Image registration using locally-weighted fitting

• US 7,855,723 B2
• Filed: 03/21/2006
• Issued: 12/21/2010
• Est. Priority Date: 03/21/2006
• Status: Active Grant

• Current Assignee: Biosense Webster Incorporated (Johnson & Johnson)
• Original Assignee: Biosense Webster Incorporated (Johnson & Johnson)
• Inventors: Patt, Zafrir, Kitlaro, Ziv, Tal, Roy, Preiss, Assaf
• Primary Examiner(s): Tung; Kee M
• Assistant Examiner(s): Crawford; Jacinta
• Application Number:
  US11/385,060
• Publication Number:
  US 20070223794A1
• Time in Patent Office: 1,736 Days
• Field of Search: 382/128, 382/293, 382/294, 382/312, 345/419
• US Class Current: 345/419
• CPC Class Codes: A61B 5/0035 adapted for acquisition of images from more than one imaging mode, e.g. combining MRI and optical tomography
• CPC Class Codes: A61B 5/0044 for the heart
• CPC Class Codes: A61B 5/0073 by tomography, i.e. reconstruction of 3D images from 2D projections A61B5/0066 takes precedence
• CPC Class Codes: A61B 8/12 in body cavities or body tracts, e.g. by using catheters catheters per se A61M25/00
• CPC Class Codes: A61B 8/4254 using sensors mounted on the probe
• CPC Class Codes: G06T 7/33 using feature-based methods
• CPC Class Codes: G06V 10/757 Matching configurations of points or features

Abstract

A computer-implemented method for image registration includes accepting a three-dimensional (3-D) image of an organ. The image includes a first plurality of image points with respective image coordinates. A map defining a distribution of values of a physiological parameter over the organ is accepted. The map includes a second plurality of map points with respective map coordinates. Confidence levels are associated with one or more of the map coordinates and image coordinates. Weights are assigned to at least some of the map points and image points responsively to the confidence levels of the respective map coordinates and image coordinates. The 3-D image is fitted with the map by calculating a geometrical transformation between the map coordinates and the image coordinates based on the weights.

Claims

1. A computer-implemented method for image registration, comprising:

accepting on a computer a three-dimensional (3-D) image of an organ, the image comprising a first plurality of image points with respective image coordinates;

accepting on the computer a map defining a distribution of values of a physiological parameter over the organ, the map comprising a second plurality of map points with respective map coordinates, said second plurality of map points being sensed values of the physiological parameters;

associating respective confidence levels with one or more of the map coordinates and image coordinates;

assigning weights to at least some of the map points and image points responsively to the confidence levels of the respective map coordinates and image coordinates;

fitting the 3-D image with the map by calculating a geometrical transformation between the map coordinates and the image coordinates based on the weights; and

displaying on the computer the fitted 3D image with the map.

2. The method according to claim 1, wherein accepting the map comprises measuring the values of the physiological parameter at the respective map coordinates using a position tracking system.

3. The method according to claim 2, wherein measuring the values and the map coordinates comprises moving a probe comprising a position sensor and a sensor measuring the physiological parameter over multiple locations in the organ, and performing respective multiple measurements of the physiological parameter and of a position of the position sensor at the locations.

4. The method according to claim 1, and comprising superimposing the values of the physiological parameter onto the 3-D image using the geometrical transformation, and displaying the superimposed values and 3-D image to a user.

5. The method according to claim 1, wherein fitting the 3-D image with the map comprises defining distances between at least some of the image points and respective map points, applying the weights assigned to the respective map points and image points to the distances to produce weighted distances, and calculating a cumulative weighted distance metric based on the distances, the metric indicative of an accuracy of the geometrical transformation in fitting the 3-D image with the map.

6. The method according to claim 5, wherein applying the weights comprises multiplying a square of the distances by the respective weights to produce the weighted distances, and wherein calculating the cumulative weighted distance metric comprises summing the weighted distances.

7. The method according to claim 5, wherein fitting the 3-D image with the map further comprises modifying the geometrical transformation so as to minimize the cumulative weighted distance metric.

8. The method according to claim 1, wherein fitting the 3-D image with the map comprises identifying an anatomical landmark external to the organ that appears in the 3-D image, acquiring a respective map point at a location of the anatomical landmark, and registering an image coordinate of the anatomical landmark in the 3-D image with a map coordinate of the acquired map point.

