Method for computed tomography of a periodically moving object to be examined, and a CT unit for carrying out this method

• US 6,937,690 B2
• Filed: 11/05/2003
• Issued: 08/30/2005
• Est. Priority Date: 11/05/2002
• Status: Expired due to Fees

• Current Assignee: Siemens Healthcare GMBH (Siemens AG)
• Original Assignee: Siemens AG
• Inventors: Bruder, Herbert, Flohr, Thomas, Stierstorfer, Karl
• Primary Examiner(s): Bruce, David V
• Application Number:
  US10/700,677
• Publication Number:
  US 20040131140A1
• Time in Patent Office: 664 Days
• Field of Search: 378/4, 378/8, 378/15, 378/19, 378/901
• US Class Current: 378/15
• CPC Class Codes: A61B 6/027 characterised by the use of a particular data acquisition trajectory, e.g. helical or spiral
• CPC Class Codes: A61B 6/032 Transmission computed tomography [CT]
• CPC Class Codes: A61B 6/4085 Cone-beams
• CPC Class Codes: A61B 6/503 for diagnosis of the heart
• CPC Class Codes: A61B 6/5288 involving retrospective matching to a physiological signal
• CPC Class Codes: A61B 6/541 involving acquisition triggered by a physiological signal
• CPC Class Codes: Y10S 378/901 Computer tomography program or processor

Abstract

A method is for producing CT images of a periodically moving object to be examined, in particular a method for cardio volume reconstruction for multi-row CT units in spiral mode. A multi-row CT unit designed is further designed therefore. A small partial revolution segments along a revolution spiral is used to calculate in each case per partial revolution segment a plurality of segment images that are converted in a second step to partial images in the target image plane (for example axial plane) and are assembled with the correct phase to form complete images.

Claims

1. A method for producing CT images of a periodically moving area to be examined, comprising:

scanning the area to be examined with the aid of a beam, emanating from a focus, with a multiplicity of rays relating to a multiplicity of detector elements of a detector array, the focus and the detector array being relatively displaced and supplying data from which a change in attenuation of the rays between the focus and detector elements is determinable;

collecting movement data from the periodically moving area to be examined, the detector data and data resulting therefrom being assignable to the movement states;

using only output data, correlated with a specific movement state of the body region, to compose at least one CT image;

dividing the output data, supplied during the movement of the focus on a spiral segment of sufficient length for reconstructing a CT image and assigned to specific movement states, into output data with reference to N_seg subsegments,

reconstructing segment images by back projection and filtering, referred to movement state, with an image plane inclined to the system axis;

combining segment images belonging to a subsegment and correlated with a movement state to form a component image, referred to movement state, with reference to a target image plane; and

combining component images, referred to movement state, for at least one target image plane to form resulting at least one CT image of a specific movement state.

2. The method as claimed in claim 1, wherein, in order to form the component image from the segment images, a spatio-temporal filter function is applied to the segment images, the spatio-temporal filter function being variable with reference to the spatial resolution in the direction of the system axis and with reference to temporal resolution.

3. The method as claimed in claim 2, wherein the set of segment images is represented by S(θ

_R,j+kπ

, Z_R,q (θ

_R,j+kπ

)), wherein 1≦

j≦

N_seg/2, with k and j representing whole numbers, θ

_R,j+kπ

representing the central angle of a jth volume stack in the kth half revolution θ

_R,j=θ

_R+(j−

1)·

π

/(N_seg/2) and $z_{R,q}<br><br>\left({\mathrm{\theta <br><br>}}_{R,j}+k<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}\right)=z_{\mathrm{start}}<br><br>\left(\mathrm{\theta <br><br>}=0\right)+p·<br><br>\frac{{\mathrm{\theta <br><br>}}_{R,j}+k<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}}{2<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}}+\left(q-N_{\mathrm{tilt}}/2\right)·<br><br>\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>z$

representing the assigned z positions of the segment images in the volume stack, and wherein p is the table feed and z_start is the starting position of the spiral revolution at the projection angle θ

=0.

