METHOD AND APPARATUS FOR ELIMINATING NOISE IN MAGNETIC RESONANCE IMAGES

• US 20140203808A1
• Filed: 01/17/2014
• Published: 07/24/2014
• Est. Priority Date: 01/18/2013
• Status: Active Grant

• Current Assignee: Industry Academic Cooperation Foundation
• Original Assignee: Industry Academic Cooperation Foundation
• Inventors: HWANG, Dosik
• Granted Patent: US 9,759,796 B2
• Time in Patent Office: Days
• US Class Current: 324/318
• CPC Class Codes: G01R 33/565 Correction of image distortions, e.g. due to magnetic field inhomogeneities
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/30004 Biomedical image processing
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction
• CPC Class Codes: G06T 5/70 Denoising; Smoothing

Abstract

A method and an apparatus for eliminating noise in magnetic resonance images are disclosed. An aspect of the invention provides a method for eliminating noise in a magnetic resonance image that includes: obtaining a multiple number of magnetic resonance images sequentially within a preset time range for the same part; calculating probability information, which relates to the probability of a first pixel and a second pixel of the multiple magnetic resonance images belonging to the same tissue, by using decay information, which relates to a decay in signal intensity values for the first pixel and the second pixel according to time; and determining a new signal intensity value for the first pixel and the second pixel based on a weighted average of signal intensity values of the first pixel and the second pixel according to the probability information.

Claims

1. A method for eliminating noise in a magnetic resonance image, the method comprising:

obtaining a plurality of magnetic resonance images sequentially within a preset time range for a same part;

calculating probability information relating to a probability of a first pixel and a second pixel of the plurality of magnetic resonance images belonging to a same tissue, by using decay information relating to a decay in signal intensity values for the first pixel and the second pixel according to time; and

determining a new signal intensity value for the first pixel and the second pixel based on a weighted average of signal intensity values of the first pixel and the second pixel according to the probability information.

2. The method of claim 1, wherein weights for the first and second pixels are determined according to the probability of the first pixel and the second pixel belonging to a same tissue.

3. The method of claim 2, wherein the weights are determined based on differences in signal intensity values of the first pixel and the second pixel.

4. The method of claim 1, wherein the determining comprises determining the new signal intensity value for the first and second pixels by calculating a weighted average of signal intensity values of the first pixel and the second pixel for all pixels in each of the plurality of magnetic resonance images.

10. A recorded medium readable by a computer, tangibly embodying a program of instructions executable to perform the method of claim 1.

5. A method for eliminating noise in a magnetic resonance image, the method comprising:

obtaining a plurality of magnetic resonance images sequentially within a preset time range for a same part;

deciding whether or not a first pixel and a second pixel of the plurality of magnetic resonance images belong to a same tissue, by using decay information relating to a decay in signal intensity values for the first pixel and the second pixel according to time; and

determining a new signal intensity value for the first pixel and the second pixel by calculating an average of signal intensity values of the first pixel and the second pixel, if the first pixel and the second pixel belong to a same tissue.

6. The method of claim 5, wherein the deciding comprises deciding that the first and second pixels belong to a same tissue if a difference in signal intensity values of the first pixel and the second pixel is between a first threshold and a second threshold.

7. The method of claim 6, wherein the deciding comprises deciding that the first and second pixels belong to a same tissue if the difference in signal intensity values is larger than the first threshold and smaller than the second threshold.

8. The method of claim 7, wherein the first threshold is selected according to noise level, and the second threshold is selected empirically.

9. A method for eliminating noise in a magnetic resonance image, the method comprising:

obtaining a plurality of magnetic resonance images sequentially within a preset time range for a same part;

calculating probability information relating to a probability of a first pixel and a second pixel of the plurality of magnetic resonance images belonging to a same tissue, by using signal intensity change information relating to a change in signal intensity values for the first pixel and the second pixel according to relaxation; and

determining a new signal intensity value for the first pixel and the second pixel based on a weighted average of signal intensity values of the first pixel and the second pixel according to the probability information.

11. An apparatus for eliminating noise in a magnetic resonance image, the apparatus comprising:

an MRI acquisition unit configured to obtain a plurality of magnetic resonance images sequentially within a preset time range for a same part;

a decay analysis unit configured to analyze decay information relating to a decay in signal intensity values for a first pixel and a second pixel of the plurality of magnetic resonance images according to time;

a same-tissue decision unit configured to calculate a probability of different pixels belonging to a same tissue by using the analyzed decay information; and

a filtering unit configured to determine a new signal intensity value for each of the pixels based on a weighted average of signal intensity values of each of the pixels by using the probability information.

12. The apparatus of claim 11, wherein the MRI acquisition unit obtains the plurality of magnetic resonance images in preset time intervals.

13. The apparatus of claim 11, wherein the same-tissue decision unit applies different weights on the first pixel and the second pixel according to a difference in signal intensity values of the first pixel and the second pixel.

14. An apparatus for eliminating noise in a magnetic resonance image, the apparatus comprising:

an MRI acquisition unit configured to obtain a plurality of magnetic resonance images sequentially within a preset time range for a same part;

a decay analysis unit configured to analyze decay information relating to a decay in signal intensity values for a first pixel and a second pixel of the plurality of magnetic resonance images according to time;

a same-tissue decision unit configured to identify pixels corresponding to a same tissue by using the analyzed decay information; and

a filtering unit configured to determine a new signal intensity value for pixels corresponding to a same tissue by calculating an average of signal intensity values of the pixels.