AN IMAGE ANALYZING SYSTEM FOR INTERSTITIAL LUNG ABNORMALITIES IN CHEST RADIOGRAPHY - DETECTION AND CLASSIFICATION BY LAPLACIAN-GAUSSIAN FILTERING AND LINEAR OPACITY JUDGMENT

RATIONALE AND OBJECTIVES. The effectiveness of a computerized analysis system to detect and characterize interstitial lung abnormalities see n on chest radiographs was evaluated. This method included a process o f four-directional Laplacian-Gaussian filtering and a process of linea r opacity judgment. For quantitative analysis of interstitial opacitie s, the radiographic index, which is the percentage of opacity areas in a region of interest, was obtained and evaluated in the images. These opacities represented interstitial lung abnormalities. METHODS. Two r egions of interest were selected in each right lung of 50 patients wit h normal lung parenchyma and 50 patients with diffuse interstitial lun g abnormalities, confirmed with high-resolution computed tomography. T hese regions of interest were processed by our computerized analysis s ystem. RESULTS. Abnormal lungs were well differentiated from normal lu ngs by the radiographic indices obtained from the images filtered by f our-directional Laplacian-Gaussian filters and from those processed by linear opacity judgment. However, honeycomb lesions and other interst itial abnormalities (interstitial changes other than honeycombing) wer e differentiated from each other only by the radiographic indices obta ined from the image processed by linear opacity judgment (P < .05). DI SCUSSION. These results indicate that this system is useful for the de tection and characterization of interstitial lung abnormalities.