Maximum-lesion-detectability reconstruction using penalized maximum likelihood

Jinyi Qi, Ronald H. Huesman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Statistical reconstruction methods based on the penalized maximum-likelihood (or maximum a posteriori) principle have been developed to improve the image quality in emission tomography. However, the penalty functions (or image priors) are often heuristically designed to encourage smoothness in reconstructed images. This paper presents a method to design the penalty function to improve lesion detection. We factorize the penalty function into pairwise potentials. Using the theoretical expressions that we have developed for evaluating lesion detectability, we are able to find the optimum weights of the pairwise potentials to achieve the maximum lesion detectability based on the numerical observers that are correlated to human performance.

Original languageEnglish (US)
Title of host publication2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano
Pages348-351
Number of pages4
Volume1
StatePublished - 2004
Externally publishedYes
Event2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano - Arlington, VA, United States
Duration: Apr 15 2004Apr 18 2004

Other

Other2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano
CountryUnited States
CityArlington, VA
Period4/15/044/18/04

Fingerprint

Maximum likelihood
Image quality
Tomography

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Qi, J., & Huesman, R. H. (2004). Maximum-lesion-detectability reconstruction using penalized maximum likelihood. In 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano (Vol. 1, pp. 348-351)

Maximum-lesion-detectability reconstruction using penalized maximum likelihood. / Qi, Jinyi; Huesman, Ronald H.

2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano. Vol. 1 2004. p. 348-351.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Qi, J & Huesman, RH 2004, Maximum-lesion-detectability reconstruction using penalized maximum likelihood. in 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano. vol. 1, pp. 348-351, 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano, Arlington, VA, United States, 4/15/04.
Qi J, Huesman RH. Maximum-lesion-detectability reconstruction using penalized maximum likelihood. In 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano. Vol. 1. 2004. p. 348-351
Qi, Jinyi ; Huesman, Ronald H. / Maximum-lesion-detectability reconstruction using penalized maximum likelihood. 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano. Vol. 1 2004. pp. 348-351
@inproceedings{feef4eeea8dc47edb7f406097adcfffc,
title = "Maximum-lesion-detectability reconstruction using penalized maximum likelihood",
abstract = "Statistical reconstruction methods based on the penalized maximum-likelihood (or maximum a posteriori) principle have been developed to improve the image quality in emission tomography. However, the penalty functions (or image priors) are often heuristically designed to encourage smoothness in reconstructed images. This paper presents a method to design the penalty function to improve lesion detection. We factorize the penalty function into pairwise potentials. Using the theoretical expressions that we have developed for evaluating lesion detectability, we are able to find the optimum weights of the pairwise potentials to achieve the maximum lesion detectability based on the numerical observers that are correlated to human performance.",
author = "Jinyi Qi and Huesman, {Ronald H.}",
year = "2004",
language = "English (US)",
isbn = "0780383885",
volume = "1",
pages = "348--351",
booktitle = "2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano",

}

TY - GEN

T1 - Maximum-lesion-detectability reconstruction using penalized maximum likelihood

AU - Qi, Jinyi

AU - Huesman, Ronald H.

PY - 2004

Y1 - 2004

N2 - Statistical reconstruction methods based on the penalized maximum-likelihood (or maximum a posteriori) principle have been developed to improve the image quality in emission tomography. However, the penalty functions (or image priors) are often heuristically designed to encourage smoothness in reconstructed images. This paper presents a method to design the penalty function to improve lesion detection. We factorize the penalty function into pairwise potentials. Using the theoretical expressions that we have developed for evaluating lesion detectability, we are able to find the optimum weights of the pairwise potentials to achieve the maximum lesion detectability based on the numerical observers that are correlated to human performance.

AB - Statistical reconstruction methods based on the penalized maximum-likelihood (or maximum a posteriori) principle have been developed to improve the image quality in emission tomography. However, the penalty functions (or image priors) are often heuristically designed to encourage smoothness in reconstructed images. This paper presents a method to design the penalty function to improve lesion detection. We factorize the penalty function into pairwise potentials. Using the theoretical expressions that we have developed for evaluating lesion detectability, we are able to find the optimum weights of the pairwise potentials to achieve the maximum lesion detectability based on the numerical observers that are correlated to human performance.

UR - http://www.scopus.com/inward/record.url?scp=17144368446&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=17144368446&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:17144368446

SN - 0780383885

VL - 1

SP - 348

EP - 351

BT - 2004 2nd IEEE International Symposium on Biomedical Imaging: Macro to Nano

ER -