Energy Subtraction Radiography
A further advance in digital chest radiography is dual energy subtraction (ES) radiography. Here, instead of a single storage phosphor plate, two plates are used with a copper filter interposed between them. The full energy spectrum of the primary beam is recorded by the first plate as usual. Radiation passing through the first plate and filter undergoes low-energy filtration before encountering the second plate. Thus, the image recorded by the second plate consists mainly of the high-energy components of the beam. By performing a weighted subtraction of the two images, soft tissue and calcium/bone components are separated. Three chest images are produced: a standard image; a soft tissue-only image; and a bone/calcium-only image. Dual ES radiography can improve detection accuracy for certain types of pathology. The soft-tissue ES image clearly improves detectability of focal soft-tissue opacities in the lungs, such as pulmonary nodules that are obscured by overlying ribs. The “bone images” are useful for confirming the presence of calcification in benign pulmonary nodules, thereby potentially obviating the need for thin-section CT. Rib abnormalities such as sclerotic metastases, which can mimic lung nodules, are correctly identified on the “bone” ES images. In addition, calcified pleural plaques easily can be differentiated from pulmonary nodules.
Digital radiography also provides the opportunity for the images to be analyzed directly by the computer for the purpose of detecting, localizing, and characterizing radiographic abnormalities. Digital imaging techniques are now allowing for the development of CAD, an exciting new area of research. More info
CAD, the application of computer techniques to radiologic diagnostic decision making, includes programs that enhance diagnostic images for visual examination by separating components of the same image (eg, ES) or by integrating different images (eg, temporal subtraction). Another group of CAD programs includes those designed to automatically detect pathologic abnormalities in the image (eg, nodule detection) and highlight them for the radiologist or clinician. These techniques have all proven, in preliminary studies, to significantly increase the film reader’s sensitivity for detecting nodules and other subtle abnormalities.
A third form of CAD also encompasses programs that use available clinical data, radiologic data, or both to determine the most probable diagnosis. Such techniques include the use of an artificial neural network for differential diagnosis of interstitial disease or pulmonary nodules. Preliminary data has shown that the use of this technology can significantly improve diagnostic accuracy, even for experienced radiologists.
The past 5 to 10 years have witnessed an explosion of new technologies for evaluating the lung. Newer techniques have recently allowed for the possibility of evaluating pulmonary function as well as anatomy. Digital imaging is rapidly changing the practice of chest radiology and is leading to the development of CAD. Although helical CT and HRCT have become the cornerstone of pulmonary imaging, newer modalities such as PET and MRI may soon become critical components in the arsenal of tests used to evaluate pulmonary disease.