The use of ultrasonic imaging for the detection and diagnosis of a variety of cancers, is presently limited by speckle which tends to reduce the resolution and detection of low contrast regions. TETRAD has designed and implemented an image processing algorithm which greatly reduces the speckle in ultrasonic images and, at the same time, preserves the sharp details associated with boundaries and well defined reflectors. The proposed research is designed to perform a preliminary evaluation of this speckle reduction algorithm using phantoms and clinical cases. A successful demonstration of the algorithm would provide a solution to one of the oldest problems associated with ultrasound images -- the grainy appearance. In addition to the obvious cosmetic improvement, it is anticipated that the probability for detecting a variety of lesions (and physician confidence) will be greatly enhanced. Speckle reduction will also enable the application of more advanced image analysis methods which presently cannot be applied to ultrasound images. Successful demonstration of the algorithm will lead to a real-time hardware implementation which can be added to the video display system of any commercial ultrasound imaging system.
