This study develops an algorithm for multiple-windows blending of CT images in red-green-blue (RGB) color space. A CT image is windowed at three different combinations of window width (WW) and window level (WL) values. The first window image is then colorized red (R), the second window image is colorized green (G), and the third window image is colorized blue (B). These three images, representing three combinations of window settings, can then be represented as a single RGB color image. The technique was applied to axial images of TOS-phantom with various objects and anthropomorphic phantom. This multiple-windows blending enables the result of multiple-windows to be presented in a single view without any subsequent changes to the windows settings. This multiple-windows blending is able to visualize many tissues of interest simultaneously with high contrast depending on the initial windows settings. The blending of soft tissue, bone, and lungs windows in thoracic images, for example, produces an image in which the details of soft tissues, bones, and lung nodules can be seen simultaneously in one view. Multiple-windows blending can potentially hasten image interpretation because more information is viewed in a single image. However, this novel method needs the radiologist to adapt to interpreting the new multiple-windows blended image. The method also requires standardization and optimization.
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition