Image processing apparatus and method for quantitatively detecting and correcting the twist between images

• US 5,800,355 A
• Filed: 09/25/1996
• Issued: 09/01/1998
• Est. Priority Date: 09/27/1995
• Status: Expired due to Term

• Current Assignee: Toshiba Medical Systems Corporation (Canon Inc.)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Hasegawa, Hyoji
• Primary Examiner(s): Lateef, Marvin M.
• Assistant Examiner(s): MANTIS MERCADER, ELENI M
• Application Number:
  US08/719,548
• Time in Patent Office: 706 Days
• Field of Search: 128/659, 128/654, 128/653.1, 128/653.2, 378/62, 378/98.5, 382/128, 382/289, 382/296, 382/131, 250/363.04, 250/363.05, 250/363.07, 250/363.09, 600/407, 600/410, 600/420, 600/436
• US Class Current: 600/436
• CPC Class Codes: A61B 6/037 Emission tomography
• CPC Class Codes: A61B 6/507 for determination of haemodynamic parameters, e.g. perfusion CT
• CPC Class Codes: A61B 6/583 using calibration phantoms

Abstract

Radiotracer is dosed to an object to detect radiation rays emitted from the radiotracer. The detection is performed with R waves of an electrocardiogram serving as a trigger at the end of diastole of the myocardium and the end of systole of the same during a predetermined number of heart beats. Detected radiation-ray data is used to reconstruct a SPECT short axis image (an ED image) at the end of the diastole and a SPECT short axis image (an ES image) at the end of systole, each of which is formed of several slices. All slices of the SPECT short axis image are developed on a polar coordinate to make functional maps (bull'"'"'s-eye map). The amount of shift of the pattern of the segment value of the functional map of the ED image and the segment value of the functional map of the ES image is obtained for each slice by calculating the sum of products. The segment value for each slice is rotated on the functional map by the obtained amount so that the influence of the twist is corrected.

Claims

1. An image processing apparatus comprising:

reconstruction means for detecting radiation rays emitted from an object in synchronization with an electrocardiogram of the object so as to reconstruct a first image of a myocardium at an end of diastole and a second image of the myocardium at an end of systole, the first and second images having first and second tomographic images of plural slices;

means for developing the first and second tomographic images on a polar coordinate system;

means for detecting an amount of shift in a rotational direction between the first and second tomograph images of each slice; and

twist correction means for rotating at least either of the first and second tomographic images of each slice in accordance with the detected amount of the shift.

2. An apparatus according to claim 1, in which said first and second tomographic images are perpendicular to an axial line connecting an apex of a heart of the object and a base of the heart.

3. An apparatus according to claim 1, in which the first and second tomographic images of each slice are divided into plural segments in a circumferential direction, and said detecting means comprises means for detecting the amount of shift of a segment with which a difference between segments is made to be minimum.

4. An apparatus according to claim 3, in which said detecting means detects an amount of shift of the segment with which a sum of products of values of segments of the first and second tomographic images is made to be maximum.

5. An apparatus according to claim 4, in which said detection means obtains a value k with which a following calculation is made to be maximum:

##EQU3## where i is the segment number, j is the slice number, ED (j,i) is the value of the segment i in the slice j of the first radiation pattern image, ES (j,i) is the value of the segment i in the slice j of the second radiation pattern image, n1 and n2 are arbitrarily integers, and k is a variable for changing the segment, the sum of products of which is obtained.

6. An apparatus according to claim 5, in which said developing means develops the first tomographic image on the same plane with changing the radial directional positions of the first tomographic image of the plural slices to form a first bull'"'"'s-eye map and develops the second tomographic images on the same plane with changing the radial directional positions of the second tomographic images of the plural slices to form a second bull'"'"'s-eye map;

said detecting means calculates the amount of shift of all slices; and

said twist correction means corrects all the slices using the amount of shift calculated of all the slices.

7. An apparatus according to claim 5, in which said developing means develops the first tomographic image on the same plane with changing the radial directional positions of the first tomographic image of the plural slices to form a first bull'"'"'s-eye map and develops the second tomographic images on the same plane with changing the radial directional positions of the second tomographic images of the plural slices to form a second bull'"'"'s-eye map;

said detecting means calculates the amount of shift of only predetermined slices and interpolates the amount of shift of the other slices; and

said twist correction means corrects all the slices using the amount of shift calculated of all the slices.

