Singular value filter for imaging or detection

A system for use with a transducer array, the transducer array configured to deliver energy to a tissue region, the system comprising:

• a processor circuit configured to:
• obtain, from the transducer array, information indicative of the delivered energy reflected from the tissue region;
• form respective input matrices representative of the obtained information indicative of the energy reflected from the tissue region, the respective input matrices including information corresponding to a spatial dimension and a temporal dimension, the input matrices respectively comprising an ensemble-of-interest and at least one ensemble corresponding to a spatial location nearby a spatial location corresponding to the ensemble-of-interest;
• perform respective singular value decompositions on the respective input matrices to obtain respective sets of singular values corresponding to respective sets of singular vectors;
• obtain respective output matrices including weighting a respective projection of a respective ensemble-of-interest onto at least one of the singular vectors included in a respective set of singular vectors using a weighting function, including adaptively determining the weighting function using information about at least one of the respective singular values;
• and
• using the respective output matrices, at least one of determine a characteristic, or construct an image, of at least a portion of the tissue region;
• wherein the weighting function:
• varies continuously between 0 and 1 without a discontinuity;
• and
• includes a modified sigmoidal weighting function.

A method, comprising:

• obtaining, from a transducer array, information indicative of the delivered energy reflected from the tissue region;
• forming respective input matrices representative of the obtained information indicative of the energy reflected from the tissue region, the respective input matrices including information corresponding to a spatial dimension and a temporal dimension, the input matrices respectively comprising an ensemble-of-interest and at least one ensemble corresponding to a spatial location nearby a spatial location corresponding to the ensemble-of-interest;
• performing respective singular value decompositions on the respective input matrices to obtain respective sets of singular values corresponding to respective sets of singular vectors;
• obtaining respective output matrices including weighting a respective projection of a respective ensemble-of-interest onto at least one of the singular vectors included in a respective set of singular vectors;
• and
• using the respective output matrices, at least one of determining a characteristic, or constructing an image, of at least a portion of the tissue region.