Systems and methods for noise suppression in microvessel ultrasound imaging

1 . A method for generating an image that depicts microvessels in a subject using an ultrasound system, the steps of the method comprising:

(a) accessing with a computer system, ultrasound data acquired from a subject with an ultrasound system, wherein the ultrasound data comprise image frames obtained at a plurality of different time points;

(b) generating spatiotemporal matrix data with the computer system by reformatting the ultrasound data as a Casorati matrix;

(c) determining image frames that meet a similarity threshold;

(d) generating local spatiotemporal clutter-filtered data with the computer system by inputting the determined image frames that meet the similarity threshold to a clutter filter;

(e) determining pixels related to flow-signal and background signal in the local spatiotemporal clutter-filtered data;

(f) generating a synthetic noise image with the computer system based on replacing select flow-signal pixels in the local spatiotemporal clutter-filtered data with background signal pixel values;

(g) estimating background noise field data from the synthetic noise image using the computer system; and

(h) generating a noise-suppressed image with the computer system by normalizing the image using the background noise field data, wherein the noise-suppressed image depicts microvessels in the subject.

2 . The method as recited in claim 1 , wherein replacing select flow-signal pixels with background signal pixel values includes randomly selecting the background signal pixels.

3 . The method as recited in claim 1 , wherein values of the background signal pixels are depth-matched to be consistent with time gain compensation settings for the ultrasound system.

4 . The method as recited in claim 1 , further comprising generating synthetic noise images for all obtained image frames.

5 . The method as recited in claim 4 , wherein estimating background noise field data includes averaging the generated synthetic noise images.

6 . The method as recited in claim 1 , wherein the similarity threshold for image frames is equal to or greater than 0.8.

7 . The method as recited in claim 1 , wherein the background noise field data are estimated by computing a low-rank approximation of the synthetic noise image.

8 . The method as recited in claim 7 , wherein the low-rank approximation is based on a singular value decomposition.

9 . The method as recited in claim 1 , wherein generating the synthetic noise image comprises replacing each flow pixel in the local spatiotemporal clutter-filtered data with a noise pixel randomly selected from different image frames in the ultrasound data.

10 . The method as recited in claim 1 , wherein the clutter filter is implemented with a singular value decomposition.

11 . A system for generating an image that depicts microvessels in a subject, the system comprising:

an ultrasound system configured to acquire ultrasound data from a subject;

a computer system configured to:

(i) access the ultrasound data acquired from the subject, wherein the ultrasound data comprise image frames obtained at a plurality of different time points;

(ii) generate spatiotemporal matrix data by reformatting the ultrasound data as a Casorati matrix;

(iii) determine image frames that meet a similarity threshold;

(iv) generate local spatiotemporal clutter-filtered data by inputting the determined image frames that meet the similarity threshold to a clutter filter;

(v) determine pixels related to flow-signal and background signal in the local spatiotemporal clutter-filtered data;

(vi) generate a synthetic noise image based on replacing select flow-signal pixels in the local spatiotemporal clutter-filtered data with background signal pixel values;

(vii) estimate background noise field data from the synthetic noise image; and

(viii) generate a noise-suppressed image by normalizing the image using the background noise field data, wherein the noise-suppressed image depicts microvessels in the subject.

12 . The system as recited in claim 11 , wherein the computer system is further configured to replace select flow-signal pixels with background signal pixel values by randomly selecting the background signal pixels.

13 . The system as recited in claim 11 , wherein the computer system is further configured to depth-match values of the background signal pixels to be consistent with time gain compensation settings for the ultrasound system.

14 . The system as recited in claim 11 , wherein the computer system is further configured to generate synthetic noise images for all obtained image frames.

15 . The system as recited in claim 14 , wherein the computer system is further configured to estimate background noise field data by averaging the generated synthetic noise images.

16 . The system as recited in claim 11 , wherein the similarity threshold for image frames is equal to or greater than 0.8.

17 . The system as recited in claim 11 , wherein the computer system is further configured to estimate the background noise field data by computing a low-rank approximation of the synthetic noise image.

18 . The system as recited in claim 17 , wherein the low-rank approximation is based on a singular value decomposition.

19 . The system as recited in claim 11 , wherein the computer system is further configured to generate the synthetic noise image by replacing each flow pixel in the local spatiotemporal clutter-filtered data with a noise pixel randomly selected from different image frames in the ultrasound data.

20 . The system as recited in claim 11 , wherein the computer system is further configured to implement the clutter filter with a singular value decomposition.