Ultrasound imaging (US) is susceptible to several types of artifacts. Most artifacts appear because of transducer limitations and simplified assumptions on the wave propagation. The artifacts are sometimes used as a component that contains tissue information; however, they often lead to a misinterpretation in the clinical diagnosis. Therefore, to improve the clinical utility of ultrasound in difficult-to-image patients and settings, a number of artifact removal methods have been proposed that aim at boosting image quality. Classical optimization-based methods have severe limitations due to their limited performance and high computation requirements. Furthermore, it is difficult to obtain parameters for producing high-quality output. A quick remedy for the aforementioned issues is the deep learning approach, which offers high performance compared with the traditional methods despite the significantly reduced runtime complexity. Another big advantage is that the same parameters as those learned during the training phase can be used to process different input images. This has motivated the scientific community to design deep-neural-network-based approaches for US artifact removal tasks.

The basics of sound propagation in three dimensions are considered. Simple models for sources are presented, including isotropic and nonisotropic sources, such as the dipole, quadruple, line sources, and oscillating baffles. Sound propagation from those sources is treated in the ray approximation, where effects due to the wave nature of sound are suppressed. For finite sources, it is shown that in the absence of reflections, sound intensity falls off following an inverse square law. Reflections in a room are considered and lead to reverberation. The time sound remains in a room is characterized using a reverberation time, and a simple equation is used to relate that time to room dimensions, materials used, and possible contents of the room. Sound in a room from multiple sources is considered using sound-energy density rather than intensity. One example of the treatment of multiple sources is known as the cocktail party effect.