Computed radiography (CR) uses storage phosphor imaging plates for digital imaging. Absorbed X-ray energy is stored in crystal defects. In read-out the energy is set free as blue photons upon optical stimulation. In the 35 years of CR history, several storage phosphor families were investigated and developed. An explanation is given as to why some materials made it to the commercial stage, while others did not. The photo stimulated luminescence mechanism of the current commercial storage phosphors, BaFBr:Eu2+ and CsBr:Eu2+ is discussed. The relation between storage phosphor plate physical characteristics and image quality is explained. It is demonstrated that the morphology of the phosphor crystals in the CR imaging plate has a very significant impact on its performance.

The modulation transfer function (MTF) describes the spatial resolution properties of imaging systems. In this work, the accuracy of our implementation of the edge method for calculating the presampled MTF was examined. Synthetic edge images with known MTF were used as gold standards for determining the robustness of the edge method. These images simulated realistic data from clinical digital mammography systems, and contained intrinsic system factors that could affect the MTF accuracy, such as noise, scatter, and flat-field nonuniformities. Our algorithm is not influenced by detector dose variations for MTF accuracy up to 1/2 the sampling frequency. We investigated several methods for noise reduction, including truncating the supersampled line spread function (LSF), windowing the LSF, applying a local exponential fit to the LSF, and applying a monotonic constraint to the supersampled edge spread function. Only the monotonic constraint did not introduce a systematic error; the other methods could result in MTF underestimation. Overall, our edge method consistently computed MTFs which were in good agreement with the true MTF. The edge method was then applied to images from a commercial storage-phosphor based digital mammography system. The calculated MTF was affected by the size (sides of 2.5, 5, or 10 cm) and the composition (lead or tungsten) of the edge device. However, the effects on the MTF were observed only with regard to the low frequency drop (LFD). Scatter nonuniformity was dependent on edge size, and could lead to slight underestimation of LFD. Nevertheless, this negative effect could be minimized by using an edge of 5 cm or larger. An edge composed of lead is susceptible to L-fluorescence, which causes overestimation of the LFD. The results of this work are intended to underline the need for clear guidelines if the MTF is to be given a more crucial role in acceptance tests and routine assessment of digital mammography systems: the MTF algorithm and edge object test tool need to be publicly validated.

The modulation transfer function (MTF) is well established as a metric to characterize the resolution performance of a digital radiographic system. Implemented by various laboratories, the edge technique is currently the most widespread approach to measure the MTF. However, there can be differences in the results attributed to differences in the analysis technique employed. The objective of this study was to determine whether comparable results can be obtained from different algorithms processing identical images representative of those of current digital radiographic systems. Five laboratories participated in a round-robin evaluation of six different algorithms including one prescribed in the International Electrotechnical Commission (IEC) 62220-1 standard. The algorithms were applied to two synthetic and 12 real edge images from different digital radiographic systems including CR, and direct- and indirect-conversion detector systems. The results were analysed in terms of variability as well as accuracy of the resulting presampled MTFs. The results indicated that differences between the individual MTFs and the mean MTF were largely below 0.02. In the case of the two simulated edge images, all algorithms yielded similar results within 0.01 of the expected true MTF. The findings indicated that all algorithms tested in this round-robin evaluation, including the IEC-prescribed algorithm, were suitable for accurate MTF determination from edge images, provided the images are not excessively noisy. The agreement of the MTF results was judged sufficient for the measurement of the MTF necessary for the determination of the DQE.

No abstract

Fast electronic processes taking place immediately after excitation of sensitized silver halide emulsions were studied with picosecond time-correlated single-photon counting (SPC) and femtosecond transient absorption (TA). The fluorescence decays of the J aggregate obtained by SPC measurements were also analyzed as a sum of three exponentials. For the J aggregates of a benchmark dye, a dimethyl-μ-ethyl-thiacarbocyanine, a component with a 20 ps decay time, which was absent in samples containing no AgBr crystals, was attributed to J aggregates of the sensitizer adsorbed on the AgBr microcrystals. The fluorescence decay results were compared with photographic quantum yields in an attempt to show the relevance of the spectroscopic measurements. The second component (150 ps) is a combination of fluorescence decays of the unaggregated sensitizer molecules not adsorbed on the AgBr microcrystals and monomers of the dye on AgBr surface. The fluorescence decay was found to be independent of the crystal coverage. Prolonged illumination bleached the fluorescence of the fast component, which is consistent with depletion of the J aggregates by electron injection. The decay time of this fast component shows strong dependence on both the structure of the dye and the composition of the silver halide. Excitation in the narrow absorption band of the J aggregates with a femtosecond pump pulse resulted in ground state bleaching followed by a recovery that was analyzed as a sum of three exponentials. The fast (170 fs and 2 ps) components of the recovery are related to the relaxation of bi- or multiexciton states and to singlet-singlet annihilation processes. The remaining component (1 ns) can be attributed to the radiative relaxation of dye molecules not adsorbed on the silver halide crystals. The observed bleaching recovery demonstrated a complex dependence on both the excitation flux and the total illumination dose.