4. The method as claimed in claim 3, wherein the filtered segment images Ŝ

(θ

_R,j,z_img) are calculated in accordance with the rule $\hat{S}<br><br>\left({\mathrm{\theta <br><br>}}_R<br><br>{,}_j,z_{\mathrm{img}}\right)=\frac{1}{W_j}·<br><br>\underset{k,q}{\sum <br><br>}<br><br>S<br><br>\left({\mathrm{\theta <br><br>}}_{R,j}+k<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>},z_{R,q}<br><br>\left({\mathrm{\theta <br><br>}}_{R,j}+k<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}\right)\right)·<br><br>w_z(d<br><br>\left(z_{R,q}<br><br>\left({\mathrm{\theta <br><br>}}_{R,j}+k<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}\right)<br><br>z_{\mathrm{img}}\right)·<br><br>w_{\mathrm{phase}}<br><br>\left(t<br><br>\left({\mathrm{\theta <br><br>}}_{R,j}+k<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}\right)-c_R<br><br>\left(k\right)\right),$

where 1≦

j≦

N_seg/2, w_z is a weighting function that determines the layer thickness, and d denotes at least one of the distance and a similar function of the segment image plane at the z position z_R,q from the z position z_img, w_phase representing the temporal spacing of the segment image planes centered at the angle θ

_R,j+kπ

from the temporal position c_R(k,j), determined relative to the movement state, in the movement cycle, assigned to the angle θ

_R,j+kπ

, of the object to be examined, and $W_j=\underset{k,q}{\sum <br><br>}<br><br>w_z(d<br><br>\left(z_q<br><br>\left({\mathrm{\theta <br><br>}}_{R,j}+k<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>},z_{\mathrm{img}}\right)\right)·<br><br>w_{\mathrm{phase}}<br><br>\left(t<br><br>\left({\mathrm{\theta <br><br>}}_{R,j}+k<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}\right)-c_R<br><br>\left(k,j\right)\right)$

corresponding to the weighted sum for correct normalization.

5. The method as claimed in claim 4, wherein the CT image at the z position z_img is yielded as:

$I<br><br>\left(x,y,z_{\mathrm{img}}\right)=\frac{1}{N_{\mathrm{seg}}/2}·<br><br>\underset{j=1}{\overset{N_{\mathrm{seg}}/2}{\sum <br><br>}}<br><br>\hat{S}<br><br>\left({\mathrm{\theta <br><br>}}_R<br><br>{,}_j,x,y,z_{\mathrm{img}}\right).$

6. The method as claimed in claim 1, wherein the segment images are reconstructed for image planes that are inclined with reference to the system axis both by an angle of inclination γ

about a first axis cutting the system axis at right angles, and by a tilt angle δ

about a second axis cutting both the first axis and the system axis at right angles.

7. The method as claimed in claim 1, wherein neighboring subsegments overlap one another, and wherein the output data belonging to the overlap regions are respectively weighted in such a way that the weights of mutually corresponding output data of the subsegments overlapping one another yield one in each case.

8. The method as claimed in claim 1, wherein segment images for a number n_ima of inclined image planes are reconstructed for each subsegment, the image planes having the different z positions z_ima.

9. The method as claimed in claim 8, wherein the plurality of inclined image planes intersect in a straight line running tangentially to the subsegment.

10. The method as claimed in claim 8, wherein it holds for the extreme values +δ

_max and −

δ

_max of the tilt angle δ

inclined image planes belonging to a subsegment that;

$\pm <br><br>{\mathrm{\delta <br><br>}}_{\max }=\arctan <br><br>\left(\frac{-\frac{\mathrm{SM}}{2}+\mathrm{Sp}<br><br>\text{\hspace{1em}<br><br>}<br><br>\frac{{\mathrm{\alpha <br><br>}}_l}{2<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}}\pm <br><br>\mathrm{RFOV}<br><br>\text{\hspace{1em}<br><br>}<br><br>\cos <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_l<br><br>\text{\hspace{1em}<br><br>}<br><br>\tan <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\gamma <br><br>}}_0}{-\frac{R_f}{\cos <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\gamma <br><br>}}_0}-\left(\pm <br><br>\mathrm{RFOV}\right)<br><br>\text{\hspace{1em}<br><br>}<br><br>\frac{\sin <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_l}{\cos <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\gamma <br><br>}}_0}}\right)$

