Use of 3D adaptive raw-data filter in CT of the lung: Effect on radiation dose reduction

OBJECTIVE. The purpose of this study was to determine the effectiveness of a 3D adaptive raw-data filter in improving image quality and the role of the filter in radiation dose reduction in lung CT. MATERIALS AND METHODS. Fifty-eight chest CT examinations were performed with a 16-MDCT scanner. Two acquisitions were performed with different tube current-exposure time settings (50 and 150 mAs, 120 kVp). Four series of lung images were prepared from two sets of raw data with and without application of a 3D adaptive filter (50 mAs, 50 mAs with filter, 150 mAs, 150 mAs with filter). Three blinded readers using a 5-point scale from 1 (nondiagnostic) to 5 (excellent) independently evaluated image quality in five lobes and the lingula. A set of images was considered acceptable when scores in all six regions were 3 (acceptable) or higher. The SD of attenuation was calculated in 24 regions of interest. RESULTS. The overall mean image quality scores were 3.09, 3.53, 4.02, and 4.38 for the 50 mAs, 50 mAs with filter, 150 mAs, and 150 mAs with filter sets, respectively. Scores were significantly better with filter application (p < 0.001). A significant decrease in SD of attenuation was observed with filter application (p < 0.001). Among the respective series of images, 18, 52, 50, and 58 sets were judged acceptable with no significant difference in acceptability between images obtained at 50 mAs with a filter and at 150 mAs (p = 0.72). With filter application, the acceptability of 50-mAs images became comparable with that of 150-mAs images, making dose reduction to 50 mAs practical. CONCLUSION. Use of a 3D adaptive raw-data filter improved the quality of lung images, making dose reduction to 50 mAs attainable with use of the filter.