| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Diagnostic Radiology, University of Heidelberg, INF 110, 69120
Heidelberg, Germany.
2
Present address: Department of Diagnostic Radiology (E0101), German Cancer
Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany.
Received May 3, 2001;
accepted after revision August 23, 2001.
Address correspondence to C. Fink.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Routine follow-up chest radiographs were obtained from 100 consecutive oncology patients using a large-area, amorphous silicon flat-panel detector digital radiography system (dose equivalent to a 400-speed film system). Hard-copy images were compared with previous examinations of the same individuals taken on a conventional film-screen system (200-speed). Patients were excluded if changes in the chest anatomy were detected or if the time interval between the examinations exceeded 1 year. Observer preference was evaluated for the image quality and the visibility of 15 anatomic structures using a five-point scale.
RESULTS. Dose measurements with a chest phantom showed a dose
reduction of approximately 50% with the digital radiography system compared
with the film-screen radiography system. The image quality and the visibility
of all but one anatomic structure of the images obtained with the digital
flat-panel detector system were rated significantly superior (p 
0.0003) to those obtained with the conventional film-screen radiography
system.
CONCLUSION. The image quality and visibility of anatomic structures on the images obtained by the flat-panel detector system were perceived as equal or superior to the images from conventional film-screen chest radiography. This was true even though the radiation dose was reduced approximately 50% with the digital flat-panel detector system.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All patients had previous posteroanterior and lateral film-screen chest radiographs within a period of 1 year (mean, 192 days; minimum, 34 days; maximum, 365 days) before the actual examination with the digital system.
Before the image evaluation, all hard copies were checked for changes in the chest anatomy (e.g., changes resulting from therapy or pulmonary disease) between the two examinations. Three patients were excluded from the study because of alterations of the chest anatomy caused by thoracic surgery. All other patients showed normal findings. Another nine patients were excluded from the study because the film-screen images were on loan to other hospitals at the time of image evaluation.
Imaging Systems
Digital radiography.—The digital radiography system
(Revolution XQ/i; General Electric Medical Systems, Milwaukee, WI) consisted
of an X-ray tube, a generator, and a large-area amorphous silicon flat-panel
detector. The structure and properties of this single-piece detector panel are
described in detail elsewhere
[5]. Briefly, the panel
consists of a cesium iodide scintillator and an amorphous silicon photodiode
array fabricated on a glass substrate. The thallium-doped cesium iodide
scintillator converts X rays into visible light. The scintillator material has
a needlelike microstructure and is applied on the silicon photodiode array.
The amorphous silicon photodiode array converts the light into an electric
charge, which is read out by electronics and converted to a 14-bit signal. The
detector has a matrix size of 2048 x 2048 pixels, with a pixel size of
200 µm. The active area of the detector measures 41 x 41 cm. The
maximum spatial resolution achieved with this detector is 2.5 line pairs per
millimeter [5]. The digital
radiography system also included a standard X-ray tube (Maxiray 100; General
Electric Medical Systems) and a standard high-voltage 80-kW generator (General
Electric Medical Systems).
Images were obtained with an automatic exposure control that was adjusted to provide a radiation dose equivalent to that provided by a 400-speed film-screen system. Exposure conditions were 125 kVp and 160 mA for the posteroanterior radiographs and 125 kVp and 250 mA for the lateral radiographs. The film focus distance was 180 cm. A grid (78 lines per centimeter; ratio, 13:1) was used. All digital images were processed with a standardized postprocessing set that was defined in collaboration with the manufacturer before the patient examinations. On the basis of the postprocessing algorithms supplied by the manufacturer, post-processing was adjusted to achieve a similar image appearance to the film-screen combinations used in our department (Cronex UV-G, DuPont, Bad Homburg, Germany; and Ultra-Vision C, Sterling Diagnostic Imaging, Newmark, Denmark). Briefly, the processing included a multiresolution algorithm to perform edge enhancement and dynamic range reduction and then applied an asymmetric sigmoidal look-up table to shape the contrast curve to match the screen-film combination. Finally, all digital images were transferred to a laser imager (LR 3300; Agfa, Mortsel, Belgium) and printed on laser film (Scopix Laser 2B DL, Agfa).
