| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Medical Physics |
1 From the Department of Radiology, Digital Imaging Research Division, Duke University Medical Center, Box 3302, Room 139, Bryan Research Building, Durham, NC 27710 (all authors); and the Department of Biomedical Engineering, Duke University, Durham, NC (C.E.F., R.J.W., J.T.D., A.H.B.). Received March 9, 2000; revision requested April 26; revision received May 25; accepted June 15. Address correspondence to H.G.C. (e-mail: harrell.chotas@duke.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: The 41 x 41-cm digital FPD is constructed on a single monolithic glass substrate with a structured cesium iodide scintillator layer and an amorphous silicon thin-film transistor array for image readout. Basic imaging characteristics of the FPD and associated image processing system were assessed on acquired images, including linearity, repeatability, uniformity of response, modulation transfer function (MTF), noise power spectrum, detective quantum efficiency (DQE), contrast sensitivity, and scatter content. Results with the FPD system were compared to those with a storage phosphor computed radiography (CR) system.
RESULTS: Images obtained with the FPD demonstrated excellent uniformity, repeatability, and linearity, as well as MTF and DQE that were superior to those with the storage phosphor CR system. The contrast and scatter content of images acquired with the FPD were equivalent to those acquired with the storage phosphor system.
CONCLUSION: The FPD provides radiographic images with excellent inherent physical image quality.
Index terms: Flat-panel detector, **.112 • Radiography, digital
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
A new generation of digital flat-panel detectors (FPDs) is now emerging. The purpose of our study was to present measurements of the basic physical imaging characteristics of one of these FPD systems.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
A detailed description of the FPD is available elsewhere (3) and is shown schematically in Figure 1. The FPD is fabricated on a single monolithic glass substrate, with an active imaging area of 41 x 41 cm. Layered on the glass is an amorphous silicon, thin-film transistor array which, in turn, is overlaid with a structured cesium iodide scintillator and a protective coating. When x rays strike the scintillator, visible light is emitted and is converted to an electrical signal in photodiodes in the thin-film transistor array. Each pixel is then read out by onboard amplifiers and sampling electronics and converted to a digital value, forming a raw image.
|
|
To characterize the performance of this FPD system, we measured the following clinically relevant physical properties: linearity, uniformity and repeatability, noise power spectrum (NPS), modulation transfer function (MTF), detective quantum efficiency (DQE), contrast sensitivity and scatter, dead pixels, and memory artifacts.
Linearity
The response of the FPD system as a function of entrance dose to the FPD was measured over a range of exposures from 0.22 to 13.7 mR (0.057 x 10-6 to 3.53 x 10-6 C/kg). Imaging parameters in each case were 120 kVp, source-to-image distance of 180 cm, 0.5 mm of added copper filtration, focal spot size of 1.25 mm, and tube current of 200 mA. Exposure intensity was varied by altering the exposure time from the lowest generator setting (0.25 mAs) to a level at which the FPD system was saturated (100 mAs). The digital images were analyzed by measuring the mean digital pixel value inside a 5-cm2 circular region of interest centered on the FPD. Exposure values were measured with the ionization chamber of a calibrated exposure meter (MDH 1015; Radcal, Monrovia, Calif), with the FPD moved out of the beam to avoid backscatter. The exposure values were adjusted by the inverse square law to give measurements corresponding to the entrance surface of the FPD housing. Measurements were made both with and without the x-ray grid in place.
Uniformity and Repeatability
The calibration protocol for the FPD system includes a per-pixel compensation to correct for differences in the amplifier gains and offsets and for beam inhomogeneities (eg, heel effect, grid nonuniformities). The goal of this compensation is to produce an image of uniformly constant value when a full-area exposure is acquired with nothing in the beam. In this study, uniformity calibration was performed without the antiscatter grid in place (because many of our characterization experiments were performed without a grid), although the standard practice for the clinical device is to calibrate with the grid in place.
To evaluate the repeatability of the FPD system response to exposure, six images were acquired over a period of several days. Each exposure was made at 120 kVp, 1.0 mAs, 180-cm source-to-image distance, with manual exposure control and no grid in place. No attenuating material was placed in the beam.
