DE10126638B4 - Procedures for computed tomography as well as computed tomography (CT) device - Google Patents

Abstract

Process for computer tomography, comprising the process steps
a) for scanning an object with a cone-shaped beam of rays emanating from a focus and with a matrix-like detector array for detecting the beam of rays, the focus is moved relative to the object on a spiral path around a system axis, the detector array supplying the received radiation with corresponding output data, and
b) images with an inclined image plane are reconstructed from the output data supplied during the movement of the focus on a spiral segment, the image planes both about a first axis that intersects the system axis at an angle of inclination γ and about a second, both the first axis and the system axis, which intersects at right angles, is also inclined by a tilt angle δ with respect to the system axis.

Description

The invention relates to a method for the Computer tomography, showing the process steps that for Scanning an object with a conical beam of rays emanating from a focus and with a matrix-like detector array for detecting the beam the focus relative to the object on a spiral path around a system axis moves, the detector array corresponding to the received radiation Output data delivers, and that each time during the movement of the focus output data supplied on a spiral segment with relative images to be reconstructed to the system axis. The invention relates Moreover a computed tomography (CT) device having a radiation source, the focus of which emits a conical beam of rays, a matrix-like detector array for detecting the beam, wherein the detector array corresponding to the received radiation output data provides means for generating a relative movement between the radiation source and detector array on the one hand and an object on the other hand and one Image computer to which the output data are supplied, the means to generate a relative movement for scanning the object with the bundle of rays and the two-dimensional detector array a relative movement of the Cause focus to a system axis such that the focus is relative moved to the system axis on a helical spiral path, the Central axis corresponds to the system axis, and being the image calculator from each during the movement of the focus on a spiral segment provided output data Reconstructed images with an image plane inclined relative to the system axis.

Such a procedure is called spiral CT; a corresponding method or CT device is from the US 5 802 134 known.

bzw.The in 1 illustrated spiral path of focus F is described by the following equations:

respectively.

gilt und h die Steigung der Spiralbahn pro Umdrehung des Fokus F ist. α ist der Projektionswinkel, wobei im Folgenden eine Bildebene betrachtet wird, die zu Daten gehört, die über einen Projektionswinkelbereich von ±α aufgenommen wurden, wobei die zu der Bildebene gehörige Referenzprojektion bei α_r = 0 liegt, also die Mitte des Projektionswinkelbereichs ±α darstellt. α_r wird im Folgenden als Referenzprojektionswinkel bezeichnet.In this case, the case that the detector elements of the detector array are arranged in rows running transverse to the system axis Z and parallel to the system axis Z, S for the extension of a detector line in the direction of the system axis and p for the pitch, where

applies and h is the slope of the spiral path per revolution of the focus F. α is the projection angle, in the following an image plane is considered that belongs to data that was recorded over a projection angle range of ± α, the reference projection belonging to the image plane being α _r = 0, i.e. representing the center of the projection angle range ± α , α _r is referred to below as the reference projection angle.

In the case of conventional spiral CT, so-called Reconstructed transverse sectional images, i.e. Images for image layers, which is perpendicular to the system axis designated z and thus contains the x and y axes, where the x and y axes are perpendicular to each other and to the system axis z stand.

In case of US 5 802 134 on the other hand, images are reconstructed for image planes, which according to 2 are inclined by an inclination angle γ about the x-axis to the system axis z. This achieves the at least theoretical advantage that the images contain fewer artifacts if the angle of inclination γ is chosen in this way is that there is a good, if possible according to a suitable error criterion, for example a minimal square mean of the distance of all points of the spiral segment measured in the z direction from the image plane, an optimal adaptation of the image plane to the spiral path.

In the case of US 5 802 134 Fan data, ie data recorded in the known fan geometry, used for the reconstruction, which were obtained by moving the focus over a spiral segment of length 180 ° plus fan or cone angle, eg 240 °. Relative to the reference projection angle α _r = 0 applies to the normal vector of the image plane

The optimal angle of inclination γ obviously depends on the slope of the spiral and thus on the pitch p.

