Speckle reduction techniques in B-mode ultrasound images

The traditional 1D spilt spectrum processing (SSP) method suffers from a relatively high-resolution loss limitation, inherent to the method. There is a trade-off between speckle reduction and resolution loss. On the other hand, the new 2D SSP with directive filters perform well. In this method, the filters are chosen to have a bandwidth larger than that of the ultrasound pulse and this helps to maintain image resolution. Even though SSP is a single-image method, it tackles the speckle reduction problem in a similar way to frequency compounding (a multi-image method), where a set of images from the same structure is acquired by using ultrasound pulses with different frequencies for each image. Likewise, the use of directive filters in the (single image) 2D SSP method has a similarity with angle compounding method (a multi-image method), where a number of images are acquired from different scanning angles in order to decorrelate the speckle. However, practical difficulties with the use of angle compounding methods in a medical context are overcome by the new 2D SSP with directive filters, which requires only a single image. The zero adjustment processing (ZAP) is a very powerful method, and is the only technique that can recover the apparent energy loss suffered in destructive interference, and promises well for further development. The hybrid method was used to demonstrate the synergy between different techniques, and the concept certainly looks ripe for further development. (6 pages)