As a measure of image uniformity, the spatial variation of the digital values was measured after application of routine image uniformity compensation corrections. Each image was subdivided into a 32 x 32 matrix of contiguous subregions, each 64 x 64 pixels (1.28 x 1.28 cm). The mean and SD of pixel values in each subregion were computed, providing a distribution of the subregional "intensity" and "noise" values. In addition for each image, the mean and SD of intensity values of all subregions were computed. This mean of the mean values provides the global image mean intensity. The SD of the distribution of mean intensity values for the subregions is a measure of the large-scale spatial nonuniformity of the corrected image when sampled and averaged every 1.28 cm. The variation of this value over the six images expresses the repeatability of the uniformity compensation over time.
NPS
Flat-field exposures were used to measure the NPS. Imaging parameters were 70 kVp, 0.5 mm of added copper filtration, and source-to-image distance of 180 cm. An x-ray exposure technique of 2.5 mAs provided an incident exposure of about 0.3 mR (8 x 10-8 C/kg), an exposure level comparable to that used in previous experiments with CR systems. This exposure is also approximately midway between the exposure behind the lungs and dense regions on a typical posteroanterior chest radiograph. These NPS measurements were made with the x-ray grid removed from the FPD housing (the faceplate remained in place) to allow an evaluation of the FPD alone and a comparison of measurements reported for other detector systems in which the grid was removed (such as CR).
The NPS was measured by using the technique described by Dobbins et al (1) and Dobbins (4). The image was subdivided into an 8 x 8 array of contiguous subregions, each 128 x 128 pixels. The squared Fourier amplitude was measured in each of the 64 subregions, and these measurements were averaged. The NPS in both the horizontal and vertical directions was determined from the two-dimensional NPS.
A second NPS experiment was performed to estimate the contribution to total noise power arising from spatially fixed pattern noise in the FPD system, such as from improperly corrected pixel-to-pixel variation in FPD response. In this experiment, 32 images were acquired, each with the same x-ray technique indicated previously. An averaged image was computed from the 32 images and was then subtracted from each of the 32 images. For each resultant difference image, the NPS was again estimated as the spatial average of 64 regions of interest. The mean of the NPS estimates from the 32 difference images (scaled by 32/31 to account for loss of variance in the subtraction process) provides an estimate of the NPS that excludes noise power from spatially fixed FPD noise.
MTF
The MTF was measured in both the horizontal and vertical directions by using the technique of Fujita et al (5), as modified by Dobbins et al (1). A 12-µm lead slit was placed at the entrance to the FPD faceplate (approximately 4 cm from the FPD itself). The slit was angled at 2.6° off the orientation of the pixel rows or columns and was imaged at a location free from bad-pixel rows or columns. Images were acquired with the same exposure parameters used for the NPS measurements with the exception that an exposure of 500 mAs was used to provide adequate intensity through the slit.
A finely sampled line spread function was computed from the pixel data in the image of the slit. To estimate the line spread function values at intensities less than 1% of the peak value (where experimental uncertainty was greater), the curve was extended by means of exponential extrapolation following standard practice (1,5). The presampled MTF was computed as the amplitude of the Fourier transform of the line spread function. A small correction was applied to compensate for the finite aperture of the slit. The expectation MTF, which includes an estimated contribution from aliasing, was determined from the presampled MTF (6).
DQE
The DQE as a function of spatial frequency f was determined from the following relation:
Contrast Sensitivity and Scatter
An acrylic chest phantom was constructed from uniform sheets of acrylic that were cut into geometric shapes, as described previously (8). The phantom possessed gross attenuation and scatter properties similar to those of an adult thorax (acrylic thicknesses: lung-equivalent region, 11.1 cm; mediastinum- and subdiaphragm-equivalent regions, 20.0 cm). Acrylic contrast objects (2.5-cm-diameter, 0.476-cm-thick disks) were attached to the front of the phantom overlying areas representing the lung and mediastinum. An array of lead beam stops, each approximately 2 mm thick and 2 mm in diameter, was also affixed to the beam-entrance side of the phantom to allow estimation of scatter fractions in selected regions of the detected images.
Digital radiographic images of the chest phantom were acquired with the FPD system and, for comparison purposes, with a storage phosphor CR system (PCR 7000; Philips Medical Systems, Shelton, Conn). All images were acquired with the same x-ray generation equipment with the same exposure conditions: 120 kVp, 20 mAs, manual exposure control, 180-cm source-to-image distance, with the antiscatter grid. The storage phosphor CR images were acquired with an imaging plate (ST-Vn; FujiFilm Medical Systems, Stamford, Conn) in a lead-backed cassette, with 20 minutes elapsed time between exposure and image readout. The imaging plate was processed with fixed latitude and sensitivity during image readout (latitude of 4.0 decades, sensitivity of 200 speed equivalent).
Images that had not been subjected to any image postprocessing, such as contrast adjustment or spatial frequency enhancement, were retrieved from both systems for analysis. To estimate the subject contrast at each disk location, the mean of pixel values was computed in small regions of interest just inside and just outside the radiographic shadow of each contrast disk (23-pixel-diameter circular regions of interest). Mean pixel values were transformed to exposure values by means of empirically derived calibration curves for each system. Contrast was computed as (Xout - Xin)/Xout, where Xout and Xin are the mean exposure measured just outside and just inside, respectively, the radiographic shadow of the contrast disk.
Scatter fractions in the lung- and mediastinum-equivalent regions were estimated near each contrast disk. With the assumption that all exposure measured behind a beam stop is due to scattered photons, these scatter fractions were estimated as Xin/Xout, where X was defined just inside and outside, respectively, the shadow of the lead beam stop.
The contrast values were compared with theoretically predicted values. The predicted contrast of the acrylic disk shadow in each chest-equivalent region was computed with a computer program that simulates the energy-dependent attenuation of a polyenergetic photon flux passing through a known thickness of materials with known elemental composition (7). This simulation computed the expected transmission spectra for the photons passing through the phantom and then estimated the relative absorbed energy in the two detectors, with the contrast computed as the fraction of energy absorbed.
Dead Pixels
The number of known dead (nonfunctional) pixels and pixel aggregations was obtained from the system calibration table (bad-pixel mask). Calibration algorithms detect individual pixels, pixel clusters, and lines of contiguous pixels (rows or columns) that fail to produce a usable output value. In routine clinical use, a proprietary correction algorithm is applied to each image at the site of each nonfunctional pixel to fill in the missing image information on the basis of surrounding image information.
Memory Artifacts
The FPD system was tested for the presence and persistence of image memory artifacts, a common phenomenon among digital detectors to varying degrees. This artifact, when present in a digital radiographic system, may be seen as a residual image shadow from a previous exposure (9). The artifact was evaluated by making an exposure of a high-contrast phantom, rapidly followed by a second exposure without the contrast phantom, and then estimating the contribution of the first exposure to the second image. The high-contrast image was formed by shielding half of the FPD with lead and allowing the other half to be exposed. Image intensity was compared in both halves on a subsequent image made with the lead shielding removed. The evaluation was performed at clinically relevant exposure levels comparable to those for a typical lateral image (120 kVp, 40 mAs) followed by a lower dose posteroanterior image (120 kVp, 2.5 mAs, 0.5 mm of added copper filtration).
A second experiment was performed in which a lateral image of an anthropomorphic chest phantom was acquired at 40 mAs (120 kVp). Twenty-seven seconds later, a posteroanterior image of the chest phantom was acquired at 2.5 mAs, to evaluate whether any visible residual signal was noticeable.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
Memory Artifact
The magnitude of the residual contrast due to FPD memory was 3.7% in the 2.5-mAs image acquired 27 seconds after the 40-mAs exposure with the high-contrast lead phantom. This residual contrast decayed to 1.7% and 0.8% after 5 and 30 minutes, respectively.
In the experiment with the anthropomorphic chest phantom, there was no visible evidence of the lateral image on the subsequent posteroanterior image, even with substantial contrast enhancement on the image. In a separate study involving chest radiographs of human subjects, there was no case in which a prior image was visible in a subsequent image.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
The most important result in this study was the relatively high DQE of the FPD system. The value of 66% for the DQE at zero frequency at 0.3 mR (8 x 10-8 C/kg) for the FPD system is substantially higher than the value for the storage phosphor CR system or screen-film radiography. The higher DQE will manifest as decreased visible noise on FPD images compared with storage phosphor CR images acquired at the same exposure levels and displayed with the same image contrast. For the test case of a simple object in a uniform background, the higher DQE would predict an improvement in the ability of a human observer to detect the object. Alternatively, the same level of image noise and, hence, detection performance could be obtained at a lower patient exposure. The DQE in this study was measured at 70 kVp to allow comparison with previous measurements at storage phosphor CR and screen-film radiography. We expect that the DQE of FPD and storage phosphor CR will both decrease at the higher tube potentials typical of chest radiography (eg, 120 kVp). We will evaluate the DQE measurements in this study with other tube potentials and exposure levels.
