Effect of noise correlation in channels of a digital X-ray detector—converter on evaluation of detection quantum efficiency

Porosev, V.V.; Зеликман, М. И.; Blinov, N. N.

doi:10.1007/s10527-005-0005-3

Cited by 2 publications

(4 citation statements)

References 2 publications

(5 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Correlation of output detector sig nals corrupts variance of output noise provided that special measures were not observed. It was suggested in [4] to divide open field image of interest into subdomains, deter mine mean brightness in subdomains, and calculate vari ance of mean brightness in subdomains. Mean brightness in subdomains is used to calculate system output noise variance.…”

Section: Estimate Of Scattering In Detector Channelsmentioning

confidence: 99%

“…Mean brightness in subdomains is used to calculate system output noise variance. The subdomains were suggested in [4] to be reduced to 16 elements (subdomain size, 4 × 4 pixels). However, further research into X ray digital detectors revealed that subdomain size of 4 × 4 pixels was insuffi cient (if noise correlation coefficient in neighboring chan nels is >0.2).…”

Section: Estimate Of Scattering In Detector Channelsmentioning

confidence: 99%

See 1 more Smart Citation

Monitoring of Quantum Efficiency of Digital X-Ray Detectors Based on Photostimulated Luminophores

Зеликман¹,

Kabanov²,

Kruchinin³

et al. 2008

Biomed Eng

Self Cite

View full text Add to dashboard Cite

Complexes based on photostimulated luminophores (CR systems) are now widely used in medical organiza tions of the Russian Federation as digital X ray systems. CR systems are rather simple and provide relatively inex pensive transition from X ray film diagnosis to digital technology: the entire X ray apparatus need not be replaced, but only reequipped with X ray cassettes of standard typical size, sampling and reading unit for mem ory screens, and a computer based automatic workplace (AWP) [1]. Imaging quality control in such systems differs from imaging quality control in digital detectors based on flat panels (amorphous silicon and amorphous selenium), CCD matrices, XRII, and scanners (solid state and gas). The goal of this work was to analyze this difference.Construction of CR systems has some specific fea tures:-transition characteristic (dependence of image brightness on detector input density of quantum flux) of some CR systems is nonlinear; -X ray detector (cassette with memory screen and sampling and reading unit for memory screens) is not controlled by the X ray radiation source. Combined cali bration of X ray detector and X ray radiation source is difficult to be implemented in practice;-interaction of IR laser beam with memory screen (during information reading) induces optical signal, scat tering resulting in signal correlation at detector output channels. Experiments demonstrated that this correlation increased upon increasing the spatial resolution of the detector.These specific features have an effect on estimation of general characteristic of image quality in a digital X ray detector (detection quantum efficiency at zero spatial fre quency DQE and Nyquist spatial frequency DQE(u,v)). Estimate of Transition Characteristic NonlinearityNonlinear signal transformation in detector chan nels should be avoided during open field image formation and detection for DQE and DQE(u,v) estimation [1 3]. It was noted above that the transition characteristic of some CR systems was nonlinear at detector output chan nels (logarithmic transition characteristics of channel amplifiers). In this case, data linearization is required. In other words, the data should be transformed to the initial form (i.e., the form prior to nonlinear transformation).Let signals before X ini and after X res nonlinear trans formation be correlated using the following expression:whereThe linearization procedure starts with dependence of mean brightness of open field image on detector input signal amplitude. Open field images detected at different detector input plane dose are used (standard [2] requires clinical use of the apparatus at dose range from 0 to four fold mean dose level). Discrete dose variation step should provide logarithm of variation range 0.1. This dependence is approximated by the following function X res = a + b ln(X ini ).Equation (1) can be recast as the sought function:Equation (2) can be used for transformation of initial data for detector plane dose providing the following cal culations of DQE and DQE(u,v). It should ...

show abstract

Section: Estimate Of Scattering In Detector Channelsmentioning

confidence: 99%

Section: Estimate Of Scattering In Detector Channelsmentioning

confidence: 99%

Monitoring of Quantum Efficiency of Digital X-Ray Detectors Based on Photostimulated Luminophores

