Clinical and phantom validation of a deep learning based denoising algorithm for F-18-FDG PET images from lower detection counting in comparison with the standard acquisition.
Deep-learning
Denoising
PET/CT
Journal
EJNMMI physics
ISSN: 2197-7364
Titre abrégé: EJNMMI Phys
Pays: Germany
ID NLM: 101658952
Informations de publication
Date de publication:
11 May 2022
11 May 2022
Historique:
received:
10
08
2021
accepted:
20
04
2022
entrez:
11
5
2022
pubmed:
12
5
2022
medline:
12
5
2022
Statut:
epublish
Résumé
PET/CT image quality is directly influenced by the F-18-FDG injected activity. The higher the injected activity, the less noise in the reconstructed images but the more radioactive staff exposition. A new FDA cleared software has been introduced to obtain clinical PET images, acquired at 25% of the count statistics considering US practices. Our aim is to determine the limits of a deep learning based denoising algorithm (SubtlePET) applied to statistically reduced PET raw data from 3 different last generation PET scanners in comparison to the regular acquisition in phantom and patients, considering the European guidelines for radiotracer injection activities. Images of low and high contrasted (SBR = 2 and 5) spheres of the IEC phantom and high contrast (SBR = 5) of micro-spheres of Jaszczak phantom were acquired on 3 different PET devices. 110 patients with different pathologies were included. The data was acquired in list-mode and retrospectively reconstructed with the regular acquisition count statistic (PET100), 50% reduction in counts (PET50) and 66% reduction in counts (PET33). These count reduced images were post-processed with SubtlePET to obtain PET50 + SP and PET33 + SP images. Patient image quality was scored by 2 senior nuclear physicians. Peak-signal-to-Noise and Structural similarity metrics were computed to compare the low count images to regular acquisition (PET100). SubtlePET reliably denoised the images and maintained the SUV Based on our results, SubtlePET is adapted in clinical practice for half-time or half-dose acquisitions based on European recommended injected dose of 3 MBq/kg without diagnostic confidence loss.
Sections du résumé
BACKGROUND
BACKGROUND
PET/CT image quality is directly influenced by the F-18-FDG injected activity. The higher the injected activity, the less noise in the reconstructed images but the more radioactive staff exposition. A new FDA cleared software has been introduced to obtain clinical PET images, acquired at 25% of the count statistics considering US practices. Our aim is to determine the limits of a deep learning based denoising algorithm (SubtlePET) applied to statistically reduced PET raw data from 3 different last generation PET scanners in comparison to the regular acquisition in phantom and patients, considering the European guidelines for radiotracer injection activities. Images of low and high contrasted (SBR = 2 and 5) spheres of the IEC phantom and high contrast (SBR = 5) of micro-spheres of Jaszczak phantom were acquired on 3 different PET devices. 110 patients with different pathologies were included. The data was acquired in list-mode and retrospectively reconstructed with the regular acquisition count statistic (PET100), 50% reduction in counts (PET50) and 66% reduction in counts (PET33). These count reduced images were post-processed with SubtlePET to obtain PET50 + SP and PET33 + SP images. Patient image quality was scored by 2 senior nuclear physicians. Peak-signal-to-Noise and Structural similarity metrics were computed to compare the low count images to regular acquisition (PET100).
RESULTS
RESULTS
SubtlePET reliably denoised the images and maintained the SUV
CONCLUSION
CONCLUSIONS
Based on our results, SubtlePET is adapted in clinical practice for half-time or half-dose acquisitions based on European recommended injected dose of 3 MBq/kg without diagnostic confidence loss.
Identifiants
pubmed: 35543894
doi: 10.1186/s40658-022-00465-z
pii: 10.1186/s40658-022-00465-z
pmc: PMC9095795
doi:
Types de publication
Journal Article
Langues
eng
Pagination
36Informations de copyright
© 2022. The Author(s).
Références
IEEE Trans Med Imaging. 2003 Sep;22(9):1042-52
pubmed: 12956260
Hell J Nucl Med. 2016 Jan-Apr;19(1):15-8
pubmed: 26929936
IEEE Trans Biomed Eng. 2017 Mar;64(3):569-579
pubmed: 27187939
J Nucl Med. 2008 Apr;49(4):657-66
pubmed: 18344430
Phys Med Biol. 2019 Aug 21;64(16):165019
pubmed: 31307019
Med Image Anal. 2013 Dec;17(8):877-91
pubmed: 23837964
EJNMMI Phys. 2019 Dec 26;6(1):28
pubmed: 31879795
EJNMMI Phys. 2020 May 12;7(1):28
pubmed: 32399752
IEEE Trans Med Imaging. 2014 Mar;33(3):636-50
pubmed: 24595339
PLoS One. 2013 Dec 05;8(12):e81390
pubmed: 24339921
Med Image Anal. 2019 May;54:253-262
pubmed: 30954852
IEEE Trans Radiat Plasma Med Sci. 2019 Mar;3(2):153-161
pubmed: 32754674
IEEE Trans Med Imaging. 2019 Mar;38(3):675-685
pubmed: 30222554
EJNMMI Res. 2020 Sep 21;10(1):105
pubmed: 32955669
J Nucl Med. 2015 Sep;56(9):1447-52
pubmed: 26159585
Med Phys. 2020 Oct;47(10):4928-4938
pubmed: 32687608
Phys Med Biol. 2016 Jan 21;61(2):791-812
pubmed: 26732849
EJNMMI Phys. 2021 Mar 9;8(1):25
pubmed: 33687602
Mol Imaging. 2019 Jan-Dec;18:1536012119869070
pubmed: 31429375