8. An image processing apparatus for correcting twist of myocardium which is divided into plural slices in a radial direction, each slice being divided into plural segments in a circumferential direction, comprising:

means for comparing segment data for each slice of a given myocardium and corresponding segment data of reference myocardium to detect a degree of twist for each slice; and

means for rotating each slice of the given myocardium based on the degree of twist detected by said comparing means.

9. An image processing apparatus according to claim 8, in which said comparing means obtains a value k with which a following calculation is made to be maximum:

##EQU4## where i is the segment number, j is the slice number, ED (j,i) is the value of the segment i in the slice j of the first radiation pattern image, ES (j,i) is the value of the segment i in the slice j of the second radiation pattern image, n1 and n2 are arbitrary integers, and k is a variable for changing the segment, the sum of products of which is obtained.

10. A method for correcting a myocardium twist, comprising the steps of:

dividing the myocardium into plural slices in a radial direction, and dividing respective of said plural slices into plural segments in a circumferential direction;

comparing segment data for each slice of the myocardium with corresponding segment data of a reference myocardium;

detecting a degree of twist for each slice based on the comparison result obtained in said comparison step; and

rotating each slice of the myocardium based on degree of twist detected in said detecting step.

11. The method of claim 10 wherein:

said comparing step includes obtaining a value k with which the following calculation is made to be maximum, ##EQU5## where i is the segment number, j is the slice number, ED (j,i) is the value of the segment i in the slice j of the first radiation pattern image, ES (j,i) is the value of the segment i in the slice j of the second radiation pattern image, n1 and n2 are arbitrary integers, and k is a variable for changing the segment, the sum of products of which is obtained.

12. An image processing method, comprising the steps of:

detecting radiation rays emitted from an object in synchronization with an electrocardiogram of the object so as to reconstruct a first image of a myocardium at an end of a diastole and a second image of the myocardium at an end of systole, including,

reconstructing the first image to have a first tomographic image of plural slices, and

reconstructing the second image to have second tomographic images of plural slices;

developing the first and second tomographic images on a polar coordinate;

detecting an amount of shift in a rotational direction between the first and second tomographic images of each slice; and

rotating at least one of the first and second tomographic images of each slice in accordance with the detected amount of the shift.

13. The method of claim 12, wherein the reconstructing steps respectively include reconstructing the first tomographic images and the second tomographic images to be perpendicular to an axial line connecting an apex of a heart of the object and a base of the heart.

14. The method of claim 12, wherein:

the reconstructing steps respectively include reconstructing the first tomographic images and the second tomographic images of each slice by dividing the respective slices into plural segments in a circumferential direction; and

said detecting step, comprises detecting the amount of shift of a segment with which a difference between segments is made to be minimum.

15. The method of claim 14, wherein said step of detecting the amount of shift, comprises detecting the amount of shift of the segment with which a sum of products of values of segments of the first tomographic images and the second tomographic images is made to be maximum.

16. The method of claim 15, wherein said step of detecting includes obtaining a value k with which the following calculation is to be made maximum, ##EQU6## where i is the segment number, j is the slice number, ED (j,i) is the value of the segment i in the slice j of the first radiation pattern image, ES (j,i) is the value of the segment i in the slice j of the second radiation pattern image, n1 and n2 are arbitrarily integers, and k is a variable for changing the segment, the sum of products of which is obtained.

17. The method of claim 16 wherein:

said developing step comprises developing the first tomographic image on the same plane with changing the radial directional positions of the first tomographic image of the plural slices to form a first bull'"'"'s-eye map and developing the second tomographic images on the same plane with changing the radial directional positions of the second tomographic images of the plural slices to form a second bull'"'"'s-eye map;

said step of detecting the shift amount, including calculating the amount of shift slices; and

correcting all the slices using the amount of shift of all slices calculated in said calculating step.

18. The method of claim 16, wherein:

said developing step comprises developing the first tomographic image on the same plane with changing the radial directional positions of the first tomographic image of the plural slices to form a first bull'"'"'s-eye map and developing the second tomographic images on the same plane with changing the radial directional positions of the second tomographic images of the plural slices to form a second bull'"'"'s-eye map;

said step of detecting the amount of shift, includes calculating the amount of shift of only predetermined slices and interpolating the amount of shift of the other slices; and

correcting all of the slices using the amount of shift of all slices calculated in said calculating step.