METHOD AND DEVICE FOR PROCESSING AND PRESENTING MEDICAL IMAGES

• US 20090257657A1
• Filed: 04/08/2009
• Published: 10/15/2009
• Est. Priority Date: 04/09/2008
• Status: Abandoned Application

• Current Assignee: AGFA Healthcare NV (Agfa-Gevaert N.V.), Barco NV, Ibbt Vzw, Vrije Universiteit Brussel
• Original Assignee: AGFA Healthcare NV (Agfa-Gevaert N.V.), Barco NV, Ibbt Vzw, Vrije Universiteit Brussel
• Inventors: SULIGA, Marek, BREUCQ, Catherine, TEMMERMANS, Frederik, DEKLERCK, Rudi, DE MEY, Johan, BEHIELS, Gert, DEWAELE, Piet
• Application Number:
  US12/420,483
• Publication Number:
  US 20090257657A1
• Time in Patent Office: Days
• US Class Current: 382/195
• CPC Class Codes: G06T 2207/10072 Tomographic images
• CPC Class Codes: G06T 2207/10116 X-ray image
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 3/147 using affine transformations
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/35 using statistical methods
• CPC Class Codes: G06T 7/41 based on statistical description of texture

Abstract

The present invention relates to a method for processing and presenting at least a first image (102) and a second image (103), these images being digital medical images. A first step in the method is performing image registration between the first image (102) and the second image (103) to generate a pixel-level mapping between both images. The registration takes into account a region-level correspondence between the first image (102) and the second image (103) A second step is presenting the first image (102) and the second image (103) simultaneously on a display (103). A third step is presenting a first magnification (104) of a region of interest (106) in the first image (102) and a second magnification (105) of a corresponding region (108) in the second image (103). The corresponding region (108) in the second image (103) being determined based on the pixel-level mapping.

Claims

1. A method for processing and presenting at least a first image (102) and a second image (103), said first image (102) and said second image (103) being digital medical images, said method comprising the steps of:

performing image registration between said first image (102) and said second image (103) to generate a pixel-level mapping between said first image (102) and said second image (103);

presenting said first image (102) and said second image (103) simultaneously on a display (101);

detecting a region of interest (106) in said first image (102); and

presenting a first magnification (104) of said region of interest (106) in said first image (102) and a second magnification (105) of a corresponding region (108) in said second image (103), said corresponding region (108) in said second image (103) being determined based on said pixel-level mapping; and

determining for said pixel-level mapping a region-level correspondence between said first image (102) and said second image (103).

2. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said regions of interest comprise one or more of the following:

a micro-calcification;

a cluster of micro-calcifications; and

a mass.

3. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said mapping of said regions of interest is based on one or more of the following:

Markov Random Field graph matching;

a feature-based comparison algorithm; and

a probabilistic comparison algorithm.

4. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said image registration comprises one or more of the following:

mapping the boundaries of said first image (102) to the boundaries in said second image (103) or vice versa;

interpolating the interior part within said boundaries of said first image (102) and/or said second image (103);

additional mapping of said corresponding regions of interest; and

interpolating the remaining areas in the image.

5. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said image registration comprises one or more of the following:

vertical alignment of said first image (102) and said second image (103);

affine transformation of said first image (102) and/or said second image (103);

higher-order deformable or elastic transformation of the first image (102) to the second image (103) or vice versa.

6. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said image registration comprises a higher-order deformable mapping like a Thin Plate Spline or TPS mapping.

7. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said image registration comprises mapping boundaries of an object in said first image (102) to boundaries of an object in said second image on the basis of characteristic feature points and/or equidistantly sampled points.

8. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1, wherein said method further comprises the step of determining statistical information (111) and/or measurements related to said region (106) in said first image (102) or a first registered image and said corresponding region (108) in said second image (103) or a second registered image.

9. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 8 wherein said method further comprises the step of presenting said statistical information (111, 112) and/or said measurements in conjunction with said first magnification (104) and/or said second magnification (105).

10. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said method further comprises the step of receiving user input for selecting said region (106) in said first image (102) or a first registered image.

11. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said method further comprises the step of iteratively presenting a region in said first image (102) or a first registered image and a corresponding region in said second image (103) or a second registered image, said region in said first image (102) or said first registered image being automatically selected from a list of regions in said first image (102) or said first registered image.

12. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said method further comprises the step of automatically moving said region in said first image (102) or a first registered image and presenting said corresponding region in said second image (103) or a second registered image.

13. The method for processing and presenting at least a first image (102) and a second image (103) according to claim 1 wherein said medical digital images are one or more of the following:

temporal mammography images;

a left and a right breast mammography image;

Cranio-Caudal projections of a breast;

Medio-Lateral Oblique projections of a breast;

a registered version of said first image (102) and/or a registered version of said second image (103) obtained by applying said pixel-level mapping to said first image (102) and/or said second image (103) respectively.

14. A device for presenting at least a first image (102) and a second image (103), said images being digital medical images and said device comprising:

means for performing image registration between said first image (102) and said second image (103) to generate a pixel-level mapping between said first image (102) and said second image (103);

means for presenting said first image (102) and said second image (103) simultaneously on a display (101);

means for receiving information related to the detection of a region of interest (106) in said first image (102);

means for presenting a first magnification (104) of said region of interest (106) in said first image (102) and presenting a second magnification (105) of a corresponding region (108) in said second image (103) wherein said corresponding region is determined based on said pixel-level mapping, and

means for determining for said pixel-level mapping a region-level correspondence between said first image (102) and said second image (103).