MDPET: A Unified Motion Correction and Denoising Adversarial Network for Low-Dose Gated PET

Overview

Journal IEEE Trans Med Imaging

Date 2021 Apr 28

PMID 33909561

Citations 20

Authors

Bo Zhou

Yu-Jung Tsai

Xiongchao Chen

James S Duncan

Chi Liu

Affiliations

Soon will be listed here.

Abstract

In positron emission tomography (PET), gating is commonly utilized to reduce respiratory motion blurring and to facilitate motion correction methods. In application where low-dose gated PET is useful, reducing injection dose causes increased noise levels in gated images that could corrupt motion estimation and subsequent corrections, leading to inferior image quality. To address these issues, we propose MDPET, a unified motion correction and denoising adversarial network for generating motion-compensated low-noise images from low-dose gated PET data. Specifically, we proposed a Temporal Siamese Pyramid Network (TSP-Net) with basic units made up of 1.) Siamese Pyramid Network (SP-Net), and 2.) a recurrent layer for motion estimation among the gates. The denoising network is unified with our motion estimation network to simultaneously correct the motion and predict a motion-compensated denoised PET reconstruction. The experimental results on human data demonstrated that our MDPET can generate accurate motion estimation directly from low-dose gated images and produce high-quality motion-compensated low-noise reconstructions. Comparative studies with previous methods also show that our MDPET is able to generate superior motion estimation and denoising performance. Our code is available at https://github.com/bbbbbbzhou/MDPET.

Citing Articles

Robust whole-body PET image denoising using 3D diffusion models: evaluation across various scanners, tracers, and dose levels.

Yu B, Ozdemir S, Dong Y, Shao W, Pan T, Shi K Eur J Nucl Med Mol Imaging. 2025; .

PMID: 39912940 DOI: 10.1007/s00259-025-07122-4.

A Review on Low-Dose Emission Tomography Post-Reconstruction Denoising with Neural Network Approaches.

Bousse A, Kandarpa V, Shi K, Gong K, Lee J, Liu C IEEE Trans Radiat Plasma Med Sci. 2024; 8(4):333-347.

PMID: 39429805 PMC: 11486494. DOI: 10.1109/trpms.2023.3349194.

Unified Noise-aware Network for Low-count PET Denoising with Varying Count Levels.

Xie H, Liu Q, Zhou B, Chen X, Guo X, Wang H IEEE Trans Radiat Plasma Med Sci. 2024; 8(4):366-378.

PMID: 39391291 PMC: 11463975. DOI: 10.1109/trpms.2023.3334105.

Deep learning-based techniques for estimating high-quality full-dose positron emission tomography images from low-dose scans: a systematic review.

Seyyedi N, Ghafari A, Seyyedi N, Sheikhzadeh P BMC Med Imaging. 2024; 24(1):238.

PMID: 39261796 PMC: 11391655. DOI: 10.1186/s12880-024-01417-y.

Deep learning applications for quantitative and qualitative PET in PET/MR: technical and clinical unmet needs.

Yang J, Afaq A, Sibley R, McMilan A, Pirasteh A MAGMA. 2024; 37(4):749-763.

PMID: 39167304 DOI: 10.1007/s10334-024-01199-y.

References

Dutta J, Leahy R, Li Q . Non-local means denoising of dynamic PET images. PLoS One. 2013; 8(12):e81390. PMC: 3855264. DOI: 10.1371/journal.pone.0081390. View

Kaplan S, Zhu Y . Full-Dose PET Image Estimation from Low-Dose PET Image Using Deep Learning: a Pilot Study. J Digit Imaging. 2018; 32(5):773-778. PMC: 6737135. DOI: 10.1007/s10278-018-0150-3. View

Chen K, Gong E, Carvalho Macruz F, Xu J, Boumis A, Khalighi M . Ultra-Low-Dose F-Florbetaben Amyloid PET Imaging Using Deep Learning with Multi-Contrast MRI Inputs. Radiology. 2018; 290(3):649-656. PMC: 6394782. DOI: 10.1148/radiol.2018180940. View

Zhou L, Schaefferkoetter J, Tham I, Huang G, Yan J . Supervised learning with cyclegan for low-dose FDG PET image denoising. Med Image Anal. 2020; 65:101770. DOI: 10.1016/j.media.2020.101770. View

Van Der Gucht A, Serrano B, Hugonnet F, Paulmier B, Garnier N, Faraggi M . Impact of a new respiratory amplitude-based gating technique in evaluation of upper abdominal PET lesions. Eur J Radiol. 2013; 83(3):509-15. DOI: 10.1016/j.ejrad.2013.11.010. View

