Skip to main content
Key Specialties
Anesthesiology Cardiology Dermatology Diabetology Endocrinology Gastroenterology and Hepatology General Medicine Geriatrics Hematology Infectious Diseases Internal Medicine Nephrology Neurology Obstetrics and Gynecology Oncology Ophthalmology Pediatrics Psychiatry Radiology Respiratory Medicine Rheumatology Surgery Urology
Congress Coverage
ASH 2024 Annual Meeting 2024 ESMO Congress 2024 ERS Congress 2024 EASD Congress 2024 ESC Congress 2024 EAN Congress 2024 ASCO Annual Meeting 2024 ACC Congress
Clinical Collections
Advances in Alzheimer’s

Podcast | Sexual health in older age

Dr Holly Thomas helps us unpack the needlessly taboo subject of sex and sexual health in women of older age.

Listen now
ADVANCED SEARCH
Log in
Top
European Journal of Nuclear Medicine and Molecular Imaging

Open Access 12-03-2025 | Positron Emission Tomography | Original Article

Impact of deep learning denoising on kinetic modelling for low-dose dynamic PET: application to single- and dual-tracer imaging protocols

Authors: Florence M. Muller, Elizabeth J. Li, Margaret E. Daube-Witherspoon, Austin R. Pantel, Corinde E. Wiers, Jacob G. Dubroff, Christian Vanhove, Stefaan Vandenberghe, Joel S. Karp

Published in: European Journal of Nuclear Medicine and Molecular Imaging

Login to get access

Abstract

Purpose

Long-axial field-of-view PET scanners capture multi-organ tracer distribution with high sensitivity, enabling lower dose dynamic protocols and dual-tracer imaging for comprehensive disease characterization. However, reducing dose may compromise data quality and time-activity curve (TAC) fitting, leading to higher bias in kinetic parameters. Parametric imaging poses further challenges due to noise amplification in voxel-based modelling. We explore the potential of deep learning denoising (DL-DN) to improve quantification for low-dose dynamic PET.

Methods

Using 16 [18F]FDG PET studies from the PennPET Explorer, we trained a DL framework on 10-min images from late-phase uptake (static data) that were sub-sampled from 1/2 to 1/300 of the counts. This model was used to denoise early-to-late dynamic frame images. Its impact on quantification was evaluated using compartmental modelling and voxel-based graphical analysis for parametric imaging for single- and dual-tracer dynamic studies with [18F]FDG and [18F]FGln at original (injected) and reduced (sub-sampled) doses. Quantification differences were evaluated for the area under the curve of TACs, Ki for [18F]FDG and VT for [18F]FGln, and parametric images.

Results

DL-DN consistently improved image quality across all dynamic frames, systematically enhancing TAC consistency and reducing tissue-dependent bias and variability in Ki and VT down to 40 MBq doses. DL-DN preserved tumor heterogeneity in Logan VT images and delineation of high-flux regions in Patlak Ki maps. In a /[18F]FDG dual-tracer study, bias trends aligned with single-tracer results but showed reduced accuracy for [¹⁸F]FGln in breast lesions at very low doses (4 MBq).

