Automated quantification of PET/CT skeletal tumor burden in prostate cancer using artificial intelligence : The PET index
(2023) In European Journal of Nuclear Medicine and Molecular Imaging 50(5). p.1510-1520- Abstract
Purpose: Consistent assessment of bone metastases is crucial for patient management and clinical trials in prostate cancer (PCa). We aimed to develop a fully automated convolutional neural network (CNN)-based model for calculating PET/CT skeletal tumor burden in patients with PCa. Methods: A total of 168 patients from three centers were divided into training, validation, and test groups. Manual annotations of skeletal lesions in [18F]fluoride PET/CT scans were used to train a CNN. The AI model was evaluated in 26 patients and compared to segmentations by physicians and to a SUV 15 threshold. PET index representing the percentage of skeletal volume taken up by lesions was estimated. Results: There was no case in which all... (More)
Purpose: Consistent assessment of bone metastases is crucial for patient management and clinical trials in prostate cancer (PCa). We aimed to develop a fully automated convolutional neural network (CNN)-based model for calculating PET/CT skeletal tumor burden in patients with PCa. Methods: A total of 168 patients from three centers were divided into training, validation, and test groups. Manual annotations of skeletal lesions in [18F]fluoride PET/CT scans were used to train a CNN. The AI model was evaluated in 26 patients and compared to segmentations by physicians and to a SUV 15 threshold. PET index representing the percentage of skeletal volume taken up by lesions was estimated. Results: There was no case in which all readers agreed on prevalence of lesions that the AI model failed to detect. PET index by the AI model correlated moderately strong to physician PET index (mean r = 0.69). Threshold PET index correlated fairly with physician PET index (mean r = 0.49). The sensitivity for lesion detection was 65–76% for AI, 68–91% for physicians, and 44–51% for threshold depending on which physician was considered reference. Conclusion: It was possible to develop an AI-based model for automated assessment of PET/CT skeletal tumor burden. The model’s performance was superior to using a threshold and provides fully automated calculation of whole-body skeletal tumor burden. It could be further developed to apply to different radiotracers. Objective scan evaluation is a first step toward developing a PET/CT imaging biomarker for PCa skeletal metastases.
(Less)
- author
- Lindgren Belal, Sarah LU ; Larsson, Måns ; Holm, Jorun ; Buch-Olsen, Karen Middelbo ; Sörensen, Jens ; Bjartell, Anders LU ; Edenbrandt, Lars and Trägårdh, Elin LU
- organization
-
- WCMM-Wallenberg Centre for Molecular Medicine
- Nuclear medicine, Malmö (research group)
- LUCC: Lund University Cancer Centre
- Urological cancer, Malmö (research group)
- eSSENCE: The e-Science Collaboration
- EpiHealth: Epidemiology for Health
- Clinical Physiology and Nuclear Medicine, Malmö (research group)
- publishing date
- 2023
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Artificial intelligence, Deep learning, PET-CT, Prostate cancer, Tumor burden
- in
- European Journal of Nuclear Medicine and Molecular Imaging
- volume
- 50
- issue
- 5
- pages
- 1510 - 1520
- publisher
- Springer
- external identifiers
-
- scopus:85146393066
- pmid:36650356
- ISSN
- 1619-7070
- DOI
- 10.1007/s00259-023-06108-4
- language
- English
- LU publication?
- yes
- id
- db3e7ea8-c8d1-4fe3-8c7a-aba718d28322
- date added to LUP
- 2023-02-20 09:05:40
- date last changed
- 2024-09-20 23:01:50
@article{db3e7ea8-c8d1-4fe3-8c7a-aba718d28322, abstract = {{<p>Purpose: Consistent assessment of bone metastases is crucial for patient management and clinical trials in prostate cancer (PCa). We aimed to develop a fully automated convolutional neural network (CNN)-based model for calculating PET/CT skeletal tumor burden in patients with PCa. Methods: A total of 168 patients from three centers were divided into training, validation, and test groups. Manual annotations of skeletal lesions in [<sup>18</sup>F]fluoride PET/CT scans were used to train a CNN. The AI model was evaluated in 26 patients and compared to segmentations by physicians and to a SUV 15 threshold. PET index representing the percentage of skeletal volume taken up by lesions was estimated. Results: There was no case in which all readers agreed on prevalence of lesions that the AI model failed to detect. PET index by the AI model correlated moderately strong to physician PET index (mean r = 0.69). Threshold PET index correlated fairly with physician PET index (mean r = 0.49). The sensitivity for lesion detection was 65–76% for AI, 68–91% for physicians, and 44–51% for threshold depending on which physician was considered reference. Conclusion: It was possible to develop an AI-based model for automated assessment of PET/CT skeletal tumor burden. The model’s performance was superior to using a threshold and provides fully automated calculation of whole-body skeletal tumor burden. It could be further developed to apply to different radiotracers. Objective scan evaluation is a first step toward developing a PET/CT imaging biomarker for PCa skeletal metastases.</p>}}, author = {{Lindgren Belal, Sarah and Larsson, Måns and Holm, Jorun and Buch-Olsen, Karen Middelbo and Sörensen, Jens and Bjartell, Anders and Edenbrandt, Lars and Trägårdh, Elin}}, issn = {{1619-7070}}, keywords = {{Artificial intelligence; Deep learning; PET-CT; Prostate cancer; Tumor burden}}, language = {{eng}}, number = {{5}}, pages = {{1510--1520}}, publisher = {{Springer}}, series = {{European Journal of Nuclear Medicine and Molecular Imaging}}, title = {{Automated quantification of PET/CT skeletal tumor burden in prostate cancer using artificial intelligence : The PET index}}, url = {{http://dx.doi.org/10.1007/s00259-023-06108-4}}, doi = {{10.1007/s00259-023-06108-4}}, volume = {{50}}, year = {{2023}}, }