Digital chest radiography with a selenium-based flat-panel detector versus a storage phosphor system: comparison of soft-copy images

OBJECTIVE. We compared the soft-copy images produced by a digital chest radiography system that uses a flat-panel X-ray detector based on amorphous selenium with images produced by a storage phosphor radiography system for the visualization of anatomic regions of the chest. MATERIALS AND METHODS. Two chest radiologists and two residents analyzed 46 pairs of posteroanterior chest radiographs on high-resolution video monitors (2560 x 2048 x 8 bits). In each pair, one radiograph was obtained with a storage phosphor radiography system, and the other radiograph was obtained with a selenium-based flat-panel detector radiography system. Each pair of radiographs was obtained at the same exposure settings. The interpreter rated the visibility and radiographic quality of 11 different anatomic regions. Each pair of images was ranked on a five-point scale (1 = prefer image A, 3 = no preference, 5 = prefer image B) for preference of technique. Statistical significance of preference was determined using the Wilcoxon's signed rank test. RESULTS. The interpreters had a statistically significant preference for the selenium-based radiography system in six (unobscured lung, hilum, rib, minor fissure, heart border, and overall appearance) of 11 anatomic regions (p<0.001) and for the storage phosphor system in two regions (proximal airway and thoracic spine) (p<0.05). Chest radiologists strongly preferred selenium-based images in eight regions, and they did not prefer storage phosphor images in any region. CONCLUSION. The soft-copy images produced by the selenium-based radiography system were perceived as equal or superior to those produced by the storage phosphor system in most but not all anatomic regions.     

Digital chest radiography with a large-area flat-panel silicon X-ray detector: clinical comparison with conventional radiography

This was a radiologists' preference study to compare a digital chest radiography system that utilizes a large-area silicon flat-panel detector with conventional radiography for visualizing anatomic regions of the chest. Conventional and digital posteroanterior (PA) and lateral chest radiographs were obtained in 115 patients. The PA and lateral image pairs were compared independently by three radiologists rating the overall appearance, 11 anatomic regions in the PA, and 9 in the lateral views. Statistical analysis was performed with the Wilcoxon signed-rank test with Bonferroni-Holm adjustment (p=0.05). For the PA view, the digital system performed significantly better for the overall appearance and for all anatomic regions except for the peripheral pulmonary vasculature and hilum, where no significant difference was found. For the lateral digital images, the regions trachea, costodiaphragmatic recess, and hilum were rated significantly worse. The regions retrosternal and retrocardiac lung were rated significantly better. The other regions and the overall appearance showed no significant differences. The described digital chest radiography system showed statistically superior visualization of anatomic regions for PA and an ambiguous performance for lateral images as compared with conventional radiography. After changing some image processing parameters for the lateral view, this system may be suitable for digitalization of chest radiography.     

Normal radiographic anatomy of thoracic structures: analysis of 316 lateral chest radiographs

PURPOSE: To study the frequency of visualization and characteristics of normal thoracic structures on lateral chest radiographs in the Japanese population. MATERIALS AND METHODS: We reviewed 316 lateral chest radiographs of men and women ranging in age from 20 to 90 years. The frequency of visualization and configuration of structures including major, minor, superior and inferior accessory fissures, and orifices of the right and left upper lobe bronchi were reviewed. RESULTS: On lateral chest radiographs, major fissure and minor fissure were visualized in 99.4% and 87.3%, respectively. Superior accessory fissure and inferior accessory fissure were visualized in 1.9% and 9.5%, respectively. Orifices of the right and left upper bronchi were seen in 92.4% and 98.4%, respectively. CONCLUSION: Frequency of visualization and characteristics of various normal anatomic structures on lateral chest radiographs in the Japanese population differ from those reported previously from the West. Familiarity with these normal thoracic structures and variations is important for daily image interpretation.     

