Abstract | OBJECTIVE: To test the hypothesis that the advanced multiple beam equalization radiography ( AMBER) imaging system is superior to conventional chest radiography in the demonstration of diffuse infiltrative lung disease, emphysema, pulmonary nodules, calcification within nodules, and mediastinal or hilar masses and lymphadenopathy. MATERIAL AND METHODS: The study involved 115 patients, each of whom underwent chest computed tomography (CT), AMBER, posteroanterior chest radiography, and conventional posteroanterior stereoscopic chest radiography (two films). All radiographs were obtained with the InSight Thoracic Imaging System. Four chest radiologists independently analyzed the 115 AMBER studies, 115 unpaired single conventional radiographs (a single film from a stereoscopic pair), and 115 stereoscopic conventional radiographs (2 films) for the presence of diffuse infiltrative lung disease, emphysema, pulmonary nodules, calcification within nodules, and mediastinal or hilar masses and lymphadenopathy. For each abnormality detected, the radiologists described their level of confidence based on a scale of 1 to 5. The 115 CT examinations were interpreted by consensus among 3 of the chest radiologists. The CT results were considered the standard. Receiver operating characteristic (ROC) techniques were used for statistical analysis. RESULTS: No statistically significant differences were found with ROC techniques between the AMBER system and single or stereoscopic conventional screen-film radiography for the abnormalities studied. CONCLUSION: We noted no clinically significant difference between AMBER and either single or stereoscopic conventional screen-film radiography in this prospective study of 115 patients in which CT (performed within 1 week of both radiographic examinations) was the standard.
|
Authors | S J Swensen, G L Aughenbaugh, L R Brown, G F Harms, P R Karsell, J E Gray, D M Ilstrup, D O Hodge |
Journal | Mayo Clinic proceedings
(Mayo Clin Proc)
Vol. 73
Issue 7
Pg. 636-41
(Jul 1998)
ISSN: 0025-6196 [Print] England |
PMID | 9663191
(Publication Type: Comparative Study, Journal Article, Research Support, Non-U.S. Gov't)
|
Topics |
- Humans
- Lung Diseases
(diagnostic imaging)
- Prospective Studies
- ROC Curve
- Radiography, Thoracic
(methods)
- Sample Size
- Sensitivity and Specificity
- Tomography, X-Ray Computed
(methods)
|