Comparing lesion detection performance for PET image reconstruction algorithms: A case study

M. T. Chan, R. M. Leahy, E. U. Mumcuoglu, Simon R Cherry, J. Czernin, A. Chatziioannou

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

We present a comparison of the performance of filtered backprojection (FBP) and maximum a posteriori (MAP) reconstruction of PET images for the task of hot lesion detection. The comparison is performed on data generated by combining FDG chest scans of normal patients (i.e., without lesions) with pseudo-Poisson "lesion" data generated from appropriately scaled sinograms collected using a separately scanned 1.25 cm3 spherical source. Scaling factors were used to achieve approximately 2.5:1 lesion-to-background activity ratios. A total of 60 "abnormal" cases were generated from their normal counterparts. A 3D non-prewhitening (NPW) observer model based on a matched filter was used to test for the presence of the lesion in the vicinity of the known lesion location. ROC curves were generated for several choices of cut-off frequency for ramp-filtered FBP and the smoothing parameter for the MAP reconstructions. The NPW detector was matched to the algorithm and smoothing parameter in each case. Our experiments show that MAP reconstruction over a range of smoothing parameter values results in statistically significant improvements, compared to FBP, for the task of lesion detection using a NPW observer.

Original languageEnglish (US)
Pages (from-to)1558-1562
Number of pages5
JournalIEEE Transactions on Nuclear Science
Volume44
Issue number4 PART 1
DOIs
StatePublished - 1997
Externally publishedYes

Fingerprint

Matched filters
Cutoff frequency
image reconstruction
Image reconstruction
lesions
Detectors
smoothing
Experiments
matched filters
chest
ramps
cut-off
scaling
detectors
curves

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Nuclear Energy and Engineering

Cite this

Comparing lesion detection performance for PET image reconstruction algorithms : A case study. / Chan, M. T.; Leahy, R. M.; Mumcuoglu, E. U.; Cherry, Simon R; Czernin, J.; Chatziioannou, A.

In: IEEE Transactions on Nuclear Science, Vol. 44, No. 4 PART 1, 1997, p. 1558-1562.

Research output: Contribution to journalArticle

Chan, MT, Leahy, RM, Mumcuoglu, EU, Cherry, SR, Czernin, J & Chatziioannou, A 1997, 'Comparing lesion detection performance for PET image reconstruction algorithms: A case study', IEEE Transactions on Nuclear Science, vol. 44, no. 4 PART 1, pp. 1558-1562. https://doi.org/10.1109/23.632731
Chan, M. T. ; Leahy, R. M. ; Mumcuoglu, E. U. ; Cherry, Simon R ; Czernin, J. ; Chatziioannou, A. / Comparing lesion detection performance for PET image reconstruction algorithms : A case study. In: IEEE Transactions on Nuclear Science. 1997 ; Vol. 44, No. 4 PART 1. pp. 1558-1562.
@article{e6d5a06fa5ea455cbfbd92e14cfc6cdc,
title = "Comparing lesion detection performance for PET image reconstruction algorithms: A case study",
abstract = "We present a comparison of the performance of filtered backprojection (FBP) and maximum a posteriori (MAP) reconstruction of PET images for the task of hot lesion detection. The comparison is performed on data generated by combining FDG chest scans of normal patients (i.e., without lesions) with pseudo-Poisson {"}lesion{"} data generated from appropriately scaled sinograms collected using a separately scanned 1.25 cm3 spherical source. Scaling factors were used to achieve approximately 2.5:1 lesion-to-background activity ratios. A total of 60 {"}abnormal{"} cases were generated from their normal counterparts. A 3D non-prewhitening (NPW) observer model based on a matched filter was used to test for the presence of the lesion in the vicinity of the known lesion location. ROC curves were generated for several choices of cut-off frequency for ramp-filtered FBP and the smoothing parameter for the MAP reconstructions. The NPW detector was matched to the algorithm and smoothing parameter in each case. Our experiments show that MAP reconstruction over a range of smoothing parameter values results in statistically significant improvements, compared to FBP, for the task of lesion detection using a NPW observer.",
author = "Chan, {M. T.} and Leahy, {R. M.} and Mumcuoglu, {E. U.} and Cherry, {Simon R} and J. Czernin and A. Chatziioannou",
year = "1997",
doi = "10.1109/23.632731",
language = "English (US)",
volume = "44",
pages = "1558--1562",
journal = "IEEE Transactions on Nuclear Science",
issn = "0018-9499",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4 PART 1",

}

TY - JOUR

T1 - Comparing lesion detection performance for PET image reconstruction algorithms

T2 - A case study

AU - Chan, M. T.

AU - Leahy, R. M.

AU - Mumcuoglu, E. U.

AU - Cherry, Simon R

AU - Czernin, J.

AU - Chatziioannou, A.

PY - 1997

Y1 - 1997

N2 - We present a comparison of the performance of filtered backprojection (FBP) and maximum a posteriori (MAP) reconstruction of PET images for the task of hot lesion detection. The comparison is performed on data generated by combining FDG chest scans of normal patients (i.e., without lesions) with pseudo-Poisson "lesion" data generated from appropriately scaled sinograms collected using a separately scanned 1.25 cm3 spherical source. Scaling factors were used to achieve approximately 2.5:1 lesion-to-background activity ratios. A total of 60 "abnormal" cases were generated from their normal counterparts. A 3D non-prewhitening (NPW) observer model based on a matched filter was used to test for the presence of the lesion in the vicinity of the known lesion location. ROC curves were generated for several choices of cut-off frequency for ramp-filtered FBP and the smoothing parameter for the MAP reconstructions. The NPW detector was matched to the algorithm and smoothing parameter in each case. Our experiments show that MAP reconstruction over a range of smoothing parameter values results in statistically significant improvements, compared to FBP, for the task of lesion detection using a NPW observer.

AB - We present a comparison of the performance of filtered backprojection (FBP) and maximum a posteriori (MAP) reconstruction of PET images for the task of hot lesion detection. The comparison is performed on data generated by combining FDG chest scans of normal patients (i.e., without lesions) with pseudo-Poisson "lesion" data generated from appropriately scaled sinograms collected using a separately scanned 1.25 cm3 spherical source. Scaling factors were used to achieve approximately 2.5:1 lesion-to-background activity ratios. A total of 60 "abnormal" cases were generated from their normal counterparts. A 3D non-prewhitening (NPW) observer model based on a matched filter was used to test for the presence of the lesion in the vicinity of the known lesion location. ROC curves were generated for several choices of cut-off frequency for ramp-filtered FBP and the smoothing parameter for the MAP reconstructions. The NPW detector was matched to the algorithm and smoothing parameter in each case. Our experiments show that MAP reconstruction over a range of smoothing parameter values results in statistically significant improvements, compared to FBP, for the task of lesion detection using a NPW observer.

UR - http://www.scopus.com/inward/record.url?scp=0031208262&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031208262&partnerID=8YFLogxK

U2 - 10.1109/23.632731

DO - 10.1109/23.632731

M3 - Article

AN - SCOPUS:0031208262

VL - 44

SP - 1558

EP - 1562

JO - IEEE Transactions on Nuclear Science

JF - IEEE Transactions on Nuclear Science

SN - 0018-9499

IS - 4 PART 1

ER -