Machine Learning in PET: From Photon Detection to Quantitative Image Reconstruction

Kuang Gong, Eric Berg, Simon R. Cherry, Jinyi Qi

Research output: Contribution to journalArticle

Abstract

Machine learning has found unique applications in nuclear medicine from photon detection to quantitative image reconstruction. Although there have been impressive strides in detector development for time-of-flight positron emission tomography (PET), most detectors still make use of simple signal processing methods to extract the time and position information from the detector signals. Now, with the availability of fast waveform digitizers, machine learning techniques have been applied to estimate the position and arrival time of high-energy photons. In quantitative image reconstruction, machine learning has been used to estimate various corrections factors, including scattered events and attenuation images, as well as to reduce statistical noise in reconstructed images. Here, machine learning either provides a faster alternative to an existing time-consuming computation, such as in the case of scatter estimation, or creates a data-driven approach to map an implicitly defined function, such as in the case of estimating the attenuation map for PET/MR scans. In this article, we will review the above-mentioned applications of machine learning in nuclear medicine.

Original languageEnglish (US)
Article number8844693
Pages (from-to)51-68
Number of pages18
JournalProceedings of the IEEE
Volume108
Issue number1
DOIs
StatePublished - Jan 2020

Fingerprint

Positron emission tomography
Image reconstruction
Learning systems
Photons
Nuclear medicine
Detectors
Signal processing
Availability

Keywords

  • Attenuation correction
  • deep learning
  • denoising
  • image reconstruction
  • machine learning
  • positron emission tomography (PET)
  • scatter correction
  • timing resolution

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Machine Learning in PET : From Photon Detection to Quantitative Image Reconstruction. / Gong, Kuang; Berg, Eric; Cherry, Simon R.; Qi, Jinyi.

In: Proceedings of the IEEE, Vol. 108, No. 1, 8844693, 01.2020, p. 51-68.

Research output: Contribution to journalArticle

@article{439da0ced3a44f81a8eb11c2a529efb1,
title = "Machine Learning in PET: From Photon Detection to Quantitative Image Reconstruction",
abstract = "Machine learning has found unique applications in nuclear medicine from photon detection to quantitative image reconstruction. Although there have been impressive strides in detector development for time-of-flight positron emission tomography (PET), most detectors still make use of simple signal processing methods to extract the time and position information from the detector signals. Now, with the availability of fast waveform digitizers, machine learning techniques have been applied to estimate the position and arrival time of high-energy photons. In quantitative image reconstruction, machine learning has been used to estimate various corrections factors, including scattered events and attenuation images, as well as to reduce statistical noise in reconstructed images. Here, machine learning either provides a faster alternative to an existing time-consuming computation, such as in the case of scatter estimation, or creates a data-driven approach to map an implicitly defined function, such as in the case of estimating the attenuation map for PET/MR scans. In this article, we will review the above-mentioned applications of machine learning in nuclear medicine.",
keywords = "Attenuation correction, deep learning, denoising, image reconstruction, machine learning, positron emission tomography (PET), scatter correction, timing resolution",
author = "Kuang Gong and Eric Berg and Cherry, {Simon R.} and Jinyi Qi",
year = "2020",
month = "1",
doi = "10.1109/JPROC.2019.2936809",
language = "English (US)",
volume = "108",
pages = "51--68",
journal = "Proceedings of the IEEE",
issn = "0018-9219",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - Machine Learning in PET

T2 - From Photon Detection to Quantitative Image Reconstruction

AU - Gong, Kuang

AU - Berg, Eric

AU - Cherry, Simon R.

AU - Qi, Jinyi

PY - 2020/1

Y1 - 2020/1

N2 - Machine learning has found unique applications in nuclear medicine from photon detection to quantitative image reconstruction. Although there have been impressive strides in detector development for time-of-flight positron emission tomography (PET), most detectors still make use of simple signal processing methods to extract the time and position information from the detector signals. Now, with the availability of fast waveform digitizers, machine learning techniques have been applied to estimate the position and arrival time of high-energy photons. In quantitative image reconstruction, machine learning has been used to estimate various corrections factors, including scattered events and attenuation images, as well as to reduce statistical noise in reconstructed images. Here, machine learning either provides a faster alternative to an existing time-consuming computation, such as in the case of scatter estimation, or creates a data-driven approach to map an implicitly defined function, such as in the case of estimating the attenuation map for PET/MR scans. In this article, we will review the above-mentioned applications of machine learning in nuclear medicine.

AB - Machine learning has found unique applications in nuclear medicine from photon detection to quantitative image reconstruction. Although there have been impressive strides in detector development for time-of-flight positron emission tomography (PET), most detectors still make use of simple signal processing methods to extract the time and position information from the detector signals. Now, with the availability of fast waveform digitizers, machine learning techniques have been applied to estimate the position and arrival time of high-energy photons. In quantitative image reconstruction, machine learning has been used to estimate various corrections factors, including scattered events and attenuation images, as well as to reduce statistical noise in reconstructed images. Here, machine learning either provides a faster alternative to an existing time-consuming computation, such as in the case of scatter estimation, or creates a data-driven approach to map an implicitly defined function, such as in the case of estimating the attenuation map for PET/MR scans. In this article, we will review the above-mentioned applications of machine learning in nuclear medicine.

KW - Attenuation correction

KW - deep learning

KW - denoising

KW - image reconstruction

KW - machine learning

KW - positron emission tomography (PET)

KW - scatter correction

KW - timing resolution

UR - http://www.scopus.com/inward/record.url?scp=85077808448&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077808448&partnerID=8YFLogxK

U2 - 10.1109/JPROC.2019.2936809

DO - 10.1109/JPROC.2019.2936809

M3 - Article

AN - SCOPUS:85077808448

VL - 108

SP - 51

EP - 68

JO - Proceedings of the IEEE

JF - Proceedings of the IEEE

SN - 0018-9219

IS - 1

M1 - 8844693

ER -