The purpose of this study was to determine whether thin-section CT could be used to differentiate small localized bronchioloalveolar carcinoma from peripheral adenocarcinoma having a bronchioloalveolar (replacement) growth pattern of alveolar lining cells and from adenocarcinoma not having a replacement growth pattern on the basis of the extent of ground-glass opacity revealed by thin-section CT.One hundred twenty-four small, surgically resected, peripheral adenocarcinomas from 119 patients (67 men and 52 women; mean age, 60 years) were studied. Lesion diameters were 0.4-2.0 cm (median, 1.5 cm). The extent of ground-glass opacity within lesions on preoperative thin-section CT was reviewed retrospectively by three thoracic radiologists. On the basis of replacement growth of alveolar lining cells, small adenocarcinomas were classified histologically as localized bronchioloalveolar carcinomas (n = 42) or as adenocarcinomas with (n = 53) or without (n = 29) a replacement growth pattern of alveolar lining cell...

Image-based computer-aided diagnosis (CADx) algorithm by use of convolutional neural network (CNN) does not necessarily require an image-feature extractor. Therefore, image-based CADx is powerful compared with feature-based CADx that requires the image-feature extractor for differential diagnosis of lung abnormalities such as lung nodules and diffuse lung diseases. We have also developed an image-based computer-aided detection (CADe) algorithm by use of regions with CNN features (R-CNN) for detection of lung abnormalities. We evaluated the performance of image-based CADx by use of CNN and that of image-based CADe by use of R-CNN for various kinds of lung abnormalities such as lung nodules and diffuse lung diseases.

The purpose of this study was to analyze the CT findings in lymphangitic carcinomatosis and to determine how the abnormalities seen on CT correlated with histologic findings and the results of pulmonary function tests. Accordingly, we reviewed CT findings in 20 proved cases and correlated them with histologic findings and with results of pulmonary function tests (available in 11). Special attention was given to the extent and degree of thickening of the peripheral and/or axial interstitium as seen on CT. In 15 of 20 patients, CT scans showed marked thickening of axial interstitium and subtle or no thickening of peripheral interstitium (four diffuse and 11 localized). In three cases, CT showed marked thickening of both axial and peripheral interstitium (two diffuse and one localized). In two cases, CT showed marked thickening of peripheral interstitium and subtle or no thickening of axial interstitium (one diffuse and one localized). These CT findings correlated with the pathologic findings. Marked impairm...

To analyze the lung-nodule interfaces on small peripheral pulmonary nodules (<2 cm) in thin-section CT (HRCT) images with fractal analysis. Thin-section CT images from 70 patients with bronchogenic carcinomas (61 adenocarcinomas and 9 squamous cell carcinomas) and 47 patients with benign pulmonary nodules (23 hamartomas, 13 organizing pneumonias, and 11 tuberculomas) were used. For calculation of fractal dimensions (FDs), the authors used a box-counting method for binary- and gray-scale images of nodules. FD(two-dimensional [2D]) was an FD obtained from the binary image, and FD(three-dimensional [3D]) was an FD obtained from the gray-scale image. The FD(2D)s of hamartomas were smaller than those of other nodules (P < 0.05). The FD(3D)s obtained from the gray-scale images of organizing pneumonias and tuberculomas were greater than those of bronchogenic carcinomas (P < 0.0001) and hamartomas (P < 0.0001). In bronchogenic carcinomas, FD(3D)s of adenocarcinomas were greater than those of squamous cell carcinomas (P < 0.05). Fractal dimensions reflect the characteristics of the lung-nodule interfaces of small peripheral pulmonary nodules. The FD(2D)s revealed the irregularities of the contours. On the other hand, FD(3D)s revealed the complexities of the heterogeneous textures. With use of FD(2D) and FD(3D), it may be possible to distinguish bronchogenic carcinomas from benign pulmonary nodules. Moreover, FD(3D) may make it possible to distinguish between adenocarcinomas and squamous cell carcinomas.