Adaptive speckle reduction with color transparency display [in medical US imaging]

We present an adaptive speckle reduction method and a display technique in medical ultrasonic imaging. The speckle reduction algorithm defines a reference histogram that is derived from a known speckle region. Smoothing is applied to each region based on the similarity of that region´s histogram to the reference histogram. Regions of the original image are then classified into one of five categories: cyst, hypoechoic, mean speckle, hyperechoic, and edge; producing a color-coded, feature-segmented image. This feature-segmented image is then merged with the original image to produce a color transparency image. This final image preserves the detail of the original grayscale image while enhancing features extracted from the speckle reduced image