Image quality for five modern chest radiography techniques: a modified FROC study with an anthropomorphic chest phantom

The purpose of the study was to compare the image quality for one conventional and four digital chest radiography techniques. Three storage phosphor systems, one selenium drum system, and one film-screen system were compared using a modified receiver-operating-characteristics method. Simulated pathology was randomly positioned over the parenchymal regions and the mediastinum of an anthropomorphic phantom. Eight observers (four chest radiologists, one specialist in general radiology, one hospital physicist, and two radiographers) evaluated 60 images for each technique. The selenium drum system (Philips, Eindhoven, The Netherlands) rated best for the detection of parenchymal nodules. Together with the storage phosphor system of generation IIIN (Philips/Fuji), the selenium drum system also rated best for detection of thin linear structures. The storage phosphor system of generation V (Fuji) rated best for the detection of mediastinal nodules. The first generation of the storage phosphor system from Agfa (Mortsel, Belgium) rated worst for the detection of parenchymal nodules and thin linear structures. These differences were significant (p < 0.0001). Averaging the results for all test objects, the selenium drum system and the storage phosphor system of generation V were significantly better than the other systems tested. The film/screen system performed significantly better than the first-generation storage phosphor system from Agfa, equal to the generation IIIN storage phosphor system (Philips/Fuji) and significantly worse than the selenium drum system (Philips) and the generation-V storage phosphor system (Fuji). The conclusion is therefore that the image quality of selenium-based digital technique and of the more recent generations of storage phosphor systems is superior to both conventional technique and storage phosphor systems using image plates of older types. Received: 11 September 1997; Revised: 25 June 1998; Accepted: 18 February 1999     

DR组织均衡对照研究

目的:对大面积非晶硅平板探测器的组织均衡软拷贝进行对照研究.方法:100位患者用数字平板探测器X线摄影系统获得后前位X线胸片,50位获得侧位X线胸片.通过后前位胸部X线照片中的结构:肺尖、心后和纵隔后肺血管、降主动脉、第六胸椎上缘等,侧位X线照片:心后肺血管、肺门、心影内的肺血管、心影上的横隔,以及第三胸椎上缘等,进行影像学评价.对组织均衡后所获影像间的显著性差异作测试.结果:对于第六胸椎上缘结构的显示,以及心后三角处肺纹理的显示上组织均衡图像更好(P＜0.05).结论:组织均衡这种宽的动态范围在穿透差的区域中可改善对比性能(如:纵隔),且能够使不同吸收特性的组织同时适当重显(如:骨、肺实质).     

Vertical fissure: computed tomographic observation and correlation with chest radiography

Purpose The purpose of this study was to assess radiographic and computed tomography (CT) findings of the vertical fissure. Materials and methods We assessed whether the vertical fissures appeared as a fine linear shadow or as a linear edge with lateral opacity and medial lucency on chest radiographs. The CT scans were evaluated for the presence of volume loss in the lower lobes and for the presence of a portion of the fissure that is orientated tangential to the X-ray beam. Results We observed vertical fissures in six patients. CT studies revealed volume loss in the lower lobes in all patients and showed the presence of a portion of the fissure that is orientated tangential to the X-ray beam in only two patients whose vertical fissures were not associated with lateral opacity and medial lucency on chest radiographs. Conclusion We believe that the vertical fissure is closely related to volume loss in the lower lobe and represents the edge of the anterobasal segment of a major fissure or a portion of the fissure that is orientated tangential to the X-ray beam, with or without lateral opacity and medial lucency.     

Clinical evaluation of a new digital chest imaging system

A digital chest imaging system based on a selenium detector has been installed in a large teaching hospital. Up to 1 May 1994, 7000 patients have been studied by this new technique. Hard copies are printed on a laser imager. Image quality of the digital system was analysed by day-to-day comparison with a conventional chest system. In a pilot study of 40 patients conventional and digital images were compared by analysing six different anatomical regions. Two readers judged the image quality of the digital system to be better than that of a conventional chest system, all regions being better visualised on the digital images (P <0.01). Acceptance of the digital chest system by radiographers, radiologists and other medical specialists is good. More than half of the normal patient load is routinely dealth with using the new system. Retakes were hardly necessary. Post-processing is possible, but was seldom used. These preliminary overall results of a digital chest imaging system based on a selenium detector in routine clinical practice are promising and further phantom studies and clinical studies using receiver operating characteristic (ROC) analysis are currently being performed. Correspondence to: J. P. M. van Heesewijk     

Normal radiographic anatomy of thoracic structures: analysis of 1000 chest radiographs in Japanese population

