A COMPUTERIZED ANALYSIS SYSTEM IN CHEST RADIOGRAPHY - EVALUATION OF INTERSTITIAL LUNG ABNORMALITIES

We evaluated the usefulness of a computerized analysis system in the d etection of interstitial lung abnormalities in digitized chest radiogr aphy, This system uses the processes of four-directional Laplacian-Gau ssian filtering, linear opacity judgment, and linear opacity subtracti on. For qualitative analysis, we employed a combined radiographic inde x, which was calculated from two normalized radiographic indices obtai ned by linear opacity judgment and subtraction of linear opacities. We selected 50 regions of interest (ROIs) in patients with mild intersti tial lung abnormalities, 50 ROIs in patients with severe interstitial lung abnormalities, and 50 ROIs in individuals with normal lung parenc hyma. High-resolution computed radiography (HRCT) findings were used a s the standard of reference for this study. These ROIs were processed by our computerized analysis system, and radiographic indices were obt ained from each ROI. The area under the receiver operating characteris tic curve (At) was used as the measure of performance. The combined ra diographic index provided better results in the mild interstitial lung abnormality group (At = 0.94 +/- 0.02), but it also yielded good resu lts in the severe interstitial lung abnormality group (At = 0.98 +/- 0 .01). These results indicate that this system of combining radiographi c indices has improved the detection performance over that with our pr evious system. Copyright (C) 1997 by W.B. Saunders Company.