EFFECTS OF REDUCED EXPOSURE ON COMPUTED RADIOGRAPHY - COMPARISON OF NODULE DETECTION ACCURACY WITH CONVENTIONAL AND ASYMMETRIC SCREEN-FILM RADIOGRAPHS OF A CHEST PHANTOM

OBJECTIVE. Radiographic exposure has been thought to have little impac t on the diagnostic quality of chest computed radiography because of a utomatic digital control of global optical density. The objectives of this study were to compare images obtained with two different exposure s in computed radiography with conventional and asymmetric screen-film images of the chest for the detection of simulated lung nodules by us e of receiver operating characteristic analysis and to relate differen ces in observer performance to parameters of image noise measured for each receptor condition. MATERIALS AND METHODS. At 110 kVp (fixed), ex posures for the two screen-film systems were those necessary to achiev e adequate optical densities over the lung and mediastinal regions of an anthropomorphic phantom. The two exposures used for the computed ra diographs corresponded to the exposure used for the conventional chest screen-film system and an exposure 22% lower. An anthropomorphic phan tom constructed of materials matched to the muscle, lung, and bone att enuation of a muscular adult man was used. Soft-tissue-equivalent plas tic nodules of various sizes were secured at multiple sites on the pha ntom to simulate lung nodules. The chest phantom was imaged in 50 conf igurations with a total of 70 superimposed nodules. The perceptual per formances of five radiologists were compared by use of receiver operat ing characteristic analysis. The signal-to-noise ratio in the mediasti num and the coefficient of variation of noise were measured for all fo ur image conditions by use of a step wedge technique to provide an exp lanation for differences in diagnostic accuracy. RESULTS. We found no significant differences in the detection of lung nodules between the t wo screen-film systems or between the conventional screen-film images and the standard-exposure computed radiographs. However, there was a s ignificant decrease in nodule detection on computed radiographs obtain ed at a reduced exposure; this result was associated with a 21% decrea se in the signal-to-noise ratio. CONCLUSION. Our results show that und erexposure of computed radiographs decreases the detection of low-cont rast objects such as lung nodules. Although consistent global optical density on computed radiographs is achieved over a wide range of expos ures, the alterations in signal-to-noise ratio that result from undere xposure can reduce the diagnostic quality of computed radiographs.