γ

₀ being the value of the angle of inclination γ

determined for the tilt angle δ

=0 in accordance with ${\mathrm{\gamma <br><br>}}_0=\arctan <br><br>\left(\frac{-S<br><br>\text{\hspace{1em}<br><br>}<br><br>p<br><br>\text{\hspace{1em}<br><br>}<br><br>\widehat{\mathrm{\alpha <br><br>}}}{2<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>R_f<br><br>\sin <br><br>\text{\hspace{1em}<br><br>}<br><br>\widehat{\mathrm{\alpha <br><br>}}}\right).$

11. The method as claimed in claim 6, wherein, for a given absolute value |δ

_max| of the maximum value of the tilt angle δ

, the associated optimal value γ

_min of the angle of inclination γ

is determined in such a way that an error criterion is fulfilled.

12. The method as claimed in claim 6, wherein the tilt angle δ

of the inclined image planes is determined according to $\mathrm{\delta <br><br>}<br><br>\left(i\right)={\mathrm{\delta <br><br>}}_{\max }<br><br>\frac{2<br><br>i-\left(n_{\mathrm{ima}}-1\right)}{n_{\mathrm{ima}}-1}.$

13. The method as claimed in claim 1, wherein the combination of the plurality of segment images to form a component image is performed by interpolation.

14. The method as claimed in claim 1, wherein the combination of the plurality of segment images to form a component image is performed by averaging.

15. The method as claimed in claim 1, wherein the combination of the plurality of segment images to form a component image is performed by weighted averaging.

16. The method as claimed in claim 1, wherein the combination of the plurality of segment images to form a component image is weighted in accordance with a desired reconstruction layer thickness of the component image.

17. The method as claimed in claim 1, wherein the number of segment images that are combined to produce a component image is selected in accordance with a respectively desired reconstruction layer thickness of the component image.

18. The method as claimed in claim 17, wherein the segment images are reconstructed with the smallest possible layer thickness.

19. The method as claimed in claim 1, wherein the component images are determined for a target image plane that cuts the system axis at right angles.

20. The method as claimed in claim 1, wherein the combination of the component images to form the resulting CT image is performed by adding.

21. The method as claimed in claim 1, wherein the data corresponding to the segment images are compressed.

22. The method as claimed in claim 21, wherein the compressed data corresponding to the segment images have a pixel matrix which is inhomogeneous in such a way that the resolution in a first direction running at least substantially in the direction of the reference projection direction belonging to the respective subsegment is higher than in a second direction which runs at least substantially orthogonal to the reference projection direction.

23. The method as claimed in claim 22, wherein the compressed data to the segment images have pixels of elongated shape, the longest extension of the pixels running at least substantially in the direction of the reference projection direction belonging to the respective subsegment.

24. The method as claimed in claim 23, wherein rectangular pixels are provided.

25. The method as claimed in claim 23, wherein the segment images are converted into the inhomogeneous pixel matrix.

26. The method as claimed in claim 23, wherein the segment images are reconstructed in the inhomogeneous pixel matrix.

27. The method as claimed in claim 26, wherein the back projection direction corresponds at least substantially to the direction of the reference project direction belonging to the respective subsegment.

28. The method as claimed in claim 21, wherein the compression is canceled in the course of the combination of the component images to form a resulting CT image having a homogeneous pixel matrix.

29. The method as claimed in claim 21, wherein the segment images are converted into the inhomogeneous pixel matrix.

30. The method as claimed in claim 21, wherein the pixels of the homogeneous pixel matrix are obtained by interpolation from the pixels of the inhomogeneous pixel matrix.

31. The method as claimed in claim 1, wherein the pixels of the homogeneous pixel matrix are obtained by averaging from the pixels of the inhomogeneous pixel matrix.

