Improved dynamic PET reconstruction using temporal decoding

Shenpeng Li; Sharna D. Jamadar; Phillip G.D. Ward; Francesco Sforazzini; Malin Premaratne; Gary F. Egan; Zhaolin Chen

doi:10.1109/NSSMIC.2018.8824624

Improved dynamic PET reconstruction using temporal decoding

Shenpeng Li, Sharna D. Jamadar, Phillip G.D. Ward, Francesco Sforazzini, Malin Premaratne, Gary F. Egan, Zhaolin Chen

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Other

Abstract

Positron Emission Tomography (PET) is a medical imaging modality which can directly measure in vivo metabolism by using various radioactive tracers. A recent development of constant infusion dynamic 18F-fluorodeoxyglucose (FDG) PET (fPET) technique can detect the dynamic changes in FDG signal in the human brain [1] , [2]. The dynamic PET is used in fPET study aiming to capture the real-time glucose utilisation in the brain. The conventional dynamic PET reconstruction with a group of 1-min independent binning frames is used in [1] , [2]. The higher temporal resolution of fPET can provide more temporal information in data analysis, but it suffers from the severe Signal-to-Noise (SNR) degradation in each frame because of the insufficient photon counts. The aim of this paper is to design a dynamic PET reconstruction method by using temporal information in projection domain. The hypothesis is that more efficient noise reduction can be achieved in projection domain because the PET reconstruction is an ill-posed problem. The proposed method is achieved by two steps: (1) Dynamic PET reconstruction on a group of encoded overlapped frames using a long temporal binning window in projection domain. (2) A robust estimation method is used to estimate the reconstruction images corresponding to the short binning window from the data obtained in step 1. The purpose of the first step is to reduce the spatial noise of PET images, and step 2 corrects the temporal distortion introduced in step 1.

Original language	English
Title of host publication	Proceedings of 2018 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Number of pages	2
ISBN (Electronic)	9781538684948
ISBN (Print)	9781538684955
DOIs	https://doi.org/10.1109/NSSMIC.2018.8824624
Publication status	Published - 5 Sep 2018
Event	IEEE Nuclear Science Symposium and Medical Imaging Conference 2018 - International Convention Centre, Sydney, Australia Duration: 10 Nov 2018 → 17 Nov 2018 https://ieeexplore.ieee.org/document/8305570

Conference

Conference	IEEE Nuclear Science Symposium and Medical Imaging Conference 2018
Abbreviated title	NSS/MIC 2018
Country/Territory	Australia
City	Sydney
Period	10/11/18 → 17/11/18
Other	The 2018 Nuclear Science Symposium will offer an outstanding opportunity for scientists and engineersworking in the field of radiation instrumentation to meet together, present new results and discuss the latestdevelopments in their field, not only with their own colleagues, but also with colleagues from related fields suchas medical imaging and room temperature semiconductors. The NSS is held contemporarily with the 2018Medical Imaging Conference and the 25th International Symposium on Room Temperature X Ray and GammaRay detectors, which brings together one of the largest collection of detector scientists and experts in the world. This very international venue, which is being held for the first time in Sydney, Australia, promises to be one of best programs in the long history of this conference. The NSS program emphasizes the latest developments in detector technology, radiation detection, detector materials, new instrumentation techniques, and their implementation in high energy and nuclear physics,astrophysics, accelerators, national nuclear security, and many other applications in various types of radiation environments. Joint sessions will be held with the MIC and RTSD, along with a number of special topic workshops covering areas of specific interest. Short Courses on a variety of topics of interest to the NSS community are also being offered.
Internet address	https://ieeexplore.ieee.org/document/8305570

Access to Document

10.1109/NSSMIC.2018.8824624

Cite this

@inproceedings{db15eb12ba1043169f949edd7c91dbd7,

title = "Improved dynamic PET reconstruction using temporal decoding",

abstract = "Positron Emission Tomography (PET) is a medical imaging modality which can directly measure in vivo metabolism by using various radioactive tracers. A recent development of constant infusion dynamic 18F-fluorodeoxyglucose (FDG) PET (fPET) technique can detect the dynamic changes in FDG signal in the human brain [1] , [2]. The dynamic PET is used in fPET study aiming to capture the real-time glucose utilisation in the brain. The conventional dynamic PET reconstruction with a group of 1-min independent binning frames is used in [1] , [2]. The higher temporal resolution of fPET can provide more temporal information in data analysis, but it suffers from the severe Signal-to-Noise (SNR) degradation in each frame because of the insufficient photon counts. The aim of this paper is to design a dynamic PET reconstruction method by using temporal information in projection domain. The hypothesis is that more efficient noise reduction can be achieved in projection domain because the PET reconstruction is an ill-posed problem. The proposed method is achieved by two steps: (1) Dynamic PET reconstruction on a group of encoded overlapped frames using a long temporal binning window in projection domain. (2) A robust estimation method is used to estimate the reconstruction images corresponding to the short binning window from the data obtained in step 1. The purpose of the first step is to reduce the spatial noise of PET images, and step 2 corrects the temporal distortion introduced in step 1.",

