Method and system for detection of interval change in temporally sequential chest images

• US 5,359,513 A
• Filed: 11/25/1992
• Issued: 10/25/1994
• Est. Priority Date: 11/25/1992
• Status: Expired due to Term

• Current Assignee: ARCH Development Corporation (University of Chicago)
• Original Assignee: ARCH Development Corporation (University of Chicago)
• Inventors: Doi, Kunio, Kano, Akiko
• Primary Examiner(s): McElheny, Jr., Donald E.
• Assistant Examiner(s): Tran, Khai
• Application Number:
  US07/981,471
• Time in Patent Office: 699 Days
• Field of Search: 364/413.13, 364/413.23, 364/413.29, 382/6, 382/44, 382/54
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/20224 Image subtraction
• CPC Class Codes: G06T 3/153 using elastic snapping
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction
• CPC Class Codes: Y10S 128/922 including image analysis

Abstract

A computerized scheme for the analysis of digitized medical images which provides enhancement of interval changes occurring in a pair of temporally sequential images. In the method and apparatus of the invention, a pair of images are digitized and then are subjected to image registration including a nonlinear warping of one of the images so that corresponding locations in the two images are aligned with each other. Subsequent to image registration, a subtraction process is performed in order to generate a difference between the warped and unwarped images. In this manner, slight opacities which are only present in the later image may be detected based on the subtraction of the registered images.

Claims

1. A method for computerized analysis of temporally sequential images, comprising:

obtaining first and second images of a patient;

aligning corresponding portions of said first and second images in relation to corresponding anatomical features present in said first and second images, including performing cross-correlation mapping to produce a plurality of shift values indicative of differences in locations between corresponding anatomical features of said first and second images with respect to corresponding pixel locations of said first and second images, mathematically fitting a predetermined surface to said shift values for correcting said shift values in order to compensate for errors introduced into said cross-correlation mapping, and shifting the anatomical features in one of said first and second images based on the corrected shift values;

performing subtraction processing between said first and second images; and

determining differences between said first and second images indicative of abnormal changes which have occurred in said patient during the time interval between the taking of said first and second images.

2. The method according to claim 1, wherein said step of performing cross-correlation mapping comprises:

selecting a template region of interest (ROI) in one of said first and second images and a search area ROI in the other image, wherein said search area ROI is larger than said template ROI;

determining cross-correlation values between said template region in said one image and plural subregions at different positions within said search area ROI in said other image;

determining which subregion of said search area ROI in said other image results in a maximum cross-correlation value with respect to said template ROI in said one region; and

determining said shift values for said template ROI based on the difference in location between the position of the template region in said one image and the position in the other image of said subregion of said search area ROI which results in the maximum cross-correlation value with respect to said template ROI.

3. The method according to claim 2, wherein said step of fitting said predetermined surface to said cross-correlation mapping of shift values comprises:

using a two-dimensional nth order polynomial function with weighting coefficients ranging from 0.0 to 1.0.

4. The method according to claim 3, wherein said step of fitting said predetermined surface comprises:

selecting said weighting coefficients based on the maximum cross-correlation values determined in said step of cross-correlation mapping.

5. The method according to claim 3, wherein said step of using a two-dimensional nth order polynomial function comprises:

selecting a weighting coefficient close to or equal to 1.0 for highly correlated portions of the first and second images obtained by said cross-correlation mapping, and 0.0 or a value close to 0.0 as a weighting coefficient for portions of the first and second images which have little correlation.

6. The method according to claim 3, further comprising:

using a look-up table for generating said weighting coefficients.

7. The method according to claim 1, further comprising:

displaying an output image representative of a subtraction image obtained by said subtraction processing.

8. A system for computerized analysis of temporally sequential digitized chest images, comprising:

means for obtaining first and second digitized chest images of a patient;

means for aligning corresponding portions of said first and second images in relation to corresponding anatomical features present in said first and second images, including means for performing cross-correlation mapping to produce a plurality of shift values indicative of differences in location between corresponding anatomical features of said first and second images with respect to corresponding pixel locations of said first and second images, means for mathematically fitting a predetermined surface to said shift values for correcting said shift values in order to compensate for errors introduced into said cross-correlation mapping, and means for shifting the anatomical features in one of said first and second images based on the corrected shift values;

means for performing subtraction processing between said first and second images, whereby differences between said first and second images are determined which indicate abnormal changes which have occurred in said patient during the time interval between the taking of said first and second images.

9. The system according to claim 8, wherein said means for performing cross-correlation mapping comprise:

means for selecting a template region of interest (ROI) in one of said first and second images and a search area ROI in the other image, wherein said search area ROI is larger than said template ROI;

means for determining cross-correlation values between said template ROI in said one image and plural subregions at different positions within said search area ROI in said other image;

means for determining which subregion of said search area ROI in said other image results in a maximum cross-correlation value with respect to said template ROI in said one region; and

means for determining said shift values for said template ROI based on the difference in location between the position of the template region in said one image and the position in the other image of said subregion of said search area ROI which results in the maximum cross-correlation value with respect to said template ROI.

10. The system according to claim 9, wherein said means for fitting said predetermined surface comprises:

means for applying a two-dimensional nth order polynomial function to said cross-correlation mapping of said shift values using weighting coefficients ranging from 0.0 to 1.0.

11. The system according to claim 10, wherein said means for fitting said predetermined surface comprises:

means for selecting said weighting coefficients based on the maximum cross-correlation values.

12. The system according to claim 10, wherein said means for applying a two-dimensional nth order polynomial function comprises:

means for selecting a weighting coefficient close to or equal to 1.0 for highly correlated portions of the first and second images obtained by said cross-correlation mapping, and 0.0 or a value close to 0.0 as a weighting coefficient for portions of the first and second images which have little correlation.

13. The system according to claim 10, further comprising:

look-up table means for generating said weighting coefficients.

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

means for displaying an output image representative of a subtraction image obtained by said subtraction processing.