Compressive sensing in forming ultrasound images

Abstract

In an image compressing ultrasound system, for generating an imaging sample, delays are applied transducer-element-wise to respective time samples. The delayed samples are summed coherently in time, the coherently summed delays being collectively non-focused. An image is sparsified based on imaging samples and, otherwise than merely via said imaging samples, on angles () upon which respectively the delays for the generating of the imaging samples are functionally dependent. An image-compressing processor () may minimize a first p-norm of a first matrix which is a product of two matrices the content of one representing the image in a compression basis. The minimizing is subject to a constraint that a second p-norm of a difference between a measurement matrix and a product of an image-to-measurement-basis transformation matrix, an image representation dictionary matrix, and the matrix representing the image in the compression basis does not exceed an allowed-error threshold. The measurement matrix is populated either by channel data, or by output of a Hilbert transform applied to the channel data in a time dimension.