Iterative filter framework for medical images

• US 6,246,783 B1
• Filed: 09/17/1997
• Issued: 06/12/2001
• Est. Priority Date: 09/17/1997
• Status: Expired due to Fees

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Avinash, Gopal B.
• Primary Examiner(s): Patel, Jayanti K.
• Application Number:
  US08/932,430
• Time in Patent Office: 1,364 Days
• Field of Search: 382/100, 382/128-134, 382/168-172, 382/173, 382/190, 382/191-192, 382/199, 382/205, 382/206, 382/210, 382/211, 382/216, 382/217, 382/218, 382/254, 382/256, 382/257-266, 382/279, 382/282, 382/308, 382/265, 382/276, 378/1, 378/62-63, 250/370.08, 250/370.09, 250/370.1, 250/370.11, 345/425, 348/665-668, 358/450, 359/559, 600/300, 600/544, 600/587, 708/300, 708/320, 708/322, 736/02, 736/06
• US Class Current: 382/128
• CPC Class Codes: G06T 1/20 Processor architectures; Processor configuration, e.g. pipelining
• CPC Class Codes: G06T 5/20 using local operators

Abstract

Functional filter elements are linked in a multi-pass filter framework that forms part of an MRI system. The filter framework is configured by a set of parameters selected by the operator to perform smoothing and edge sharpening functions on acquired MR images. The functional filter elements used during each pass through the filter framework and the number of passes through it are configurable to obtain a wide variety of visual effects on medical images.

Claims

1. In a medical imaging system, the combination comprising:

means for acquiring image data from a patient;

means for reconstructing an image data set from the acquired image data; and

a filter framework for receiving the image data set and in response to a set of parameters selected by an operator, producing a filtered image data set;

wherein the filter framework comprises;

a) a smoothing filter;

b) a sharpening filter;

c) means responsive to the set of parameters for processing the image data set a plurality of passes through the filter framework to produce the filtered image data set, the means for processing being operable to determine the filters to be employed during each pass through the framework and the number of passes through the framework.

2. The medical imaging system as recited in claim 1 in which the filter framework includes means for combining a portion of the filtered image data with a portion of the image data to form the filtered image data set.

3. The medical imaging system as recited in claim 2 in which the portion of the filtered image data is determined by a scaling factor parameter α

and the portion of the image data is determined by the value 1-α

.

4. The medical imaging system as recited in claim 1 in which the filter framework includes:

segmentation means for separating the filtered image data into first region containing edge information and second regions containing substantially no edge information; and

means for combining the image data and the filtered image data in a manner determined by the first and second regions identified by the segmentation means.

5. The medical imaging system as recited in claim 1 in which the filters employ a 3×

1 filter kernel.

6. A method for producing a medical image, the steps comprising:

a) acquiring image data from a patient;

b) reconstructing an image data set from the acquired image data;

c) producing an intermediate image by filtering the image data set;

d) blending the intermediate image with the image data set to produce a filtered image data set;

by;

i) multiplying the intermediate image by a factor α

;

ii) multiplying the image data set by a factor (1-α

); and

iii) combining the results of steps i) and ii) to form the filtered image data set;

e) repeating steps c) and d) a set number of times using the filtered image data set in place of the image data set; and

f) producing the medical image from the filtered image data set.

7. The method as recited in claim 6 in which the blending in step d) is performed by:

multiplying the intermediate image by a scaling factor α

; and

multiplying the image data set by a factor (1-α

).

8. The method as recited in claim 7 in which the scaling factor α

is changed during each pass through step d).

9. The method as recited in claim 7 in which the scaling factor α

is a manually selected value.

10. The method as recited in claim 6 in which step e) is performed a plurality of times.

11. The method as recited in claim 6 in which the filtering in step c) includes using a smoothing filter during one pass.

12. The method as recited in claim 6 in which the filtering in step c) includes using a sharpening filter during one pass.

13. The method as recited in claim 6 in which the filtering in step c) includes using a directional sharpening and segmentation process during the last process.

14. The method as recited in claim 6 in which the filtering in step c) includes using a smoothing filter during one pass and a sharpening filter during another pass.

15. The method as recited in claim 6 which includes combining the filtered image data set with the image data set before producing the medical image therefrom in step f).

16. The method as recited in claim 15 which includes segmenting the filtered image data set into a first set of regions having strong edges and a second set of regions having weak edges, and the image data set is combined with the second set of regions.

17. The method as recited in claim 15 in which the data in the image data set corresponding to the second set of regions replaces the data in the filtered image data set corresponding to the second set of regions.