Film-screen radiography.—As a standard reference, a 200-speed (Cronex UV-G) conventional film-screen system was used in this study. The conventional chest radiography system consisted of an automatic film changer (Thoramat; Siemens, Erlangen, Germany), a standard X-ray tube (Opti 150, Siemens), and a high-voltage 70-kW generator (Tridoros 712 MP, Siemens). The system worked with an automatic exposure control using 125 kVp for the posteroanterior and lateral radiographs. The film-focus distance was 220 cm. A moving grid (40 lines per centimeter; ratio, 12:1) was used.
Dose Measurements
According to the manufacturer's specifications, the exposure settings of
the flat-panel detector radiography system used in this study were adjusted to
provide a radiation dose equivalent to that of a 400-speed film-screen system.
We assumed this would result in a dose reduction of approximately 50% compared
with the conventional film-screen radiographs obtained with 200-speed film.
Dose measurements were performed with a dosimeter (RadCal, Monrovia, CA) using
an anthropomorphic chest-phantom (Radiology Support Devices, Long Beach, CA)
before the patient examinations, which were performed under identical exposure
conditions (automatic exposure control using the same peak kilovoltage, tube
current, grid, etc.) as for the patient examinations. Dose measurements for
the patient examinations were not performed.
Image Evaluation
Hard-copy images were interpreted independently by three observers (a
board-certified general radiologist and two residents in the sixth and second
years of training). All observers had training in thoracic radiology. As
described previously, the postprocessing of the digital radiographs was
optimized to achieve a similar image appearance to the conventional
radiographs. However, the images of both modalities were easy to distinguish
because of the different film material and format. Images from the two
modalities were analyzed separately to exclude a potential bias from the
direct intermodality comparison. Patient names were either removed or
obscured, and no information about patient history was given. Furthermore, all
cases were presented in a randomized order.
The observers were allowed to handle the films and to use a spotlight if desired. No time or viewing distance constraints were given. The image quality and the visibility of defined anatomic structures, consisting of lung parenchyma, soft tissue, and bone structures (Fig. 1A,1B) were evaluated using a 5-point scale (1 = unsatisfactory, 2 = poor, 3 = fair, 4 = good, 5 = excellent).
|
|
Image quality was defined by the quality of exposure (i.e., the presence of under- or overexposed areas throughout the image) and the overall image quality, which was influenced, for example, by contrast and the presence of artifacts. The anatomic structures used for the image evaluation were similar to those of previous studies dealing with chest radiography [2, 8, 10] and were chosen to allow the evaluation of a range of anatomic structures. These structures included small structures such as the peripheral vessels and low-contrast structures such as the retrocardiac vessels.
Specifically, the structures included in the posteroanterior chest radiographs were the main bronchi on both sides, the peripheral bronchi, peripheral lung vasculature of the left apex and the right base, retrocardiac and retrodiaphragmatic lung vessels, the right pulmonary artery, the descending aorta, the paravertebral diaphragm on the left side, and the pedicles of the upper six thoracic vertebrae (Fig. 1A). In the lateral radiographs, the anatomic structures consisted of retrocardiac lung vessels, the hilus, lung vessels in the heart shadow, the diaphragm at the heart shadow, and the upper third of the thoracic spine (Fig. 1B).
Statistical Analysis
Mean values and standard deviations were calculated for the image quality
and anatomic structures for each modality. Differences between the images
obtained with the digital radiography system and the film-screen radiography
system were tested for significance using the Wilcoxon's signed rank test.
