Ideal versus human observer for long-tailed point spread functions: Does deconvolution help?

J. P. Rolland, Harrison H Barrett, G. W. Seeley

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

The ideal observer represents a Bayesian approach to performing detection tasks. Since such tasks are frequently used as prototype tasks for radiological imaging systems, the detectability measured at the output of an ideal detector can be used as a figure of merit to characterize the imaging system. For the detectability achieved by the ideal observer to be a good figure of merit, it should predict the ability of the human observer to perform the same detection task. Of great general interest, especially to the medical community, are imaging devices with long-tailed point spread functions (PSFs). Such PSFs may occur due to septal penetration in collimators, veiling glare in image intensifiers or scattered radiation in the body. The authors have investigated the effect that this type of PSF has on human visual signal detection and whether any improvement in performance can be gained by deconvolving the tails of the PSF. For the ideal observer, it is straightforward to show that the performance is independent of any linear, invertible deconvolution filter. The authors' psychophysical studies show, however, that performance of the human observer is indeed improved by deconvolution. The ideal observer is, therefore, not a good predictor of human observer performance for detection of a signal imaged through a long-tailed PSF.

Original languageEnglish (US)
Article number005
Pages (from-to)1091-1109
Number of pages19
JournalPhysics in Medicine and Biology
Volume36
Issue number8
DOIs
StatePublished - 1991

Fingerprint

Optical transfer function
Deconvolution
point spread functions
figure of merit
Imaging systems
Glare
visual signals
Bayes Theorem
glare
Diagnostic Imaging
image intensifiers
signal detection
Signal detection
Tail
collimators
Radiation
Equipment and Supplies
penetration
prototypes
Detectors

ASJC Scopus subject areas

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Physics and Astronomy (miscellaneous)
  • Biomedical Engineering

Cite this

Ideal versus human observer for long-tailed point spread functions : Does deconvolution help? / Rolland, J. P.; Barrett, Harrison H; Seeley, G. W.

In: Physics in Medicine and Biology, Vol. 36, No. 8, 005, 1991, p. 1091-1109.

Research output: Contribution to journalArticle

@article{8b6cc7d87ff94107bd8ecb77b67c94bf,
title = "Ideal versus human observer for long-tailed point spread functions: Does deconvolution help?",
abstract = "The ideal observer represents a Bayesian approach to performing detection tasks. Since such tasks are frequently used as prototype tasks for radiological imaging systems, the detectability measured at the output of an ideal detector can be used as a figure of merit to characterize the imaging system. For the detectability achieved by the ideal observer to be a good figure of merit, it should predict the ability of the human observer to perform the same detection task. Of great general interest, especially to the medical community, are imaging devices with long-tailed point spread functions (PSFs). Such PSFs may occur due to septal penetration in collimators, veiling glare in image intensifiers or scattered radiation in the body. The authors have investigated the effect that this type of PSF has on human visual signal detection and whether any improvement in performance can be gained by deconvolving the tails of the PSF. For the ideal observer, it is straightforward to show that the performance is independent of any linear, invertible deconvolution filter. The authors' psychophysical studies show, however, that performance of the human observer is indeed improved by deconvolution. The ideal observer is, therefore, not a good predictor of human observer performance for detection of a signal imaged through a long-tailed PSF.",
author = "Rolland, {J. P.} and Barrett, {Harrison H} and Seeley, {G. W.}",
year = "1991",
doi = "10.1088/0031-9155/36/8/005",
language = "English (US)",
volume = "36",
pages = "1091--1109",
journal = "Physics in Medicine and Biology",
issn = "0031-9155",
publisher = "IOP Publishing Ltd.",
number = "8",

}

TY - JOUR

T1 - Ideal versus human observer for long-tailed point spread functions

T2 - Does deconvolution help?

AU - Rolland, J. P.

AU - Barrett, Harrison H

AU - Seeley, G. W.

PY - 1991

Y1 - 1991

N2 - The ideal observer represents a Bayesian approach to performing detection tasks. Since such tasks are frequently used as prototype tasks for radiological imaging systems, the detectability measured at the output of an ideal detector can be used as a figure of merit to characterize the imaging system. For the detectability achieved by the ideal observer to be a good figure of merit, it should predict the ability of the human observer to perform the same detection task. Of great general interest, especially to the medical community, are imaging devices with long-tailed point spread functions (PSFs). Such PSFs may occur due to septal penetration in collimators, veiling glare in image intensifiers or scattered radiation in the body. The authors have investigated the effect that this type of PSF has on human visual signal detection and whether any improvement in performance can be gained by deconvolving the tails of the PSF. For the ideal observer, it is straightforward to show that the performance is independent of any linear, invertible deconvolution filter. The authors' psychophysical studies show, however, that performance of the human observer is indeed improved by deconvolution. The ideal observer is, therefore, not a good predictor of human observer performance for detection of a signal imaged through a long-tailed PSF.

AB - The ideal observer represents a Bayesian approach to performing detection tasks. Since such tasks are frequently used as prototype tasks for radiological imaging systems, the detectability measured at the output of an ideal detector can be used as a figure of merit to characterize the imaging system. For the detectability achieved by the ideal observer to be a good figure of merit, it should predict the ability of the human observer to perform the same detection task. Of great general interest, especially to the medical community, are imaging devices with long-tailed point spread functions (PSFs). Such PSFs may occur due to septal penetration in collimators, veiling glare in image intensifiers or scattered radiation in the body. The authors have investigated the effect that this type of PSF has on human visual signal detection and whether any improvement in performance can be gained by deconvolving the tails of the PSF. For the ideal observer, it is straightforward to show that the performance is independent of any linear, invertible deconvolution filter. The authors' psychophysical studies show, however, that performance of the human observer is indeed improved by deconvolution. The ideal observer is, therefore, not a good predictor of human observer performance for detection of a signal imaged through a long-tailed PSF.

UR - http://www.scopus.com/inward/record.url?scp=0025880308&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0025880308&partnerID=8YFLogxK

U2 - 10.1088/0031-9155/36/8/005

DO - 10.1088/0031-9155/36/8/005

M3 - Article

C2 - 1924544

AN - SCOPUS:0025880308

VL - 36

SP - 1091

EP - 1109

JO - Physics in Medicine and Biology

JF - Physics in Medicine and Biology

SN - 0031-9155

IS - 8

M1 - 005

ER -