9. The method according to claim 1, wherein associating the confidence levels comprises accepting the confidence levels from a user.

10. The method according to claim 1, wherein associating the confidence levels comprises automatically determining the confidence levels based on the map points and image points.

11. The method according to claim 10, wherein automatically determining the confidence levels comprises detecting a distortion in at least one of the map coordinates and image coordinates and assigning the confidence levels responsively to a level of the distortion.

12. The method according to claim 10, wherein automatically determining the confidence levels comprises assigning the confidence levels responsively to at least one of:

a density of the respective one or more of the map coordinates in the map;

an electrical tissue characteristic in a vicinity of the respective map coordinates and image coordinates;

a physiological tissue characteristic in the vicinity of the respective map coordinates and image coordinates;

a distance of the respective map coordinates and image coordinates from a registration landmark identified in both the map and the 3-D image; and

a physiological characteristic of the registration landmark.

13. The method according to claim 1, wherein accepting the 3-D image comprises accepting at least one of a solid model, a surface model and a wire-frame model of the organ defined by at least some of the image points.

14. Apparatus for image registration, comprising:

an interface, which is arranged to accept a three-dimensional (3-D) image of an organ, the image comprising a first plurality of image points with respective image coordinates, and to further accept a map defining a distribution of values of a physiological parameter over the organ, the map comprising a second plurality of map points with respective map coordinates, said second plurality of map points being sensed values of the physiological parameters; and

a processor, which is arranged to associate respective confidence levels with one or more of the map coordinates and image coordinates, to assign weights to at least some of the map points and image points responsively to the confidence levels of the respective map coordinates and image coordinates, and to fit the 3-D image with the map by calculating a geometrical transformation between the map coordinates and the image coordinates based on the weights.

15. The apparatus according to claim 14, wherein the interface is arranged to accept the map from a position tracking system, which measures the values of the physiological parameter at the respective map coordinates.

16. The apparatus according to claim 14, wherein the position tracking system comprises a probe comprising a position sensor and a sensor measuring the physiological parameter, which is moved over multiple locations in the organ so as to perform respective multiple measurements of the physiological parameter and of a position of the position sensor at the locations.

17. The apparatus according to claim 14, wherein the processor is further arranged to superimpose the values of the physiological parameter onto the 3-D image using the geometrical transformation and to display the superimposed values and 3-D image to a user.

18. The apparatus according to claim 14, wherein the processor is arranged to define distances between at least some of the image points and respective map points, to apply the weights assigned to the respective map points and image points to the distances to produce weighted distances, and to calculate a cumulative weighted distance metric based on the distances, the metric indicative of an accuracy of the geometrical transformation in fitting the 3-D image with the map.

19. The apparatus according to claim 18, wherein the processor is arranged to multiply a square of the distances by the respective weights to produce the weighted distances, and to sum the weighted distances to produce the cumulative weighted distance metric.

20. The apparatus according to claim 18, wherein the processor is arranged to fit the 3-D image with the map by modifying the geometrical transformation so as to minimize the cumulative weighted distance metric.

21. The apparatus according to claim 14, wherein the processor is arranged to accept an identification of an anatomical landmark external to the organ that appears in the 3-D image and of a respective map point acquired at a location of the anatomical landmark, and to register an image coordinate of the anatomical landmark in the 3-D image with a map coordinate of the acquired map point.

22. The apparatus according to claim 14, wherein the processor is arranged to accept the confidence levels from a user via the interface.

23. The apparatus according to claim 14, wherein the processor is arranged to automatically determine the confidence levels based on the map points and image points.

24. The apparatus according to claim 14, wherein the processor is arranged to detect a distortion in at least one of the map points and image points and to assign the confidence levels responsively to a level of the distortion.

25. The apparatus according to claim 14, wherein the processor is arranged to assign the confidence levels responsively to at least one of:

a density of the respective one or more of the map coordinates in the map;

an electrical tissue characteristic in a vicinity of the respective map coordinates and image coordinates;

a physiological tissue characteristic in the vicinity of the respective map coordinates and image coordinates;

a distance of the respective map coordinates and image coordinates from a registration landmark identified in both the map and the 3-D image; and

a physiological characteristic of the registration landmark.

26. The apparatus according to claim 14, wherein the 3-D image comprises at least one of a solid model, a surface model and a wire-frame model of the organ defined by at least some of the image points.