Because the image quality and visibility of the anatomic structures in this study were rated subjectively by the observers, an assessment of interobserver agreement was desirable to get an impression of the objectivity and reliability of the image evaluation. Kappa statistics were not applicable because of the large number of multivariate responses caused by the number of observers and response categories [11]. Instead, Spearman's rank correlation was performed for all criteria for all observer combinations. The level of significance was calculated for each correlation. A significant correlation indicates that the scores of the observers may be combined for an overall score for image quality and anatomic visibility. As a result of the high level of correlation measured in this study, we have reported averaged results of all three observers.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Image Evaluation
The quality of the images obtained with the digital flat-panel detector
system was rated significantly superior to the quality of those obtained with
the conventional film-screen radiography system
(Table 2). The mean scores for
the image quality of the digital images were 4.11 and 3.93 (posteroanterior
and lateral views, respectively) compared with 3.74 and 3.51 for the
film-screen images. The difference (
) of the mean scores between the
two systems for the image quality was 0.37 and 0.42 (posteroanterior and
lateral views, respectively). For the image quality, a significant
interobserver agreement was shown by the Spearman's rank correlation
(p 0.02).
|
The visibility of all but one anatomic structure on the images obtained with the digital system was rated significantly superior to the images obtained with the reference film-screen system (Tables 3 and 4). One exception was the basal peripheral lung vessels in the posteroanterior view. An insignificant advantage was noted for the digital system for these structures.
|
|
For the lung parenchymal structures, the biggest difference noted between
the two systems was the visibility of the small lung vessels in the
retrocardiac space on the lateral view. The mean score for the digital images
of this structure was 3.73, compared with 3.09 of the conventional film-screen
images ( = 0.64). For the soft-tissue structures, the biggest
differences between the two radiography systems were observed for the
delineation of the diaphragm in the posteroanterior view. The mean score for
the digital images of this structure was 4.03, compared with 3.54 for the
film-screen system ( = 0.49) (Table
4).
For the bony structures, the biggest differences between the two
radiography systems were observed for the visibility of the pedicles of the
upper thoracic spine on the posteroanterior view. For these structures, the
mean score of the digital images was 3.38, compared with 3.09 for the images
of the film-screen radiography system ( = 0.29)
(Table 4). For most anatomic
structures, a significant interobserver agreement was shown by the Spearman's
rank correlation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One feature of these new digital detectors is the higher detective quantum efficiency compared with film-screen radiography. "Detective quantum efficiency" describes the efficiency of the transduction of the input signal into the output signal. In particular, detective quantum efficiency is defined as the ratio of the squared signal-to-noise ratio at the output of the detector to the squared signal-to-noise ratio at the input of the detector. Usually detective quantum efficiency is measured as a function of spatial frequency [1, 2, 5, 6]. Detective quantum efficiency is generally accepted as the most important parameter for characterizing the performance of a detector system [5, 9]. The detector used in this study showed a significantly higher detective quantum efficiency in theoretic analysis and experimental testing than did film-screen systems [1, 2, 5]. The high detective quantum efficiency of these flat-panel digital detectors is the basis for a potential dose reduction.
In this study, the exposure settings of the digital system were adjusted to provide a radiation dose equivalent to that of a 400-speed film system. This adjustment resulted in an approximately 50% dose reduction with the digital images, as shown in the dose measurements using an anthropomorphic chest phantom. However, dose measurements were not performed for the patient examinations.
The images obtained by the flat-panel digital system were considered better or at least equivalent to the images obtained with the reference conventional film-screen system. The potential of a dose reduction without substantial loss of image quality was shown in various experimental studies with this kind of digital detector [3, 7, 9]. In a clinical study with 15 patients, Strotzer et al. [8] reported that at 50% dose reduction, a flat-panel detector system produced equivalent or superior image quality compared with the reference film-screen radiography system.
One limitation of our study is that, like similar studies dealing with this new technology [2, 8,9,10], our results are based on observer preference rather than on diagnostic performance of the radiography systems. Typically, observer preference studies are performed first and are then followed by diagnostic accuracy studies. The advantage of observer studies is that they allow a wide range of anatomic features to be compared, and they may be used to determine which diagnostic studies are required to further validate a new technology.
Another limitation of our study is that although the postprocessing of the digital images was adjusted to produce a similar image appearance to conventional radiographs, the images of both modalities could be easily distinguished because of their different film material and format. This limitation is a potential source of bias that could have influenced our results if the observers preferred one radiography system to the other. To reduce this source of bias and to avoid direct intermodality comparison, the images of the different radiography systems were evaluated in separate sessions, patient names were either removed or obscured, and all cases were presented in a random order.
