Challenging the lambda-Half Element Pitch Limit by a Sparse Least-Squares Wrapping Inversion in k-Space

In ultrasound imaging, a transducer element pitch that is larger than half the minimum wavelength is prone to the occurrence of grating lobe artifacts. Most linear arrays fail to fulfill this limit by far, which strongly restricts steering capabilities. On the other hand, increasing the pitch even further has the potential to greatly reduce the required channel data throughput. We propose that the lambda-half limit can be circumvented by a dedicated compounding approach that leverages redundancies in k-space coverage for different transmit angles to correct for ambiguities in the receive angles. The performance of this approach was evaluated in two simulation studies. In a first "inverse crime" study, the contrast of an anechoic inclusion was increased by 11.6 dB along with a gCNR improvement from 0.94 to 0.99. For a public reference dataset (PICMUS), an artifact suppression of up to 18.1 dB was shown, while the resolution was preserved.