Basically, this can be done from the US 5 802 134 known methods can be used for any values of the pitch p. However, below the maximum pitch p _max, optimal use of the available detector area and thus the radiation dose supplied to the patient for image acquisition (detector and thus dose use) is not possible, because although a given transverse layer, i.e. a layer of the Object, which is scanned via a spiral segment that is longer than 180 ° plus fan or cone angle, can be from the US 5 802 134 known methods for values of the pitch p below the maximum pitch pmax, only a spiral segment of length 180 ° plus cone angle can be used, since the use of a longer spiral segment would make it impossible to adapt the image plane sufficiently well to the spiral path.

The invention is based on the object of designing a method and a CT device of the type mentioned at the outset in such a way that the prerequisites for optimum detector and thus dose use are also given for values of the pitch p below the maximum pitch p _max .

According to the invention, this is a method task in question solved by a method with the features of claim 1.

So in the case of the invention from each during the movement of the focus on a spiral segment provided output data Reconstructed images with an inclined image plane, their image planes both around a first axis that intersects the system axis at right angles an angle of inclination γ as also around a second one, both the first axis and the system axis right-angled axis by a tilt angle δ with respect to System axis are inclined.

This is also at the maximum Pitch values below the pitch are possible, an at least approximately complete detector and achieve dose usage.

According to a first alternative embodiment of the invention, output _data for a total segment of length [−α _max , + α _max ] are obtained for a given pitch p and a given z position z _ima , where α _max = Mπ / p and M is the number the detector lines is. This total segment is divided into a number n _ima of overlapping spiral segments, each of which has the length of 180 ° plus cone angle. A separate image with an inclined image plane is reconstructed at the point z _ima for each of the spiral segments. By reconstructing an image with an inclined image plane for each of the spiral segments, it is possible to optimally adapt the image plane of the image for each of these spiral segments to the corresponding section of the spiral path, and theoretically both the detector array and the dose, by appropriately selecting the inclination angle γ and the tilt angle δ to be used completely and as far as possible in practice.

In an alternative second embodiment, on the basis of the output data obtained for a spiral segment of length 180 ° plus cone angle centered with respect to the reference projection angle α _r = 0, a number of n _ima images with a differently inclined image plane for different z positions is obtained. By reconstructing several images with differently inclined image planes for different z-positions, it is possible to optimally adapt the image plane of the image for each of these z-positions to the spiral segment and the detector array as well as the dose by appropriately selecting the inclination angle γ and the tilt angle δ theoretically fully and to the greatest extent possible in practice. According to a preferred embodiment of the invention, the plurality of inclined image planes intersect in a straight line that is tangent to the spiral.

wobei γ₀ der gemäß

für den Kippwinkel δ = 0 ermittelte Wert des Neigungswinkels γ ist.In order to obtain the most complete possible use of detector and dose, according to a variant of the invention, the extreme values + δ _max and −δ _{max of} the tilt angle δ of the inclined image planes belonging to a spiral segment apply:

where γ ₀ according to

for the tilt angle δ = 0 is the value of the inclination angle γ.

In the interest of high image quality, according to a further variant of the invention it is provided that | δ _max | of the maximum value of the tilt angle δ, the associated optimum value γ _{min of} the inclination angle γ is determined in such a way that an error criterion, for example a minimum square mean of the distance of all points of the spiral segment from the image plane measured in the z direction, is met.

If the axis of rotation about which the focus rotates around the system axis is not identical to the system axis, but intersects it at a so-called gantry angle ρ, then the angle of inclination γ 'to be selected applies

Here, too, there is the possibility, for a given amount of the maximum value of the tilt angle | δ _max | to determine the associated optimum value of the angle of inclination γ 'in such a way that an error criterion, for example a minimum mean value of the distances of all points of the spiral segment from the image plane measured in the z direction, is fulfilled.

Ebenfalls im Interesse einer möglichst vollständigen Detektor- und Dosisnutzung werden unter der Voraussetzung von Detektorzeilen gleicher Breite gemäß einer Variante der Erfindung die Kippwinkel δ der geneigten Bildebenen nach

ermittelt.In order to obtain the most complete use of detector and dose, the following also applies according to a variant of the invention for the number n _{ima of} the inclined image planes for which images with an inclined image plane are generated for each spiral segment:

Likewise, in the interest of the most complete possible use of detector and dose, the tilting angles δ of the inclined image planes are provided, provided detector lines of the same width are used, according to a variant of the invention

determined.