Spearman's rank correlation was performed to get an impression of the interobserver agreement and therefore of the objectivity and reliability of the image evaluation. (Kappa statistics were not applicable in this study because of the large number of multivariate responses [11].) A significant positive correlation was observed for the image quality and for most anatomic structures for all observer combinations.
In this study, all digital images were printed on hard-copy laser film after a standard postprocessing procedure in order to have similar conditions for the image evaluation of the digital images and the film-screen images. This procedure causes a potential bias against the digital system because possible advantages of the digital technology using various postprocessing features (e.g., digital magnification, edge-enhancement) were not implemented.
In this study, the largest differences for the visibility of anatomic structures between the two modalities were noted for the mediastinal structures (such as the delineation of the descending aorta in the posteroanterior view) (Fig. 2A,2B) or for structures with high-contrast differences (such as the retrocardiac vessels in the lateral view) (Fig. 3A,3B). This finding may be due to the wide dynamic range of these digital detectors compared with conventional film-screen radiography systems [1, 2, 6]. The wide dynamic range improves the contrast of features in poorly penetrated regions (e.g., the mediastinum) and enables a simultaneous adequate representation of tissues with substantially different absorption characteristics (e.g., bone, lung parenchyma) [1, 6]. Garmer et al. [4] reported similar findings when the imaging of mediastinal abnormalities was significantly improved with the flat-panel digital radiography system compared with film-screen radiography.
|
|
|
|
In conclusion, the image quality and the visibility of structures on the images obtained with a flat-panel detector based on amorphous silicon were perceived as equal or superior to images obtained with a conventional film-screen radiography system. This finding was true even though the radiation dose of the digital radiographs was reduced approximately 50% compared with the film-screen radiographs. Further clinical studies are desirable to compare the diagnostic accuracy of the flat-panel detector system to that of conventional film-screen systems.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
er HJ, et al.
Digital radiography versus conventional radiography in chest imaging:
diagnostic performance of a large-area silicon flat-panel detector in a
clinical CT-controlled study. AJR
2000;174:75
-80
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  L J M Kroft, W J H Veldkamp, B J A Mertens, J-P A van Delft, and J Geleijns Dose reduction in digital chest radiography and perceived image quality Br. J. Radiol., December 1, 2007; 80(960): 984 - 988. [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]
|
|
|
|
 |
|
 L. J. M. Kroft, W. J. H. Veldkamp, B. J. A. Mertens, J. P. A. van Delft, and J. Geleijns Detection of simulated nodules on clinical radiographs: dose reduction at digital posteroanterior chest radiography. Radiology, November 1, 2006; 241(2): 392 - 398. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Bacher, P. Smeets, L. Vereecken, A. De Hauwere, P. Duyck, R. De Man, K. Verstraete, and H. Thierens Image quality and radiation dose on digital chest imaging: comparison of amorphous silicon and amorphous selenium flat-panel systems. Am. J. Roentgenol., September 1, 2006; 187(3): 630 - 637. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. De Hauwere, K. Bacher, P. Smeets, K. Verstraete, and H. Thierens Analysis of image quality in digital chest imaging Radiat Prot Dosimetry, December 1, 2005; 117(1-3): 174 - 177. [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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
W. Huda The Current Concept of Speed Should Not Be Used to Describe Digital Imaging Systems Radiology, February 1, 2005; 234(2): 345 - 346. [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]
|
|
|
|
|
|
F. Fischbach, T. Freund, R. Rottgen, U. Engert, R. Felix, and J. Ricke Dual-Energy Chest Radiography with a Flat-Panel Digital Detector: Revealing Calcified Chest Abnormalities Am. J. Roentgenol., December 1, 2003; 181(6): 1519 - 1524. [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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
M. Ganten, B. Radeleff, A. Kampschulte, M. D. Daniels, G. W. Kauffmann, and J. Hansmann Comparing Image Quality of Flat-Panel Chest Radiography with Storage Phosphor Radiography and Film-Screen Radiography Am. J. Roentgenol., July 1, 2003; 181(1): 171 - 176. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