A slight increase was noted in the NPS below 0.2 cycles per millimeter for the single-image as compared with the 32-image measurement with spatially fixed pattern noise removed. This increase at low frequencies is likely the result of spatial nonuniformity in the x-ray beam that was not fully compensated by the calibration in our prototype system. Our prototype system allowed multidirectional translation of the x-ray tube, which may have resulted in a slight misalignment with the position of the tube during the calibration procedure. Our understanding is that the commercial implementation of the system does not allow such independent tube movement. Also in our prototype system, the gain correction was acquired at 120 kVp, although the DQE measurement was performed at 70 kVp. Some potential low-frequency contribution to NPS could result from this difference in tube potential. The commercial version of this system makes a correction for tube potential when it makes the gain correction. For all these reasons, there may be substantially less low-frequency residual NPS in the corrected images acquired with the commercial system, although further experiments are required to confirm these conjectures.
The NPS and DQE measurements in this study demonstrate the FPD performance without the x-ray antiscatter grid in place. Measurements of the NPS with the grid in place should reveal an overall degradation in the NPS at all spatial frequencies owing to the 40% loss of primary x-ray flux resulting from the grid, as well as a corresponding degradation of the DQE by 40% at all frequencies. We chose not to report results with the grid, even though a grid is used in the standard configuration of the FPD system, because DQE with most other digital devices has been reported without a grid. The loss in DQE with a grid would be true with any detector, film or digital, although the loss of inherent DQE might be offset by the substantially improved scatter properties of actual clinical images when a grid is used.
The measurement of inherent contrast sensitivity showed little difference between the FPD and storage phosphor CR systems. The contrast that was measured was the inherent contrast of an acrylic object and reflects the dependence of contrast on the energy response of the detectors. Of course, the contrast of a digital image as viewed by an observer may be arbitrarily varied, so a comparison of the inherent contrast sensitivity to image noise is often performed. We did not perform a separate contrast-to-noise ratio experiment, but because the inherent contrast of the two detectors was comparable, and because the NPS of the FPD system was far superior to that of the storage phosphor system, one may conclude that the contrast-to-noise ratio of the former should also be superior to that of the latter.
There are several differences between the configuration of the prototype FPD system evaluated in this study and that of the commercially available FPD system, although none of these differences are expected to substantially affect the measurements. First, there is a slight difference in grid specifications. Grid spacing was 70 lines per centimeter for the system evaluated in this study, but it is 78 lines per centimeter for the commercially available system. Second, the DQE was measured at 70 kVp, whereas the uniformity correction map was obtained at 120 kVp (the tube potential at which other measurements were performed). The prototype FPD system as evaluated in this study did not change the correction map as the energy changed. Energy-specific uniformity correction is provided by the commercially available system, however, and its DQE may vary slightly from the values in this study. Third, the algorithms to correct for the effects due to dead pixels differ slightly. It is difficult to say what effect a different bad-pixel correction procedure would have on the measurements we report herein, although we would expect any such differences to be negligibly small.
With so few FPD systems in general clinical use, no consensus has yet formed regarding an acceptable fraction and distribution of dead pixels. In dead-pixel measurements in this study, an outer border of 13 pixels was excluded; this is also true for the commercially available system, and its images have an active area of 2,022 x 2,022 pixels. A triangular region of 968 pixels in each corner was excluded from the analysis of bad pixels with the prototype system, because the manufacturer originally intended to exclude these regions from clinical images. Subsequently, the regions were not excluded in the commercially available system. In considering the effect of bad pixels, it should be noted that the FPD system evaluated in this study was produced as a single panel. In FPD systems being developed by other manufacturers, two or four separate detector tiles are used, and they must be physically joined and stitched together algorithmically to form the composite image. Thus, in addition to any individual bad pixels, there is the potential for as many as 2,000 or more absent or degraded pixels in the multiple-tile FPDs as a result of joining and stitching.
