AUTOMATED IMAGE ANALYSIS FOR MAGNETIC RESONANCE IMAGING

• US 20100284595A1
• Filed: 02/02/2009
• Published: 11/11/2010
• Est. Priority Date: 01/31/2008
• Status: Active Grant

• Current Assignee: Johns Hopkins University
• Original Assignee: Johns Hopkins University
• Inventors: Miller, Michael I., Oishi, Kenichi, Mori, Susumu, Zhang, Jiangyang
• Granted Patent: US 8,731,256 B2
• Time in Patent Office: Days
• US Class Current: 382/131
• CPC Class Codes: G01R 33/56 Image enhancement or correction, e.g. subtraction or averaging techniques , e.g. improvement of signal-to-noise ratio and resolution
• CPC Class Codes: G01R 33/5608 Data processing and visualization specially adapted for MR, e.g. for feature analysis and pattern recognition on the basis of measured MR data, segmentation of measured MR data, edge contour detection on the basis of measured MR data, for enhancing measured MR data in terms of signal-to-noise ratio by means of noise filtering or apodization, for enhancing measured MR data in terms of resolution by means for deblurring, windowing, zero filling, or generation of gray-scaled images, colour-coded images or images displaying vectors instead of pixels image data processing or generation, in general G06T
• CPC Class Codes: G01R 33/56341 Diffusion imaging
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• CPC Class Codes: G06T 2207/30016 Brain
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/30 Determination of transform parameters for the alignment of images, i.e. image registration

Abstract

A magnetic resonance imaging (MRI) system, comprising: a magnetic resonance imaging scanner configured to generate a plurality of signals for forming at least one magnetic resonance image of a soft tissue region from a subject under observation, wherein the at least one magnetic resonance image provides at least one integrating feature to facilitate automatic segmentation; a signal processing system in communication with the magnetic resonance imaging scanner to receive the plurality of signals; and a data storage unit in communication with the signal processing system, wherein the data storage unit contains at least one template corresponding to the soft tissue region, wherein the signal processing system is adapted to process the plurality of signals received from the magnetic resonance imaging scanner to automatically perform segmentation for the soft tissue region of the subject under observation by utilizing the at least one template and the at least one integrating feature.

Claims

1. A magnetic resonance imaging (MRI) system, comprising:

a magnetic resonance imaging scanner configured to generate a plurality of signals for forming at least one magnetic resonance image of a soft tissue region from a subject under observation, wherein said at least one magnetic resonance image provides at least one integrating feature to facilitate automatic segmentation;

a signal processing system in communication with said magnetic resonance imaging scanner to receive said plurality of signals; and

a data storage unit in communication with said signal processing system, wherein said data storage unit contains at least one template corresponding to said soft tissue region,

wherein said signal processing system is adapted to process said plurality of signals received from said magnetic resonance imaging scanner to automatically perform segmentation for said soft tissue region of said subject under observation by utilizing said at least one template and said at least one integrating feature.

2. The magnetic resonance imaging system according to claim 1, wherein said at least one integrating feature provided by said at least one magnetic resonance image facilitates discerning said soft tissue region by at least one of:

a correction of intensity homogeneity;

a plurality of contrasts;

a suppression of at least one tissue component;

a quantification of at least one magnetic resonance (MR) contrast parameter, or variations thereof.

3. The magnetic resonance imaging system according to claim 1, wherein said signal processing system is at least partially incorporated within a structure housing said magnetic resonance scanner.

4. The magnetic resonance imaging system according to claim 1, wherein said signal processing system is at least partially incorporated in a workstation that is structurally separate and in communication with said magnetic resonance imaging scanner.

5. The magnetic resonance imaging system according to claim 1, wherein said at least one template is obtained from magnetic resonance imaging.

6. The magnetic resonance imaging system according to claim 1, wherein said at least one template incorporates at least one magnetic resonance image from at least one subject that is different from said subject under observation.

7. The magnetic resonance imaging system according to claim 1, wherein said at least one template incorporates at least one magnetic resonance image from a previous scan of said subject under observation.

