SINGLE-EXPOSURE DUAL-ENERGY CHEST IMAGES WITH COMPUTED RADIOGRAPHY - EVALUATION WITH SIMULATED PULMONARY NODULES

RATIONALE AND OBJECTIVES. Dual-energy subtraction radiography using co mputed radiography (CR) can aid in the detection of pulmonary abnormal ities such as nodules, but the subtracted image requires higher x-ray exposure than usual to reduce quantum mottle. To reduce quantum mottle without increasing x-ray exposure, a new dual-energy subtraction algo rithm was investigated that included an edge-adaptive smoothing proces s and a subtraction process. The signal-to-noise ratio and the image q uality of this new subtracted image was significantly superior to that of conventional subtracted images. METHODS. Observer performance of t he subtracted digital radiography in detecting simulated pulmonary nod ules was compared with original CR images and conventional subtracted digital radiography of 50 patients. RESULTS. A combination of an origi nal CR image and a new subtracted CR image was significantly superior to a single original CR image or a combination of an original CR image and a conventional subtracted CR image (P < .01). DISCUSSION. The sin gle-exposure dual-energy subtraction method is superior to the convent ional subtraction method in the detection of pulmonary nodules.