Premium
Absolute measures of physical image quality: Measurement and application to radiographic magnification
Author(s) -
Sandrik John M.,
Wagner Robert F.
Publication year - 1982
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.595099
Subject(s) - detective quantum efficiency , optical transfer function , magnification , optics , image quality , spatial frequency , physics , noise (video) , mathematics , image (mathematics) , computer science , artificial intelligence
Two absolute measures of physical image quality—noise equivalent quanta (NEQ) and detective quantum efficiency (DQE)—are reviewed and calculated as functions of spatial frequency from measurements of the imaging characteristics of radiographic screen–film systems. NEQ is the effective number of information bearing quanta used to form an image, and DQE is the ratio of NEQ to the actual number of quanta used. NEQ and DQE are derived from the sensitometric characteristic, modulation transfer function (MTF), and noise power spectrum (NPS) of a screen–film system. Spatial frequency ( f ) distributions of NEQ and DQE were determined for three calcium tungstate screens—Du Pont Cronex Detail, Par Speed, and Hi‐Plus—used with Kodak XRP film. The relative ranking of NEQ( f ) of these three systems followed the ranking of MTF and the exposure required to produce a given film density (Detail>Par Speed>Hi Plus). The ranking of DQE( f ) followed that of system speed—Hi Plus>Par Speed>Detail—at frequencies <2.5 cycles/mm, but was the same as the ranking of NEQ( f ) at frequencies >5 cycles/mm. Noise equivalent apertures defined in terms of MTF were larger than analogous apertures defined in terms of NEQ for these screen–film systems, suggesting that, for a specific imaging task, these systems introduce more degradation in image quality due to loss of contrast than to loss of signal‐to‐noise ratio (SNR). Finally, the application of NEQ( f ) and DQE( f ) distributions to imaging system analysis is demonstrated by applying them to an optimization of magnification radiography with respect to image SNR.