PET image reconstruction using kernel method

Guobao Wang, Jinyi Qi

Research output: Contribution to journalArticle

48 Citations (Scopus)

Abstract

Image reconstruction from low-count positron emission tomography (PET) projection data is challenging because the inverse problem is ill-posed. Prior information can be used to improve image quality. Inspired by the kernel methods in machine learning, this paper proposes a kernel based method that models PET image intensity in each pixel as a function of a set of features obtained from prior information. The kernel-based image model is incorporated into the forward model of PET projection data and the coefficients can be readily estimated by the maximum likelihood (ML) or penalized likelihood image reconstruction. A kernelized expectation-maximization algorithm is presented to obtain the ML estimate. Computer simulations show that the proposed approach can achieve better bias versus variance trade-off and higher contrast recovery for dynamic PET image reconstruction than the conventional maximum likelihood method with and without post-reconstruction denoising. Compared with other regularization-based methods, the kernel method is easier to implement and provides better image quality for low-count data. Application of the proposed kernel method to a 4-D dynamic PET patient dataset showed promising results.

Original languageEnglish (US)
Article number6868314
Pages (from-to)61-71
Number of pages11
JournalIEEE Transactions on Medical Imaging
Volume34
Issue number1
DOIs
StatePublished - Jan 1 2015

Fingerprint

Positron emission tomography
Computer-Assisted Image Processing
Image reconstruction
Positron-Emission Tomography
Maximum likelihood
Image quality
Likelihood Functions
Inverse problems
Learning systems
Computer Simulation
Pixels
Recovery
Computer simulation

Keywords

  • Expectation maximization (EM)
  • image prior
  • image reconstruction
  • kernel method
  • positron emission tomography (PET)

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Radiological and Ultrasound Technology
  • Software

Cite this

PET image reconstruction using kernel method. / Wang, Guobao; Qi, Jinyi.

In: IEEE Transactions on Medical Imaging, Vol. 34, No. 1, 6868314, 01.01.2015, p. 61-71.

Research output: Contribution to journalArticle

@article{c05d09f442d04e9db3c57ee6f8bd43e5,
title = "PET image reconstruction using kernel method",
abstract = "Image reconstruction from low-count positron emission tomography (PET) projection data is challenging because the inverse problem is ill-posed. Prior information can be used to improve image quality. Inspired by the kernel methods in machine learning, this paper proposes a kernel based method that models PET image intensity in each pixel as a function of a set of features obtained from prior information. The kernel-based image model is incorporated into the forward model of PET projection data and the coefficients can be readily estimated by the maximum likelihood (ML) or penalized likelihood image reconstruction. A kernelized expectation-maximization algorithm is presented to obtain the ML estimate. Computer simulations show that the proposed approach can achieve better bias versus variance trade-off and higher contrast recovery for dynamic PET image reconstruction than the conventional maximum likelihood method with and without post-reconstruction denoising. Compared with other regularization-based methods, the kernel method is easier to implement and provides better image quality for low-count data. Application of the proposed kernel method to a 4-D dynamic PET patient dataset showed promising results.",
keywords = "Expectation maximization (EM), image prior, image reconstruction, kernel method, positron emission tomography (PET)",
author = "Guobao Wang and Jinyi Qi",
year = "2015",
month = "1",
day = "1",
doi = "10.1109/TMI.2014.2343916",
language = "English (US)",
volume = "34",
pages = "61--71",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - PET image reconstruction using kernel method

AU - Wang, Guobao

AU - Qi, Jinyi

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Image reconstruction from low-count positron emission tomography (PET) projection data is challenging because the inverse problem is ill-posed. Prior information can be used to improve image quality. Inspired by the kernel methods in machine learning, this paper proposes a kernel based method that models PET image intensity in each pixel as a function of a set of features obtained from prior information. The kernel-based image model is incorporated into the forward model of PET projection data and the coefficients can be readily estimated by the maximum likelihood (ML) or penalized likelihood image reconstruction. A kernelized expectation-maximization algorithm is presented to obtain the ML estimate. Computer simulations show that the proposed approach can achieve better bias versus variance trade-off and higher contrast recovery for dynamic PET image reconstruction than the conventional maximum likelihood method with and without post-reconstruction denoising. Compared with other regularization-based methods, the kernel method is easier to implement and provides better image quality for low-count data. Application of the proposed kernel method to a 4-D dynamic PET patient dataset showed promising results.

AB - Image reconstruction from low-count positron emission tomography (PET) projection data is challenging because the inverse problem is ill-posed. Prior information can be used to improve image quality. Inspired by the kernel methods in machine learning, this paper proposes a kernel based method that models PET image intensity in each pixel as a function of a set of features obtained from prior information. The kernel-based image model is incorporated into the forward model of PET projection data and the coefficients can be readily estimated by the maximum likelihood (ML) or penalized likelihood image reconstruction. A kernelized expectation-maximization algorithm is presented to obtain the ML estimate. Computer simulations show that the proposed approach can achieve better bias versus variance trade-off and higher contrast recovery for dynamic PET image reconstruction than the conventional maximum likelihood method with and without post-reconstruction denoising. Compared with other regularization-based methods, the kernel method is easier to implement and provides better image quality for low-count data. Application of the proposed kernel method to a 4-D dynamic PET patient dataset showed promising results.

KW - Expectation maximization (EM)

KW - image prior

KW - image reconstruction

KW - kernel method

KW - positron emission tomography (PET)

UR - http://www.scopus.com/inward/record.url?scp=84920141881&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920141881&partnerID=8YFLogxK

U2 - 10.1109/TMI.2014.2343916

DO - 10.1109/TMI.2014.2343916

M3 - Article

VL - 34

SP - 61

EP - 71

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 1

M1 - 6868314

ER -