The purpose of this paper was to study the frequency of visualization and characteristics of normal thoracic structures on posteroanterior (PA) chest radiographs in Japanese population. 1000 consecutive normal PA chest radiographs of men and women ranging in age from 20 years to 90 years were reviewed. Frequency of visualization and configuration of structures including (1) fissure lines such as major, minor, vertical fissure line, and accessory fissures, (2) vascular structures including normal apical opacity, aortic nipple, and descending aortic interface, and (3) other structures including air in the oesophagus, aortic pulmonary stripe, and diaphragm were studied. On PA chest radiographs: (1) minor fissure, superolateral major fissure, superomedial major fissure, vertical fissure line, superior accessory fissure, and inferior accessory fissure were visualized in 74.7%, 19.7%, 15.4%, 1.6%, 2.9% and 13.1%, respectively. (2) Normal apical opacity was seen in 3.7%, while aortic nipple was seen in 0.9%. Descending aortic interface was obliterated in 13.7%. (3) Air in the oesophagus and aortic pulmonary stripe were seen in 8.9% and 17.7%, respectively. Hemidiaphragm was obliterated in 10.3% on the right, and in 32.4% on the left. Scalloping of the diaphragm was seen in 10.6% on the right, 6.5% on the left, and 4.3% bilaterally. Frequency of visualization and characteristics of various normal anatomical structures on chest radiographs in Japanese population differ from those reported previously from the West. Familiarity with these normal thoracic structures and variations is important for our daily image interpretation.     

Comparing image quality of flat-panel chest radiography with storage phosphor radiography and film-screen radiography

OBJECTIVE: To evaluate image quality of a large-area direct-readout flat-panel detector system in chest radiography, we conducted an observer preference study. A clinical comparative study was conducted of the flat-panel system versus the storage phosphor and standard film-screen systems. MATERIALS AND METHODS: Routine chest radiographs (posteroanterior) of 30 patients that were obtained using flat-panel, storage phosphor, and film screen systems were compared. The visibility of 10 anatomic regions and the overall image quality criteria were rated independently by three radiologists using a 5-point scale. The significance of the differences in diagnostic performance was tested with a Wilcoxon's signed rank test. Dose measurements for the three modalities were performed. RESULTS: The flat-panel radiography system showed an improved visibility in most anatomic structures when compared with a state-of-the-art conventional film-screen system and an equal visibility when compared with a storage phosphor system. The flat-panel system showed the greatest enhancement in the depiction of small detailed structures (p < 0.05) and achieved this with a reduction in overall radiation dose of more than 50%. CONCLUSION: The visibility of anatomic structures provided by this flat-panel detector system is as good as if not better than that provided by conventional or storage phosphor systems while emitting a reduced radiation dose.     

Chest radiography based on the interlobar fissures

Recognition of which interlobar fissure in the right lung is major or minor is important in chest radiography. Based on assessment of the interlobar fissure, the location of diseases or the degree of expansion of the lobes is comprehensible. As a general rule, the major fissure is not seen in normal cases. However, the major fissure is visualized in the presence of volume loss of the lower lobe. Although the medial end of the minor fissure is at the intermediate artery between the upper and lower lobe bronchi, that of the major fissure is continuous to the central portion of the mediastinum. The following cases will be presented: post-pleuritis, atelectasis of the superior segment of the lower lobe (S6), obstructive pneumonia in the middle and lower lobes due to stenosis of the intermediate bronchus, pneumonia of the upper lobe, and superior accessory fissure with atelectasis of the upper lobe.     

Using edge enhancement to identify subtle findings on soft-copy neonatal chest radiographs.

OBJECTIVE: The purpose of this study was to evaluate whether edge enhancement could improve the visibility of subtle findings on soft copies of neonatal chest radiographs. MATERIALS AND METHODS: Two radiologists reviewed 82 soft-copy neonatal chest radiographs before and after the application of edge enhancement on our picture archiving and communication system (PACS). The visibility of a pneumothorax (n = 22), central venous catheter (n = 32), umbilical arterial catheter (n = 36), endotracheal tube (n = 40), and normal anatomic structures (the minor fissure, anterior segmental bronchus of the right upper lobe, and aortic arch, n = 57) was evaluated. Six of 22 soft-copy images depicting a pneumothorax were excluded from the evaluation of image quality either because of the large size of the pneumothorax itself (n = 7) or because of the lack of confirmatory evidence that would have been provided by an additional lateral decubitus (n = 6) or cross-table lateral radiograph (n = 3). Image quality was evaluated by visual grading analysis. RESULTS: The visibility of a pneumothorax (p < 0.01), vascular catheters (p < 0.001), the minor fissure (p < 0.001), and the anterior segmental bronchus of the right upper lobe (p < 0.001) improved significantly after applying edge enhancement to soft copies of neonatal chest radiographs, whereas the visibility of the aortic arch did not improve. Evaluations of the improvements in the visibility of the endotracheal tube were inconsistent. CONCLUSION: Application of edge enhancement to soft copies of neonatal chest radiographs helps radiologists to identify small pneumothoraces, vascular catheters, and delicate normal structures, thereby improving the detection of subtle chest findings in the neonatal intensive care unit.     