author = "Shenpeng Li and Jamadar, {Sharna D.} and Ward, {Phillip G.D.} and Francesco Sforazzini and Malin Premaratne and Egan, {Gary F.} and Zhaolin Chen",

year = "2018",

month = sep,

day = "5",

doi = "10.1109/NSSMIC.2018.8824624",

language = "English",

isbn = "9781538684955",

booktitle = "Proceedings of 2018 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

note = "IEEE Nuclear Science Symposium and Medical Imaging Conference 2018, NSS/MIC 2018 ; Conference date: 10-11-2018 Through 17-11-2018",

url = "https://ieeexplore.ieee.org/document/8305570",

}

Li, S, Jamadar, SD, Ward, PGD, Sforazzini, F, Premaratne, M , Egan, GF & Chen, Z 2018, Improved dynamic PET reconstruction using temporal decoding. in Proceedings of 2018 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC., 8824624, IEEE, Institute of Electrical and Electronics Engineers, IEEE Nuclear Science Symposium and Medical Imaging Conference 2018, Sydney, New South Wales, Australia, 10/11/18. https://doi.org/10.1109/NSSMIC.2018.8824624

Improved dynamic PET reconstruction using temporal decoding. / Li, Shenpeng; Jamadar, Sharna D.; Ward, Phillip G.D. et al.

Proceedings of 2018 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC. IEEE, Institute of Electrical and Electronics Engineers, 2018. 8824624.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Other

TY - GEN

T1 - Improved dynamic PET reconstruction using temporal decoding

AU - Li, Shenpeng

AU - Jamadar, Sharna D.

AU - Ward, Phillip G.D.

AU - Sforazzini, Francesco

AU - Premaratne, Malin

AU - Egan, Gary F.

AU - Chen, Zhaolin

PY - 2018/9/5

Y1 - 2018/9/5

N2 - Positron Emission Tomography (PET) is a medical imaging modality which can directly measure in vivo metabolism by using various radioactive tracers. A recent development of constant infusion dynamic 18F-fluorodeoxyglucose (FDG) PET (fPET) technique can detect the dynamic changes in FDG signal in the human brain [1] , [2]. The dynamic PET is used in fPET study aiming to capture the real-time glucose utilisation in the brain. The conventional dynamic PET reconstruction with a group of 1-min independent binning frames is used in [1] , [2]. The higher temporal resolution of fPET can provide more temporal information in data analysis, but it suffers from the severe Signal-to-Noise (SNR) degradation in each frame because of the insufficient photon counts. The aim of this paper is to design a dynamic PET reconstruction method by using temporal information in projection domain. The hypothesis is that more efficient noise reduction can be achieved in projection domain because the PET reconstruction is an ill-posed problem. The proposed method is achieved by two steps: (1) Dynamic PET reconstruction on a group of encoded overlapped frames using a long temporal binning window in projection domain. (2) A robust estimation method is used to estimate the reconstruction images corresponding to the short binning window from the data obtained in step 1. The purpose of the first step is to reduce the spatial noise of PET images, and step 2 corrects the temporal distortion introduced in step 1.

AB - Positron Emission Tomography (PET) is a medical imaging modality which can directly measure in vivo metabolism by using various radioactive tracers. A recent development of constant infusion dynamic 18F-fluorodeoxyglucose (FDG) PET (fPET) technique can detect the dynamic changes in FDG signal in the human brain [1] , [2]. The dynamic PET is used in fPET study aiming to capture the real-time glucose utilisation in the brain. The conventional dynamic PET reconstruction with a group of 1-min independent binning frames is used in [1] , [2]. The higher temporal resolution of fPET can provide more temporal information in data analysis, but it suffers from the severe Signal-to-Noise (SNR) degradation in each frame because of the insufficient photon counts. The aim of this paper is to design a dynamic PET reconstruction method by using temporal information in projection domain. The hypothesis is that more efficient noise reduction can be achieved in projection domain because the PET reconstruction is an ill-posed problem. The proposed method is achieved by two steps: (1) Dynamic PET reconstruction on a group of encoded overlapped frames using a long temporal binning window in projection domain. (2) A robust estimation method is used to estimate the reconstruction images corresponding to the short binning window from the data obtained in step 1. The purpose of the first step is to reduce the spatial noise of PET images, and step 2 corrects the temporal distortion introduced in step 1.

UR - http://www.scopus.com/inward/record.url?scp=85073119803&partnerID=8YFLogxK

U2 - 10.1109/NSSMIC.2018.8824624

DO - 10.1109/NSSMIC.2018.8824624

M3 - Conference Paper

SN - 9781538684955

BT - Proceedings of 2018 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - IEEE Nuclear Science Symposium and Medical Imaging Conference 2018

Y2 - 10 November 2018 through 17 November 2018

ER -

Improved dynamic PET reconstruction using temporal decoding

Abstract

Conference

Access to Document

Other files and links

ARC Centre of Excellence for Integrative Brain Function

Cite this

Improved dynamic PET reconstruction using temporal decoding

Abstract

Conference

Access to Document

Other files and links

Projects

ARC Centre of Excellence for Integrative Brain Function

Cite this