Conclusion

This study demonstrates that applying DL-DN trained on static [18F]FDG PET images to dynamic [18F]FDG and [18F]FGln PET can permit significantly reduced doses, preserving accurate FDG Ki and FGln VT measurements, and enhancing parametric image quality. DL-DN shows promise for improving dynamic PET quantification at reduced doses, including novel dual-tracer studies.
Appendix
Available only for authorised users
Literature
1.
2.
3.
4.
Doot RK. Getting the most out of 18F-FDG PET scans: the predictive value of 18F-FDG PET–derived blood flow estimates for breast cancer. J Nucl Med. 2016;57:1667–8.CrossRefPubMedPubMedCentral
5.
Yang M, Lin Z, Xu Z, Li D, Lv W, Yang S, et al. Influx rate constant of 18 F-FDG increases in metastatic lymph nodes of non-small cell lung cancer patients. Eur J Nucl Med Mol Imaging. 2020;47:1198–208.CrossRefPubMed
6.
Chen R, Ng YL, Zhao H, Ge Q, Shi F, Cao K, et al. Quantitative parametric imaging added diagnostic values in lesion detection of 68Ga-PSMA-11 in prostate cancer patients identified by dynamic total-body PET/CT. Soc Nuclear Med; 2022.
7.
8.
Viswanath V, Pantel AR, Daube-Witherspoon ME, Doot R, Muzi M, Mankoff DA, et al. Quantifying bias and precision of kinetic parameter Estimation on the PennPET explorer, a long axial field-of-view scanner. IEEE Trans Radiation Plasma Med Sci. 2020;4:735–49.CrossRef
9.
Tan H, Qi C, Cao Y, Cai D, Mao W, Yu H, et al. Ultralow-dose [18F] FDG PET/CT imaging: demonstration of feasibility in dynamic and static images. Eur Radiol. 2023;33:5017–27.CrossRefPubMed
10.
Wang G, Rahmim A, Gunn RN. PET parametric imaging: past, present, and future. IEEE Trans Radiation Plasma Med Sci. 2020;4:663–75.CrossRef
11.
12.
Mankoff DA, Pantel AR, Viswanath V, Karp JS. Advances in PET diagnostics for guiding targeted cancer therapy and studying in vivo cancer biology. Curr Pathobiology Rep. 2019;7:97–108.CrossRef
13.
Badawi RD, Shi H, Hu P, Chen S, Xu T, Price PM, et al. First human imaging studies with the EXPLORER total-body PET scanner. J Nucl Med. 2019;60:299–303.CrossRefPubMedPubMedCentral
14.
15.
Sari H, Eriksson L, Mingels C, Alberts I, Casey ME, Afshar-Oromieh A, et al. Feasibility of using abbreviated scan protocols with population-based input functions for accurate kinetic modeling of [18F]-FDG datasets from a long axial FOV PET scanner. Eur J Nucl Med Mol Imaging. 2023;50:257–65.CrossRefPubMed
16.
Viswanath V, Sari H, Pantel AR, Conti M, Daube-Witherspoon ME, Mingels C, et al. Abbreviated scan protocols to capture 18F-FDG kinetics for long axial FOV PET scanners. Eur J Nucl Med Mol Imaging. 2022;49:3215–25.CrossRefPubMedPubMedCentral
17.
Surti S, Pantel AR, Karp JS. Total body PET: why, how, what for? IEEE Trans Radiation Plasma Med Sci. 2020;4:283–92.CrossRef
18.
19.
Wang G. High temporal-resolution dynamic PET image reconstruction using a new Spatiotemporal kernel method. IEEE Trans Med Imaging. 2018;38:664–74.CrossRefPubMedCentral
20.
21.
22.
Ote K, Hashimoto F, Kakimoto A, Isobe T, Inubushi T, Ota R, et al. Kinetics-induced block matching and 5-D transform domain filtering for dynamic PET image denoising. IEEE Trans Radiation Plasma Med Sci. 2020;4:720–8.CrossRef
23.
Cheng JC, Bevington C, Rahmim A, Klyuzhin I, Matthews J, Boellaard R, et al. Dynamic PET image reconstruction utilizing intrinsic data-driven HYPR4D denoising kernel. Med Phys. 2021;48:2230–44.CrossRefPubMed
24.
Vashistha R, Moradi H, Hammond A, O’Brien K, Rominger A, Sari H, et al. ParaPET: non-invasive deep learning method for direct parametric brain PET reconstruction using histoimages. EJNMMI Res. 2024;14:10.CrossRefPubMedPubMedCentral
25.
26.
Gong K, Catana C, Qi J, Li Q. Direct reconstruction of linear parametric images from dynamic PET using nonlocal deep image prior. IEEE Trans Med Imaging. 2021;41:680–9.CrossRef
27.
Hashimoto F, Ohba H, Ote K, Teramoto A, Tsukada H. Dynamic PET image denoising using deep convolutional neural networks without prior training datasets. IEEE Access. 2019;7:96594–603.CrossRef
28.
Hashimoto F, Ohba H, Ote K, Kakimoto A, Tsukada H, Ouchi Y. 4D deep image prior: dynamic PET image denoising using an unsupervised four-dimensional branch convolutional neural network. Phys Med Biol. 2021;66:015006.CrossRefPubMed
29.
30.
31.
Kwon D, Dulal C, Daube-Witherspoon ME, Viswanath V, Hou-Liu J, Mankoff DA et al. Dual-Tracer Dynamic PET Imaging on the Long-Axial Field-Of-View PennPET Explorer: A Proof-Of-Principle Study With [18F] Fluoroglutamine and [18F] Fluorodeoxyglucose. In RSNA Annual Meeting. 2023.
32.
33.
Li X, Young AJ, Pereira-Rufino LS, Shi Z, Byanyima JI, Vesslee S, et al. Pharmacokinetic effects of a single-dose nutritional ketone ester supplement on brain ketone and glucose metabolism in alcohol use disorder. MedRxiv. 2023;202309:25–23296090.
34.