32. A CT unit for scanning a periodically moving object to be examined, comprising:

a beam emanating from at least one focus;

a detector array including a multiplicity of distributed detector elements for detecting the rays of the beam, the at least one focus being moved relative to the object to be examined on at least one focal path, running around the object to be examined, with a detector array situated opposite thereof; and

functional means for detecting the movement state of the object to be examined, for collecting the detector data, and for filtering and back projection in accordance with the method of claim 1.

33. The CT unit as claimed in claim 32, wherein the functional means are implemented at least partially by at least one of programs and program modules.

34. The CT unit as claimed in claim 32, wherein an ECG is provided for detecting the movement state of the object to be examined.

35. The method as claimed in claim 9, wherein it holds for the extreme values +δ

_max and −

δ

_max of the tilt angle δ

inclined image planes belonging to a subsegment that;

$\pm <br><br>{\mathrm{\delta <br><br>}}_{\max }=\arctan <br><br>\left(\frac{-\frac{S<br><br>\text{\hspace{1em}<br><br>}<br><br>M}{2}+S<br><br>\text{\hspace{1em}<br><br>}<br><br>p<br><br>\frac{{\mathrm{\alpha <br><br>}}_l}{2<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}}\pm <br><br>\mathrm{RFOV}<br><br>\text{\hspace{1em}<br><br>}<br><br>\cos <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_l<br><br>\tan <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\gamma <br><br>}}_0}{-\frac{R_f}{\cos <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\gamma <br><br>}}_0}-\left(\pm <br><br>\mathrm{RFOV}\right)<br><br>\frac{\sin <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_l}{\cos <br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\gamma <br><br>}}_0}}\right)$

γ

₀ being the value of the angle of inclination γ

determined for the tilt angle δ

=0 in accordance with ${\mathrm{\gamma <br><br>}}_0=\arctan <br><br>\left(\frac{-S<br><br>\text{\hspace{1em}<br><br>}<br><br>p<br><br>\text{\hspace{1em}<br><br>}<br><br>\widehat{\mathrm{\alpha <br><br>}}}{2<br><br>\mathrm{\pi <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>R_f<br><br>\sin <br><br>\text{\hspace{1em}<br><br>}<br><br>\widehat{\mathrm{\alpha <br><br>}}}\right).$

36. The method as claimed in claim 24, wherein the segment images are converted into the inhomogeneous pixel matrix.

37. The method as claimed in claim 24, wherein the segment images are reconstructed in the inhomogeneous pixel matrix.

38. The method as claimed in claim 37, wherein the back projection direction corresponds at least substantially to the direction of the reference project direction belonging to the respective subsegment.

39. The CT unit as claimed in claim 33, wherein an ECG is provided for detecting the movement state of the object to be examined.

40. The method for producing CT images as claimed in claim 1, wherein the periodically moving area to be examined includes a cardiac region of a living being.

41. The method for producing CT images as claimed in claim 1, wherein the focus and the detector array are relatively displaced in a fashion rotating on an equidistant spiral path, including a multiplicity of spiral segments, and parallel to the system axis.

42. The CT unit as claimed in claim 32, wherein the CT unit is for scanning a heart.

43. A method for producing CT images of a periodically moving area to be examined, comprising:

supplying data, from which a change in attenuation of rays between a relatively displaceable focus and detector elements of an array is determinable;

collecting movement data from the periodically moving area to be examined, detector data and collected movement data being assignable to the movement states;

using only output data, correlated with a specific movement state, to compose at least one CT image;

dividing the output data, supplied during the movement of the focus on a spiral segment of sufficient length for reconstructing a CT image and assigned to specific movement states, into output data with reference to N_seg subsegments,

reconstructing segment images with an image plane;

combining segment images belonging to a subsegment and correlated with a movement state to form a component image with reference to a target image plane; and

combining component images, referred to movement state, for at least one target image plane to form resulting at least one CT image of a specific movement state.

44. The method as claimed in claim 43, wherein, in order to form the component image from the segment images, a spatio-temporal filter function is applied to the segment images, the spatio-temporal filter function being variable with reference to the spatial resolution in the direction of the system axis and with reference to temporal resolution.