Comparison of an automatic versus a semiautomatic mode for gray-scale adaptation for digital chest radiography

PURPOSE: To compare image quality of digital chest radiographs using 2 modes of gradation adjustment. METHODS: We compared image quality and visualization of anatomic landmarks of 50 chest radiographs after digital processing using a semiautomatic mode with a fixed gamma of 2.6 and an automatic mode with individual gamma adaptation. RESULTS: The mean gamma was significantly higher (P < 0.001) with the automatic mode (3.0 vs. 2.6, respectively). Patient constitution had no impact on the automatically adapted gamma for PA but showed a significant correlation (P < 0.001) for lateral images. For PA, there was a preference (P < 0.016) of the semiautomatic mode in heavier patients whereas no difference was seen in slim patients. For the lateral projection, there was a general preference (P = 0.001) of the automatic mode. CONCLUSION: For PA radiographs, the semiautomatic mode provides superior results for heavier patients without compromising the quality in slim patients. For lateral radiographs, the automatic mode provides generally superior results.     

Vertebral pseudolesion on lateral chest radiograph

OBJECTIVE: The purpose of this study was to describe and explain the basis for the lateral chest radiographic finding of a pseudolesion simulating a sclerotic vertebral lesion. CONCLUSION: Superimposition of the scapula on the upper thoracic spine causes a vertebral pseudolesion that simulates a sclerotic lesion.     

Digital imaging with a photostimulable phosphor in the chest of newborns

To evaluate use of a digital photostimulable phosphor imaging system in the neonatal nursery, 150 newborns were divided into three groups of 50. In the first two groups, screen-film and computed radiographs of the chest were obtained at the same radiation exposure; in the third group, computed radiographs were obtained with a 50% dose reduction (half-exposure computed radiographs). All images were blindly evaluated by three readers who scored the quality of visualization of the mediastinum, lung, bone, soft tissues, and endotracheal and nasogastric tubes, and also image density. No statistical differences in visualization of tubes existed among the three groups. Visualization of the mediastinum, lung, bones, and soft tissues was statistically significantly better on computed radiographs than on half-exposure computed radiographs; visualization of the lungs, bones, and soft tissues was statistically significantly better on screen-film radiographs than half-exposure computed radiographs. Image density was statistically better on computed and half-exposure computed radiographs than on screen-film radiographs.     

Digital storage phosphor imaging versus conventional film radiography in CT-documented chest disease

The advantages of imaging the chest with digital storage phosphor radiography (SR) may be nullified by its spatial resolution, which is lower than that of conventional film radiography (FR). To test the reader detection performance with the two modalities under clinical conditions, the authors compared 140-kVp isoexposure SR (system resolution: 0.2 mm, 10 bits) and FR images of a variety of chest abnormalities proved by computed tomography (CT) (157 patients, 244 abnormalities, 5,652 observations, six readers). In all tests, SR was as good as or better than FR (P less than .05). In overall detection, indicated by the average area of receiver operating characteristics, SR and FR were equivalent. SR was superior for mediastinal lesions and for pulmonary opacities greater than 2 cm in diameter. For all other types of pulmonary lesions and pleural abnormalities, SR and FR were equivalent. Currently available commercial SR systems can replace film radiographic systems in the detection of a wide variety of chest lesions. SR is likely to enable better visualization than FR in the detection of mediastinal and large pulmonary abnormalities.     

Sectional neuroanatomy of the upper thoracic spine and chest

This is the second in a series of articles on the spine. The first reviewed the anatomy of the neck. Subsequent articles are planned to cover the anatomy of the middle and lower thoracic spine. Procedures and trauma of the upper thoracic spine and chest are fraught with potentially serious complications. Hemothorax, pneumothorax, nerve damage, pulmonary collapse, and thoracic aortic aneurysm are included in the list. This article provides anatomically accurate schematics of innervation of the upper thoracic chest and spine that can be used to interpret magnetic resonance images of the muscles and nerves. Cross-sectional schematics of the upper chest and spine were drawn as they appear in imaging projections. The relevant nerves were color coded. The muscles and skin surfaces were labeled and assigned the color of the appropriate nerves. An organized comprehensive map of the motor innervation of the upper chest and spine allows the physician to increase the accuracy and efficacy of interventional procedures. This could also assist the electromyographer in correlating the clinical and electrophysiologic findings with magnetic resonance images.     

