ROC ANALYSIS FOR POST-PROCESSING OF IMAGE DATA OF DIGITAL LUMINESCENCE RADIOGRAPHS OF THE THORAX

Purpose: To examine the extent to which digital luminescence radiograp hy (DLR) can be used for the imaging of pulmonary nodules and intersti tial lung disease in chest radiography without any loss of image quali ty. Additionally: to examine whether post-processing of image data can optimise the recognisability of varied image details. Materials and M ethods: Detail perceptibility studies were performed on an anthropomor phic thorax phantom with simulated nodules and small linear and reticu lar details. Under standard conditions, digital luminescence radiograp hs were obtained in 7 different image modes, and these were compared w ith a 200-speed screen-film system. The detection of these systems was evaluated in an ROC analysis on the basis of 19 200 individual observ ations. Results: Edge enhancement or application of high-frequency-enh ancing small filter kernels (S 5) slightly improves the detection of l inear structures; however, the illustration of nodular details is mark edly reduced. Larger filter kernels (S 20, S 40) make a definitive det ection possible - not only of circular, but also of linear details. Co nclusions: Storage phosphor radiographs are equal to the tested analog screen-film-system, The optimization of post-processing can be helpfu l in the prevention of routine multiple documentations.