# US 10,863,972 B2 · Image domain ultrasound imaging denoising filter > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,863,972 B2 - Document ID: US10863972B2 - Status: Granted - Kind: B2 - Application number: 15/469,733 - Filed: 2017-03-27 - Published: 2018-09-27 - Granted: 2020-12-15 - Earliest claimed: 2017-03-27 - Assignees: B-K Medical Aps - Applicants: B-K Medical Aps - Inventors: Drake A. Guenther - Web page: https://patentgrep.com/patent/US10863972B2 ## CPC - A61B 8/5269: {involving detection or reduction of artifacts} - A61B 8/5207: {involving processing of raw data to produce diagnostic data, e.g. for generating an image} - G01S 7/52046: {Techniques for image enhancement involving transmitter or receiver (image enhancement by image data processing G06T5/00)} - G01S 7/52077: {with means for elimination of unwanted signals, e.g. noise or interference} - G01S 15/8915: {using a transducer array} - G01S 15/8995: {Combining images from different aspect angles, e.g. spatial compounding} - A61B 8/4488: {the transducer being a phased array} - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES ## Abstract An apparatus includes a memory device with computer readable instructions and a processor configured to execute the computer readable instructions encoded on the memory device. The processor, in response to executing the computer readable instructions, obtains an ensemble of ultrasound images with diversity in an ensemble dimension, extracts a sub-set of data from each of the images, constructs a data matrix with the extracted data, wherein the data matrix has a dimension of space versus the ensemble dimension, identifies a largest eigenvalue(s) and a corresponding eigenvector(s) in the data matrix, computes a coherent signal projection matrix with the identified corresponding eigenvector(s), filters the data matrix with the coherent signal projection, and generates an ultrasound image with the filtered data matrix. ## Claims ### Claim 1 An apparatus, comprising: a memory device with computer readable instructions; a processor configured to execute the computer readable instructions encoded on the memory device, wherein the processor is configured to, in response to executing the computer readable instructions: obtain an ensemble of ultrasound images with diversity in an ensemble dimension; extract a sub-set of data from each of the images; construct a data matrix with the extracted data, wherein the data matrix has a dimension of space versus the ensemble dimension; identify a largest eigenvalue(s) and a corresponding eigenvector(s) in the data matrix; compute a coherent signal projection matrix with the identified corresponding eigenvector(s); filter the data matrix with the coherent signal projection; and generate an ultrasound image with the filtered data matrix. ### Claim 2 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes a signal of interest that is coherent across the ensemble dimension, and wherein the ensemble of ultrasound images includes clutter, reverberation, and/or electronic noise across which are incoherent across the ensemble dimension. ### Claim 3 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes images generated from data received in response to transmitted plane waves at different tilt angles, and the data matrix has a dimension of space versus the tilt angle. ### Claim 4 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes images generated from synthetic aperture datasets in response to transmissions from different transmit elements, and the data matrix has a dimension of space versus the transmit element. ### Claim 5 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes images generated with different apodizations, and the data matrix has a dimension of space versus apodization profile. ### Claim 6 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes images generated in response to transmissions from diverging waves, and the data matrix has a dimension of space versus diverging wave, where each diverging wave is defined by its virtual source. ### Claim 7 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes images generated with a fixed transmit focus, and the data matrix has a dimension of space versus transmit event. ### Claim 8 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes images generated from a first stage beamformer, and the data matrix has a dimension of space versus first stage beamformer output. ### Claim 9 (depends on 1) The apparatus of claim 1, wherein the sub-set is less than an entire set of the data from each of the images. ### Claim 10 (depends on 1) The apparatus of claim 1, wherein the sub-set is equal to an entire set of the data from each of the images. ### Claim 11 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes one-dimensional images. ### Claim 12 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes two-dimensional images. ### Claim 13 (depends on 1) The apparatus of claim 1, wherein the ensemble of ultrasound images includes three-dimensional images. ### Claim 14 (depends on 1) The apparatus of claim 1, further comprising: a transducer array configured to transmit ultrasound signals. ### Claim 15 (depends on 14) The apparatus of claim 14, further comprising: a signal processor configured to produce the ensemble from echo signals generated in response to the transmitted ultrasound signals and received by the transducer array. ### Claim 16 A method, comprising: extracting a sub-set of data from each image of an ensemble of ultrasound images having diversity in an ensemble dimension; constructing a data matrix with the extracted data, wherein the data matrix has a dimension of space versus the ensemble dimension; identifying largest eigenvalue(s) in the data matrix; computing a coherent signal projection matrix with eigenvector(s) of the identified eigenvalue(s); filtering the data matrix with the coherent signal projection; and generating a decluttered ultrasound image with the filtered data. ### Claim 17 (depends on 16) The method of claim 16, wherein the ensemble of ultrasound images includes a signal of interest that is coherent across the ensemble dimension, and wherein the ensemble of ultrasound images includes clutter, reverberation, and/or electronic noise across which are incoherent across the ensemble dimension. ### Claim 18 (depends on 17) The method of claim 17, further comprising: computing a covariance matrix with the data matrix, wherein the largest eigenvalue(s) and the corresponding eigenvector(s) are identified from the covariance matrix. ### Claim 19 (depends on 17) The method of claim 17, further comprising: compounding the filtered data across the ensemble dimension, wherein the decluttered ultrasound image is generated with the compounded filtered data. ### Claim 20 A non-transitory computer readable medium encoded with computer executable instructions which when executed by a processor causes the processor to: extract a sub-set of data from each image of an ensemble of ultrasound images having diversity in an ensemble dimension; construct a data matrix with the extracted data, wherein the data matrix has a dimension of space versus the ensemble dimension; compute a covariance matrix with the data matrix; identify largest eigenvalue(s) and a corresponding eigenvector(s) in the covariance matrix; compute a coherent signal projection matrix with the eigenvector(s) of the identified eigenvalue(s); filter the data matrix with the coherent signal projection; compound the filtered data across the ensemble dimension; and generate a decluttered ultrasound image with the compounded filtered data.