Neofissure after right upper lobectomy: radiographic evaluation

Following lobectomy of the right upper lobe of the lung, a single fissure, the neofissure, separates the right middle and lower lobes. In 25 patients in whom chest radiographs and computed tomographic (CT) scans were obtained after right upper lobectomy, the neofissure was identifiable to some extent on the radiographs in seven (28%) and on CT scans in 24 (96%). Among those patients in whom the neofissure was seen in its entirety, in 12 of 20 (60%) it was oriented predominantly in the coronal plane, paralleling the course of the original major fissure but displaced anteriorly and superiorly. In seven of 20 patients (35%) the superior aspect was coronal, rotated clockwise toward the sagittal plane inferiorly. In one of 20 patients (5%) the neofissure was oriented predominantly in the sagittal plane. The right middle lobe lies anterior and medial to the neofissure; the right lower lobe lies posterior and lateral.     

Sectional neuroanatomy of the lower thoracic spine and chest

This is the fourth in a series of articles on the spine. The first reviewed the anatomy of the neck; the second reviewed the upper thoracic spine and chest (T1-T4); and the third reviewed the middle thoracic spine and chest (T5-T8). The procedures performed in the lower thoracic spine include percutaneous biopsies of the liver and kidneys, percutaneous nephrolithotomy, spinal injections, radiofrequency ablations, electromyography of the diaphragm, trigger point injections, chemodenervation with botulinum toxin, acupuncture, aneurysm repair, and, occasionally, chest tube placement in the lower lung fields. Complications include subcapsular hematomas, infections, pneumothorax, hemothorax, spinal cord ischemia and resultant paraplegia, and, rarely, nephropleural fistulas. This article provides anatomically accurate schematics of innervations of the lower thoracic chest and spine (T9-T12) that can be used to interpret the magnetic resonance images of the muscles and the nerves. Cross-sectional schematics of the lower thoracic chest and spine were drawn as they appear on imaging projections. The relevant nerves were color coded. The muscles and skin surfaces were labeled and assigned the color of the appropriate nerves. An organized comprehensive map of the motor innervation of the lower thoracic chest and spine allows the physician to increase the accuracy and the efficacy of interventional procedures. This could also assist the electromyographer in correlating the clinical and electrophysiological findings with magnetic resonance images.     

Digital chest radiography system with amorphous selenium flat-panel detectors: Qualitative and dosimetric comparison with a dedicated film-screen system

PURPOSE: To compare the quality and radiation dose of a conventional film-screen system and a digital system with amorphous selenium detectors in the study of the chest, by verifying overall performance and exposure levels for the main chest structures in patients of different sizes. MATERIALS AND METHODS: An analogic system (Chest-Changer, Dupont, Day-light model 1000) and a digital system (Directray Rad 1000C, Hologic) were tested on a total of 1000 patients randomly assigned to one of two groups of 500 subjects. The patients were further subdivided according to BMI (Body Mass Index). Image quality was determined by two chest radiologists who evaluated eight anatomical structures. The entrance surface dose (skin-dose), calculated based on the exposure parameters, was taken as the patient dose. RESULTS: Mean dose delivered was very similar for both techniques in the PA view (0.28 mGy), but it was greater in the LL projections obtained with the digital system (1.20 mGy versus 0.83 mGy). The highest overall scores were assigned to 43% and 23.2% analogic radiograms and 64% and 70.2% digital radiograms, for the PA and LL projections respectively. The scores assigned to the various anatomical structures confirmed the better performance of the digital system in almost all of the regions considered. CONCLUSIONS: The mean quality of radiograms is definitely higher with the digital system, in particular in the LL projections, where the higher patient doses are counterbalanced by fewer repeated scans. The greater level of exposure in the digital system appears nonetheless tolerable on account of the greater informativeness and therefore diagnostic gain and also considering the possibilities for improving the system.     

The prone chest film.

Frontal chest radiographs in the prone position enhanced visualization of the posterior lung base in the presence of pleural effusions. Though the lateral decubitus view often displaced fluid adequately, in several cases only the prone film revealed basilar lung pathology; this was due to gravitational shift of the fluid away from this region as well as improved aeration of the lower lobe in the prone position, as shown by horizontal-beam lateral views. The prone chest film is recommended for evaluation of the lung base when obscured by fluid.