We did not perform a detailed comparison of uniformity with the FPD and storage phosphor CR systems. Isolation of spatial nonuniformity of the x-ray beam from other sources of nonuniformity is not practical with the storage phosphor system; thus, the earlier measurements (10) are not directly comparable. As a rough comparison, however, overall variation with the FPD system was about 2% compared with the previously reported 6% for the storage phosphor system.
To our knowledge, ours is the first independent study of the physical imaging characteristics of this FPD, although previous studies have evaluated this FPD with smaller format for mammography (11) and other prototypes of the larger FPD evaluated herein (3). The FPD used in our study meets all the manufacturing production specifications; therefore, our results are indicative of the imaging performance that can be expected from installed clinical systems.
Physical measurements with this FPD system have been encouraging, but the true importance of the technology lies in its ability to produce diagnostically useful images. We are currently evaluating the system in the clinical setting.
The results of this study indicate that digital radiographic images obtained with this FPD have excellent uniformity, repeatability, and linearity, as well as MTF and DQE that are superior to those obtained with a storage phosphor CR system. The contrast and scatter of acquired images are equivalent to those of CR. These improvements come in addition to the general advantages of digital imaging, including adjustable image appearance, image transmission, and digital storage.
|
FOOTNOTES |
|---|
Abbreviations: CR = computed radiography, DQE = detective quantum efficiency, FPD = flat-panel detector, MTF = modulation transfer function, NPS = noise power spectrum
Author contributions: Guarantors of integrity of entire study, C.E.F., J.T.D.; study concepts and design, all authors; definition of intellectual content, all authors; literature research, all authors; clinical studies, C.E.R.; experimental studies, all authors; data acquisition and analysis, all authors; statistical analysis, all authors; manuscript preparation, editing, and review, all authors; manuscript final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Wirth, M. Treitl, M. F. Reiser, and M. Korner Imaging Performance with Different Doses in Skeletal Radiography: Comparison of a Needle-structured and a Conventional Storage Phosphor System with a Flat-Panel Detector Radiology, November 10, 2008; (2008) 2493080640. [Abstract] [Full Text]
|
|
|
|
 |
|
 C S MOORE, A W BEAVIS, and J R SAUNDERSON Investigation of optimum X-ray beam tube voltage and filtration for chest radiography with a computed radiography system Br. J. Radiol., October 1, 2008; 81(970): 771 - 777. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. T. Ranger, E. Samei, J. T. Dobbins III, and C. E. Ravin Assessment of Detective Quantum Efficiency: Intercomparison of a Recently Introduced International Standard with Prior Methods Radiology, June 1, 2007; 243(3): 785 - 795. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Korner, C. H. Weber, S. Wirth, K.-J. Pfeifer, M. F. Reiser, and M. Treitl Advances in Digital Radiography: Physical Principles and System Overview RadioGraphics, May 1, 2007; 27(3): 675 - 686. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Hatakeyama, S. Kakeda, N. Ohnari, J. Moriya, N. Oda, K. Nishino, W. Miyamoto, and Y. Korogi Reduction of Radiation Dose for Cerebral Angiography Using Flat Panel Detector of Direct Conversion Type: A Vascular Phantom Study AJNR Am. J. Neuroradiol., April 1, 2007; 28(4): 645 - 650. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. P. Busch and K. Faulkner Image quality and dose management in digital radiography: a new paradigm for optimisation Radiat Prot Dosimetry, December 1, 2005; 117(1-3): 143 - 147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Rapp-Bernhardt, T. M. Bernhardt, H. Lenzen, R. Esseling, F. W. Roehl, M. Schiborr, I. Theobald-Hormann, and W. Heindel Experimental Evaluation of a Portable Indirect Flat-Panel Detector for the Pediatric Chest: Comparison with Storage Phosphor Radiography at Different Exposures by Using a Chest Phantom Radiology, November 1, 2005; 237(2): 485 - 491. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. W. Hamer, C. B. Sirlin, M. Strotzer, I. Borisch, N. Zorger, S. Feuerbach, and M. Volk Chest Radiography with a Flat-Panel Detector: Image Quality with Dose Reduction after Copper Filtration Radiology, November 1, 2005; 237(2): 691 - 700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. M. Kroft, W. J. H. Veldkamp, B. J. A. Mertens, M. V. Boot, and J. Geleijns Comparison of Eight Different Digital Chest Radiography Systems: Variation in Detection of Simulated Chest Disease Am. J. Roentgenol., August 1, 2005; 185(2): 339 - 346. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Samei, J. Y. Lo, T. T. Yoshizumi, J. L. Jesneck, J. T. Dobbins II, C. E. Floyd Jr, H. P. McAdams, and C. E. Ravin Comparative Scatter and Dose Performance of Slot-Scan and Full-Field Digital Chest Radiography Systems Radiology, June 1, 2005; 235(3): 940 - 949. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Samei, J. T. Dobbins III, J. Y. Lo, and M. P. Tornai A framework for optimising the radiographic technique in digital X-ray imaging Radiat Prot Dosimetry, May 17, 2005; 114(1-3): 220 - 229. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Metz, P. Damoser, R. Hollweck, R. Roggel, C. Engelke, K. Woertler, B. Renger, E. J. Rummeny, and T. M. Link Chest Radiography with a Digital Flat-Panel Detector: Experimental Receiver Operating Characteristic Analysis Radiology, March 1, 2005; 234(3): 776 - 784. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. J. M. Kroft, J. Geleijns, B. J. A. Mertens, W. J. H. Veldkamp, H. M. Zonderland, and A. de Roos Digital Slot-Scan Charge-coupled Device Radiography versus AMBER and Bucky Screen-Film Radiography for Detection of Simulated Nodules and Interstitial Disease in a Chest Phantom Radiology, April 1, 2004; 231(1): 156 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P Meyer, E Buffard, L Mertz, C Kennel, A Constantinesco, and P Siffert Evaluation of the use of six diagnostic X-ray spectra computer codes Br. J. Radiol., March 1, 2004; 77(915): 224 - 230. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Bacher, P. Smeets, K. Bonnarens, An De Hauwere, K. Verstraete, and H. Thierens Dose Reduction in Patients Undergoing Chest Imaging: Digital Amorphous Silicon Flat-Panel Detector Radiography Versus Conventional Film-Screen Radiography and Phosphor-Based Computed Radiography Am. J. Roentgenol., October 1, 2003; 181(4): 923 - 929. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zahringer, V. Hesselmann, O. Schulte, K. F. Kamm, W. Braun, G. Haupt, B. Krug, and K. Lackner Reducing the Radiation Dose During Excretory Urography: Flat-Panel Silicon X-Ray Detector Versus Computed Radiography Am. J. Roentgenol., October 1, 2003; 181(4): 931 - 937. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ludwig, K. Ahlers, D. Wormanns, M. Freund, T. M. Bernhardt, S. Diederich, and W. Heindel Lumbar Spine Radiography: Digital Flat-Panel Detector versus Screen-Film and Storage-Phosphor Systems in Monkeys as a Pediatric Model Radiology, October 1, 2003; 229(1): 140 - 144. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ludwig, C. Schulke, S. Diederich, D. Wormanns, H. Lenzen, T. M. Bernhardt, P. Brinckmann, and W. Heindel Detection of Subtle Undisplaced Rib Fractures in a Porcine Model: Radiation Dose Requirement--Digital Flat-Panel versus Screen-Film and Storage-Phosphor Systems Radiology, April 1, 2003; 227(1): 163 - 168. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Dobbins III, E. Samei, H. G. Chotas, R. J. Warp, A. H. Baydush, C. E. Floyd Jr, and C. E. Ravin Chest Radiography: Optimization of X-ray Spectrum for Cesium Iodide-Amorphous Silicon Flat-Panel Detector Radiology, January 1, 2003; 226(1): 221 - 230. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Goo, J.-G. Im, H. J. Lee, M. J. Chung, J. B. Seo, H. Y. Kim, Y.-J. Lee, J.-W. Kang, and J. H. Kim Detection of Simulated Chest Lesions by Using Soft-Copy Reading: Comparison of an Amorphous Silicon Flat-Panel-Detector System and a Storage-Phosphor System Radiology, July 1, 2002; 224(1): 242 - 246. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||