Liu H, Wu J, Lu W, Onofrey J, Liu Y, Liu C . Noise reduction with cross-tracer and cross-protocol deep transfer learning for low-dose PET. Phys Med Biol. 2020; 65(18):185006. DOI: 10.1088/1361-6560/abae08. View

Strauss K, Kaste S . The ALARA (as low as reasonably achievable) concept in pediatric interventional and fluoroscopic imaging: striving to keep radiation doses as low as possible during fluoroscopy of pediatric patients--a white paper executive summary. Radiology. 2006; 240(3):621-2. DOI: 10.1148/radiol.2403060698. View

Ouyang J, Chen K, Gong E, Pauly J, Zaharchuk G . Ultra-low-dose PET reconstruction using generative adversarial network with feature matching and task-specific perceptual loss. Med Phys. 2019; 46(8):3555-3564. PMC: 6692211. DOI: 10.1002/mp.13626. View

Shan H, Padole A, Homayounieh F, Kruger U, Doda Khera R, Nitiwarangkul C . Competitive performance of a modularized deep neural network compared to commercial algorithms for low-dose CT image reconstruction. Nat Mach Intell. 2020; 1(6):269-276. PMC: 7687920. DOI: 10.1038/s42256-019-0057-9. View

10.

Normandin M, Petersen K, Ding Y, Lin S, Naik S, Fowles K . In vivo imaging of endogenous pancreatic β-cell mass in healthy and type 1 diabetic subjects using 18F-fluoropropyl-dihydrotetrabenazine and PET. J Nucl Med. 2012; 53(6):908-16. PMC: 3737743. DOI: 10.2967/jnumed.111.100545. View

11.

Xiang L, Qiao Y, Nie D, An L, Wang Q, Shen D . Deep Auto-context Convolutional Neural Networks for Standard-Dose PET Image Estimation from Low-Dose PET/MRI. Neurocomputing (Amst). 2017; 267:406-416. PMC: 5714510. DOI: 10.1016/j.neucom.2017.06.048. View

12.

Dawood M, Buther F, Jiang X, Schafers K . Respiratory motion correction in 3-D PET data with advanced optical flow algorithms. IEEE Trans Med Imaging. 2008; 27(8):1164-75. DOI: 10.1109/TMI.2008.918321. View

13.

Cui J, Gong K, Guo N, Wu C, Meng X, Kim K . PET image denoising using unsupervised deep learning. Eur J Nucl Med Mol Imaging. 2019; 46(13):2780-2789. PMC: 7814987. DOI: 10.1007/s00259-019-04468-4. View

14.

Papademetris X, Jackowski M, Rajeevan N, DiStasio M, Okuda H, Constable R . BioImage Suite: An integrated medical image analysis suite: An update. Insight J. 2014; 2006:209. PMC: 4213804. View

15.

Lu W, Onofrey J, Lu Y, Shi L, Ma T, Liu Y . An investigation of quantitative accuracy for deep learning based denoising in oncological PET. Phys Med Biol. 2019; 64(16):165019. DOI: 10.1088/1361-6560/ab3242. View

16.

Chan C, Onofrey J, Jian Y, Germino M, Papademetris X, Carson R . Non-Rigid Event-by-Event Continuous Respiratory Motion Compensated List-Mode Reconstruction for PET. IEEE Trans Med Imaging. 2017; 37(2):504-515. PMC: 7304524. DOI: 10.1109/TMI.2017.2761756. View

17.

Gong Y, Shan H, Teng Y, Tu N, Li M, Liang G . Parameter-Transferred Wasserstein Generative Adversarial Network (PT-WGAN) for Low-Dose PET Image Denoising. IEEE Trans Radiat Plasma Med Sci. 2022; 5(2):213-223. PMC: 8993163. DOI: 10.1109/trpms.2020.3025071. View

18.

Wang Y, Yu B, Wang L, Zu C, Lalush D, Lin W . 3D conditional generative adversarial networks for high-quality PET image estimation at low dose. Neuroimage. 2018; 174:550-562. PMC: 6410574. DOI: 10.1016/j.neuroimage.2018.03.045. View

19.

Callahan J, Kron T, Siva S, Simoens N, Edgar A, Everitt S . Geographic miss of lung tumours due to respiratory motion: a comparison of 3D vs 4D PET/CT defined target volumes. Radiat Oncol. 2014; 9:291. PMC: 4278238. DOI: 10.1186/s13014-014-0291-6. View

20.

Maggioni M, Katkovnik V, Egiazarian K, Foi A . Nonlocal transform-domain filter for volumetric data denoising and reconstruction. IEEE Trans Image Process. 2012; 22(1):119-33. DOI: 10.1109/TIP.2012.2210725. View