There are two types of wavefront distortion inside the tion produced by breast, liver, and abdominal wall [9,[13][14][16]. female breast: incoherent scattering and coherent interference. Adap-Basic deaberration algorithms [5-7] are recognized to be suitable tive wavefront compensation algorithms developed so far are useful only for weak scattering that can be modeled as a thin randomfor correction of incoherent distortion caused by scattering. The perphase screen located in the plane of the receiving aperture [8formances of these algorithms differ according to the extent of wave-9,18]. Extensions of the basic algorithms are made to correct front amplitude distortion. It is shown analytically and experimentally
amplitude distortion caused by strong scattering or distributed in this article that the matched filtering approach, while optimal for scattering 9]. A model-based approach that uses a priori infordetection, is not so for imaging when the wavefront amplitude is mation of the speeds of the rectus muscle layers inside the abdomdistorted in addition to phasefront distortion. Matched filtering algoinal wall has been developed to correct double image artifacts rithms increase wavefront amplitude variance and therefore decrease image contrast. The inverse filtering approach, while ideal for fidelity, caused by refraction . The proposed technique is not model is not stable when the signal-to-noise ratio is low. An approach toward based and therefore is freed from the necessity of having a priori inverse filtering, amplitude compression operation in addition to information about the source of the distortion.
phase deaberration, is introduced in this article. Analysis and experi-
The motivation of this study stems from the following. The ments show that its performance is superior to matched filtering algoalgorithms [5-9] that require no a priori knowledge about the rithms and to time-delay type correction algorithms without introducpropagating medium function differently upon incoherent scattering stability problems.