8. A workstation, comprising:

a preprocessing engine to receive and preprocess magnetic resonance signals, from a magnetic resonance scanner, wherein said magnetic resonance signals form at least one magnetic resonance image showing a soft tissue region from at least one subject under observation in said magnetic resonance scanner; and

a normalizing engine to automatically segment said soft tissue region by normalizing the formed at least one magnetic resonance image based on at least one template corresponding to said soft tissue region,

wherein said at least one magnetic resonance image provides at least one integrating feature to facilitate automatic segmentation.

9. The workstation according to claim 8, wherein said at least one integrating feature provided by said at least one magnetic resonance image facilitates automatic segmentation of said soft tissue region by at least one of:

a correction of intensity homogeneity;

a plurality of contrasts;

a suppression of at least one tissue component;

a quantification of at least one magnetic resonance (MR) parameter, or variations thereof.

10. The workstation according to claim 8, wherein said at least one template and the at least one magnetic resonance image being normalized are based on at least one common magnetic resonance (MR) contrast.

11. The workstation according to claim 8, wherein said normalizing engine is constructed to implement at least one transformation using an algorithm of Large Deformation Diffeomorphic Metric Mapping (LDDMM).

12. The workstation according to claim 8, further comprising:

an analyzing engine to perform quantitative analysis associated with the segmented soft tissue region.

13. The workstation according to claim 12, wherein said analysis is related to at least one transformation matrix used by said normalizing engine.

14. The workstation according to claim 8, further comprising:

a diagnosing engine constructed to compare the results from said quantitative analysis with a knowledge database.

15. The workstation according to claim 14, wherein said diagnosing engine is constructed to compare by correlating statistics of said quantitative analysis and said knowledge database.

16. The workstation according to claim 14, wherein said knowledge database comprises data from a normal population and a clinical population.

17. A method of automatic segmentation, comprising:

receiving magnetic resonance signals thr forming at least one magnetic resonance image showing a soft tissue region from a subject, wherein said at least one magnetic resonance image provides at least one integrating feature to facilitate automatic segmentation;

preprocessing said magnetic resonance signals to generate at least one magnetic resonance image by utilizing said at least one integrating feature;

receiving at least one template corresponding to said soft tissue region;

normalizing the generated at least one magnetic resonance image based on said at least one template to generate a segmented view of said soft tissue region.

18. The method according to claim 17, wherein said at least one integrating feature provided by said at least one magnetic resonance image facilitates automatic segmentation by at least one of:

a correction of intensity homogeneity;

a plurality of contrasts;

a suppression of at least one tissue component;

a quantification of at least one magnetic resonance (MR) parameter, and variations thereof.

19. The method according to claim 18, wherein said plurality of contrasts relates to at least one of a relaxation time, a diffusion property, a perfusion property, a blood oxygenation level dependent property, a proton-density property, and variations thereof.

20. The method according to claim 17, wherein said at least one template and the at least one magnetic resonance image are based on at least one common magnetic resonance (MR) contrast.

21. The method according to claim 17, wherein said template contains information identifying said soft tissue region.

22. The method according to claim 21, wherein said information comprises at least one probabilistic map of said soft tissue region.

23. The method according to claim 17, wherein said normalizing comprises at least one transformation using an algorithm of Large Deformation Diffeomorphic Metric Mapping (LDDMM).

24. The method according to claim 17, further comprising:

using at least one existing transformation matrix to normalize a new magnetic resonance image.

25. The method according to claim 17, further comprising;

analyzing data associated with the segmented soft tissue region.

26. The method according to claim 25, wherein said data comprises at least one transformation matrix used during said normalizing.

27. The method according to claim 25, further comprising:

diagnosing by comparing the results from said analyzing with a knowledge database.

28. The method according to claim 27, wherein said comparing comprises correlating based on statistics associated with said analyzing and said knowledge database.

29. The method according to claim 27, wherein said knowledge database comprises data from a normal population and a clinical population.

30. The method according to claim 29, further comprising:

adding the results from said analyzing to said knowledge database if said diagnosing is positive.

31. A computer-readable medium comprising software, which when executed by a computer system, causes the computer system to perform operations to perform the method of 17.