Matej S, Lewitt RM. Practical considerations for 3-D image reconstruction using spherically symmetric volume elements. IEEE Trans Med Imaging. 1996;15:68–78.CrossRefPubMed
35.
Popescu LM, Matej S, Lewitt RM. Iterative image reconstruction using geometrically ordered subsets with list-mode data. IEEE Symposium Conference Record Nuclear Science 2004: IEEE; 2004. pp. 3536-40.
36.
Werner ME, Surti S, Karp JS. Implementation and evaluation of a 3D PET single scatter simulation with TOF modeling. 2006 IEEE nuclear science symposium conference record: IEEE; 2006. pp. 1768-73.
37.
Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5–9, 2015, Proceedings, Part III 18: Springer; 2015. pp. 234– 41.
38.
He K, Zhang X, Ren S, Sun J. Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. Proceedings of the IEEE international conference on computer vision; 2015. pp. 1026-34.
39.
He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition; 2016. pp. 770-8.
40.
Singh G, Mittal A, Aggarwal N. ResDNN: deep residual learning for natural image denoising. IET Image Proc. 2020;14:2425–34.CrossRef
41.
Oktay O, Schlemper J, Folgoc LL, Lee M, Heinrich M, Misawa K et al. Attention u-net: Learning where to look for the pancreas. arXiv preprint arXiv:180403999. 2018.
42.
Chen T, Hou J, Zhou Y, Xie H, Chen X, Liu Q et al. 2.5 D Multi-view averaging diffusion model for 3D medical image translation: application to Low-count PET reconstruction with CT-less Attenuation correction. arXiv preprint arXiv:240608374. 2024.
43.
Kingma DP, Ba J, Adam. A method for stochastic optimization. ArXiv Preprint arXiv:14126980. 2014.
44.
Wang G, Corwin MT, Olson KA, Badawi RD, Sarkar S. Dynamic PET of human liver inflammation: impact of kinetic modeling with optimization-derived dual-blood input function. Phys Med Biol. 2018;63:155004.CrossRefPubMedPubMedCentral
45.
Phelps ME, Huang S, Hoffman E, Selin C, Sokoloff L, Kuhl DE. Tomographic measurement of local cerebral glucose metabolic rate in humans with (F-18) 2‐fluoro‐2‐deoxy‐D‐glucose: validation of method. Annals Neurology: Official J Am Neurol Association Child Neurol Soc. 1979;6:371–88.CrossRef
46.
Patlak CS, Blasberg RG, Fenstermacher JD. Graphical evaluation of blood-to-brain transfer constants from multiple-time uptake data. J Cereb Blood Flow Metabolism. 1983;3:1–7.CrossRef
47.
Logan J. Graphical analysis of PET data applied to reversible and irreversible tracers. Nucl Med Biol. 2000;27:661–70.CrossRefPubMed
48.
Dahlbom M. Estimation of image noise in PET using the bootstrap method. 2001 IEEE Nuclear Science Symposium Conference Record (Cat No 01CH37310): IEEE; 2001. pp. 2075-9.
49.
Buvat I. A non-parametric bootstrap approach for analysing the statistical properties of SPECT and PET images. Phys Med Biol. 2002;47:1761.CrossRefPubMed
50.
Chen GP, Branch KR, Alessio AM, Pham P, Tabibiazar R, Kinahan P, et al. Effect of reconstruction algorithms on myocardial blood flow measurement with 13 N-ammonia PET. J Nucl Med. 2007;48:1259–65.CrossRefPubMed
51.
Li EJ, Spencer BA, Schmall JP, Abdelhafez Y, Badawi RD, Wang G, et al. Efficient delay correction for total-body PET kinetic modeling using pulse timing methods. J Nucl Med. 2022;63:1266–73.CrossRefPubMedPubMedCentral
52.
Chicklore S, Goh V, Siddique M, Roy A, Marsden PK, Cook GJ. Quantifying tumour heterogeneity in 18 F-FDG PET/CT imaging by texture analysis. Eur J Nucl Med Mol Imaging. 2013;40:133–40.CrossRefPubMed
53.
Chitalia R, Viswanath V, Pantel AR, Peterson LM, Gastounioti A, Cohen EA, et al. Functional 4-D clustering for characterizing intratumor heterogeneity in dynamic imaging: evaluation in FDG PET as a prognostic biomarker for breast cancer. Eur J Nucl Med Mol Imaging. 2021;48:3990–4001.CrossRefPubMedPubMedCentral
54.
Chen Z, Wu Y, Zhang N, Sun T, Shen Y, Zheng H, et al. High Temporal resolution total-body dynamic PET imaging based on pixel-level time-activity curve correction. IEEE Trans Biomed Eng. 2022;69:3689–702.CrossRefPubMed
55.
De Benetti F, Simson W, Paschali M, Sari H, Rominger A, Shi K et al. Self-Supervised Learning for Physiologically-Based Pharmacokinetic Modeling in Dynamic PET. International Conference on Medical Image Computing and Computer-Assisted Intervention: Springer; 2023. pp. 290-9.
Metadata
Title
Impact of deep learning denoising on kinetic modelling for low-dose dynamic PET: application to single- and dual-tracer imaging protocols
Authors
Florence M. Muller
Elizabeth J. Li
Margaret E. Daube-Witherspoon
Austin R. Pantel
Corinde E. Wiers
Jacob G. Dubroff
Christian Vanhove
Stefaan Vandenberghe
Joel S. Karp
Publication date
12-03-2025
Publisher
Springer Berlin Heidelberg
Published in
European Journal of Nuclear Medicine and Molecular Imaging
Print ISSN: 1619-7070
Electronic ISSN: 1619-7089
DOI
https://doi.org/10.1007/s00259-025-07182-6