the different regions were presented.<br/><br/><b>Conclusion:</b> We present a deep-learning approach for evaluation of normal radiotracer activity in whole bone and bone marrow. Knowledge about radiotracer uptake in the normal bone prior to cancerous involvement is a necessary first step for subsequent tumor assessment and could be of value in the implementation of future tracers.}},
  author       = {{Lindgren Belal, Sarah and Enqvist, Olof and Ulén, Johannes and Edenbrandt, Lars and Trägårdh, Elin}},
  issn         = {{1869-4101}},
  language     = {{eng}},
  note         = {{Conference Abstract}},
  number       = {{Suppl. 1}},
  pages        = {{349--350}},
  publisher    = {{Springer}},
  series       = {{Insights into Imaging}},
  title        = {{Deep learning-based evaluation of normal bone marrow activity in <sup>18</sup>F-NaF PET/CT in patients with prostate cancer}},
  url          = {{http://dx.doi.org/10.1186/s13244-020-00851-0}},
  doi          = {{10.1186/s13244-020-00851-0}},
  volume       = {{11}},
  year         = {{2020}},
}