| Medical ultrasound imaging is in widespread use due to its low cost, portability, and lack of side effects. As transducer apertures and operating frequencies grow, however, resolving power continues to be limited by variations in the speed of sound within tissue. Concepts borrowed from other imaging disciplines can provide new insights into the aberration problem in medical ultrasound; in particular, some of the same issues have been studied for many years to improve acoustic imaging of the nonhomogeneous Earth.; The seismic imaging community has long understood that layered media may be approximated by constant-sound-speed media for beamforming purposes, leading to so-called time-migration algorithms. This raises the possibility of medical ultrasound applications—for example, brain imaging through the adult human skull. While our simulation results have been encouraging, experiments with animal skulls have been inconclusive due to the high attenuation of ultrasound in skull bone.; Complete data sets contain the raw reflection data from every combination of single transmit and receive elements in an array. The tremendous redundancy of complete data can be exploited for aberration correction by analyzing the time shifts on common-midpoint gathers. Until now, however, the wide-angle, random-scattering nature of biological targets has limited the accuracy and robustness of this approach. Prefiltering the data with two-dimensional fan filters largely solves this problem; the resulting algorithm (OFF) outperforms the most popular existing algorithms in almost all cases.; The concept of focusing-operator updating, recently popular in seismic imaging, provides insight into iterative aberration-correction algorithms using a transmit focus. We develop a new updating procedure based on dynamic programming. With careful selection of initial focus points, the resulting algorithm outperforms existing algorithms in some experiments.; Our results imply that single-valued focusing operators may be able to correct for most of the aberration encountered in soft tissues; that increasing aperture should not be viewed merely as a source of aberration, but as an opportunity to more fully correct it; and that the noise penalty for using complete data sets may not be as severe as commonly assumed. |
