Spatio-temporal treatment of noisy images using brushlets

• US 7,542,622 B1
• Filed: 06/02/2004
• Issued: 06/02/2009
• Est. Priority Date: 06/02/2003
• Status: Expired due to Fees

• Current Assignee: Trustees Of Columbia University In The City Of New York (Columbia University)
• Original Assignee: Trustees Of Columbia University In The City Of New York (Columbia University)
• Inventors: Angelini, Elsa D., Laine, Andrew F.
• Primary Examiner(s): Bella; Matthew C
• Assistant Examiner(s): Ge; Yuzhen
• Application Number:
  US10/858,764
• Time in Patent Office: 1,826 Days
• Field of Search: None
• US Class Current: 382/275
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/10076 4D tomography; Time-sequential 3D tomography
• CPC Class Codes: G06T 2207/10132 Ultrasound image
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/20056 Discrete and fast Fourier transform, [DFT, FFT]
• CPC Class Codes: G06T 2207/30048 Heart; Cardiac
• CPC Class Codes: G06T 5/10 using non-spatial domain filtering
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06V 10/30 Noise filtering
• CPC Class Codes: G06V 2201/03 Recognition of patterns in medical or anatomical images

Abstract

Treatment and mitigation or reduction of noise effects in noisy image data and data sets is described. Various aspects include treatment of noisy data with brushlet transforms and thresholding operations along with a favorable sequence of spatial and temporal processing and thresholding. Hard and minimax thresholding operators mitigate the noise in the image data. In medical applications this can be useful in removing noise that impairs diagnosis and treatment of patient conditions. In one application, cardiac function is better studied and understood through improved imaging of the heart and cardiac structures. In an exemplary case, a favorable sequence including spatial filtering using a brushlet filter, spatial thresholding of brushlet coefficients, then temporal filtering (first in the time domain then in the frequency domain) and thresholding of temporal coefficients yields an acceptable denoised image data set.

Claims

1. A method for producing denoised multi-dimensional spatio-temporal image data comprising a plurality of multi-dimensional frames taken at different times, the method comprising:

performing multi-dimensional, spatial brushlet filtering, in the frequency domain, of the plurality of multi-dimensional frames to generate a set of spatial brushlet coefficients;

thresholding the set of spatial brushlet coefficients to produce a set of spatially-denoised image data corresponding to the multi-dimensional spatio-temporal image data;

performing a first temporal brushlet filtering, in the time domain, of the spatially-denoised spatio-temporal image data to generate a first set of temporal brushlet coefficients;

thresholding the first set of temporal brushlet coefficients to produce a first set of spatio-temporally denoised image data corresponding to the multi-dimensional spatio-temporal image data;

performing a second temporal brushlet filtering, in the frequency domain, of the first set of spatio-temporally denoised image data to generate a second set of temporal brushlet coefficients;

thresholding the second set of temporal brushlet coefficients to produce a second set of spatio-temporally denoised image data corresponding to the multi-dimensional spatio-temporal image data; and

outputting the second set of spatio-temporally denoised image data;

wherein the above said filtering steps are performed using a computer system.

2. The method of claim 1, wherein the multi-dimensional spatio-temporal image data comprises three-dimensional spatio-temporal image data comprising two-dimensional frames taken at different instants in time.

3. The method of claim 1, wherein the multi-dimensional spatio-temporal image data comprises four-dimensional spatio-temporal image data comprising three-dimensional frames taken at different instants in time.

4. The method of claim 1, wherein producing denoised image data comprises removing speckle noise out of the image data.

5. The method of claim 1, wherein producing denoised image data comprises denoising noisy ultrasound image data.

6. The method of claim 1, wherein any of the thresholding steps comprises performing a hard thresholding operation on the respective coefficients.

7. The method of claim 1, wherein any of the thresholding steps comprises performing a Minimax thresholding operation on the respective coefficients.

8. The method of claim 1, further comprising segmentation of an image feature following denoising of the image data.

9. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, cause the processor to perform a method for producing denoised multi-dimensional spatio-temporal image data comprising a plurality of multi-dimensional frames taken at different times, the method comprising:

performing multi-dimensional, spatial brushlet filtering, in the frequency domain, of the plurality of multi-dimensional frames to generate a set of spatial brushlet coefficients;

thresholding the set of spatial brushlet coefficients to produce a set of spatially-denoised image data corresponding to the multi-dimensional spatio-temporal image data;

performing a first temporal brushlet filtering, in the time domain, of the spatially-denoised spatio-temporal image data to generate a first set of temporal brushlet coefficients;

thresholding the first set of temporal brushlet coefficients to produce a first set of spatio-temporally denoised image data corresponding to the multi-dimensional spatio-temporal image data;

performing a second temporal brushlet filtering, in the frequency domain, of the first set of spatio-temporally denoised image data to generate a second set of temporal brushlet coefficients;

thresholding the second set of temporal brushlet coefficients to produce a second set of spatio-temporally denoised image data corresponding to the multi-dimensional spatio-temporal image data; and

outputting the second set of spatio-temporally denoised image data.

10. The computer-readable medium of claim 9, wherein the multi-dimensional spatio-temporal image data comprises three-dimensional spatio-temporal image data comprising two-dimensional frames taken at different instants in time.

11. The computer-readable medium of claim 9, wherein the multi-dimensional spatio-temporal image data comprises four-dimensional spatio-temporal image data comprising three-dimensional frames taken at different instants in time.

12. The computer-readable medium of claim 9, wherein producing denoised image data comprises removing speckle noise out of the image data.

13. The computer-readable medium of claim 9, wherein producing denoised image data comprises denoising noisy ultrasound image data.

14. The computer-readable medium of claim 9, wherein any of the thresholding steps of the method comprises performing a hard thresholding operation on the respective coefficients.

15. The computer-readable medium of claim 9, wherein any of the thresholding steps of the method comprises performing a Minimax thresholding operation on the respective coefficients.

16. The computer-readable medium of claim 9, wherein the method further comprises segmentation of an image feature following denoising of the image data.