Зеликман¹,

Kabanov²,

Kruchinin³

et al. 2008

Biomed Eng

Self Cite

View full text Add to dashboard Cite

show abstract

“…Substituting the obtained value of N into Eq. (2) we obtain a generalized formula for the signal/noise ratio for an ideal system of X ray detection: (4) Taking into account that the detective sensitivity is usually determined from the exposure dose D in the input plane, Eq. (4) can be rewritten as follows: D = 7.3⋅10 −9 ⋅(signal/noise) 2 ⋅E⋅µ a ⋅(pairs of lines/mm) 2 /K 2 .…”

mentioning

confidence: 99%

“…If this value cannot be attained at anode voltage of 70 kV (for example, if an automatic brightness stabilization system is used), the actual value of anode voltage and current and the thick ness of intrinsic and additional filters should be recorded in the protocol. 4. The resolution test object should be placed at the site of the patient.…”

mentioning

confidence: 99%

Determination of the Detective Efficiency of X-Ray Diagnostic Systems by Visual Assessment of the Maximal Resolution

Vladimirov¹,

Kanter²,

Kozlov³

et al. 2005

Biomed Eng

View full text Add to dashboard Cite

Determination of the detective efficiency of X ray diagnostic systems is a rather difficult task. It has been widely discussed in the literature [1 3, 7]. An internation al standard unifies the method for determination of the detective efficiency of X ray detectors, in particular, matrix detectors used in medical X ray diagnosis [8]. The standard can be used only under laboratory conditions and requires the use of special expensive equipment, which makes it virtually impossible to apply this standard during acceptance testing, as well as in maintenance test ing of X ray apparatuses at the site of their use. A rela tively simple method for determination of detective effi ciency of X ray diagnostic systems from visually evaluat ed minimal radiation contrast was suggested in our previ ous work [2]. This method is based on the use of a con trast sensitivity test object and an X ray dosimeter provid ing measurement within the required sensitivity range. In [2], the detective efficiency was determined only at quasi zero frequency. The goal of this work was to extend the range of applicability of this method to virtually all spatial frequencies used in X ray diagnosis without eliminating such advantages of the method as simplicity and possibil ity of use for acceptance, periodical, and maintenance tests. The improved method is based on the use of a reso lution test object and an X ray dosimeter installed at the input of the X ray receiver to measure the exposure dose.The signal/noise ratio is the main characteristic determining the efficiency of detection. It can be described by the following equation:where (1 − exp(−µd)) = K is the radiation contrast pro vided by a lead test object with given thickness d and extinction coefficient µ for given effective energy of X ray radiation. Anode voltage should be set to 70 kV, and a 20 mm Al filter should be placed at the X ray collimator output. This corresponds to half attenuation Al layer thickness 7 mm and effective energy of 50 keV. For this energy, the mass coefficient µ/ρ of radiation extinction in lead is 7.94 cm 2 /g [7]. Lead density ρ Pb is 11.3 g/cm 3 , so that the linear extinction coefficient µ is 89.7 cm -1 . Taking into account that the stripes of resolution test objects used for testing of medical X ray equipment are usually 50, 70, or 100 µm thick, the theoretically calculated value of radi ation contrast for effective energy of 50 keV is 0.36 (36%), 0.47 (47%), or 0.59 (59%), respectively.The frequency-contrast characteristic (FCC) of an X ray diagnostic system is determined by its focus size, geometric characteristics of the test procedure, distance between focus and resolution test object, and distance between resolution test object and input plane of the receiver, as well as the parameters of the X ray detection and imaging system. In Eq. (1), FCC characterizes the rate of decrease in radiation contrast with increasing fre quency of lead stripes in the test object. For an ideal sys tem, FCC = 1 for all spatial frequencies. N (quantum/ mm 2 ) is the is the nu...

show abstract

Effect of noise correlation in channels of a digital X-ray detector—converter on evaluation of detection quantum efficiency

Cited by 2 publications

References 2 publications

Monitoring of Quantum Efficiency of Digital X-Ray Detectors Based on Photostimulated Luminophores

Monitoring of Quantum Efficiency of Digital X-Ray Detectors Based on Photostimulated Luminophores

Determination of the Detective Efficiency of X-Ray Diagnostic Systems by Visual Assessment of the Maximal Resolution

Contact Info

Product

Resources

About