For those who are used to using CT equipment Obtaining transverse sectional images is according to a variant of the invention reformatting is provided, that is to say that a transverse sectional image is generated in a further process step by several Images with an inclined image plane can be combined. It can the summary in an embodiment of the invention take place by several images with an inclined image plane to form a transverse sectional image by interpolation or by weighted averaging in particular be summarized.

When combining several images with an inclined image plane to form a transverse sectional image, according to a preferred variant of the invention, there is also the possibility that the method step also involves the number of images with an inclined image plane which are combined to produce a transverse sectional image corresponding to the respectively desired layer thickness of the in the layer shown in the transverse layer is selected. In the interest of the highest possible image quality of the transverse sectional images, there is the possibility of using the images with an inclined image plane with the least possible to reconstruct the layer thickness.

gewählt wird.According to a further preferred variant of the invention, a desired layer thickness of the transverse layer shown in a transverse sectional image can be set by the number of images with an inclined image plane, which are combined to produce a transverse sectional image

is chosen.

According to the invention, the one relating to a CT device Task solved through a CT scanner solved with the features of one of the claims 18 to 34. Regarding the Operation and the advantages of the CT device according to the invention to the above statements in terms of of the method according to the invention directed.

The invention is exemplified below based on the attached schematic drawings explained. Show it:

1 and 2 diagrams illustrating the geometry of processes according to the prior art,

3 and 4 a schematic, partially block diagram-like representation of a CT device according to the invention, which operates according to the method according to the invention,

5 in to the 1 and 2 analog representation of a diagram illustrating the geometry of the method according to the invention,

6 the distances of all points of a spiral segment from the image plane, which are measured in the direction of the z-axis and are related to the width S of a row of the detector array, are shown for different inclination angles γ above the sine of the projection angle α multiplied by the radius R _f , specifically for the tilt angle δ = 0

7 the square roots of the square mean values, plotted over the quotient γ / γ ₀ and related to the width S of a detector line, of the distances of all points of the spiral segment under consideration from the image plane measured in the direction of the z-axis,

8th for a spiral segment, the distances of all points of the spiral segment, measured in the direction of the z-axis and related to the width S of a row of detector elements, from the image planes of the two images belonging to this spiral segment inclined by −δ _max and + δ _max and by γ _min the sine of the projection angle a multiplied by the radius Rf,

9 and 10 the image planes belonging to a spiral segment from different angles in perspective.

11 illustrates the detector and thus dose usage for a CT device according to the prior art using the virtual detector for M = 12 and p = 8.

12 and 13 illustrate the use of the detector and thus the dose for a CT device according to the invention using the virtual detector also for M = 12 and p = 8 or for M = 12 and p = 12.

14 the maximum tilt angle δ _max as a function of the pitch p for M = 12,

15 the time interval covered for any z position in as a function of the pitch p for T _rot = 0.5 s,

16 the image planes of the images obtained for the same reference projection angle α _r , ie from the same cardiac cycle, namely for M = 12, p = 3,

17 in to the 4 analog representation of the CT device according to the invention in an operating state with a gantry inclined with respect to the system axis,

18 in to the 1 analogous representation according to the geometry of the CT device according to the invention for the operating state with the gantry inclined with respect to the system axis 17 .

19 the course of the angle of inclination γ 'for the case of the tilted gantry as a function of the reference projection angle α _r , specifically for M = 16, p = 16 and a gantry angle of ρ = 30 °,

20 the course of the maximum tilt angle δ _max in the case of the tilted gantry as a function of the reference projection angle α _r for the M = 16, p = 16, specifically for ρ = 30 ° and ρ = 0 ° and ± RFOV, and

21 the to the 19 analog representation for a CT device according to the prior art.

In the 3 and 4 A multi-layer CT device of the third generation according to the invention which is suitable for carrying out the method according to the invention is shown. Whole with 1 designated measuring arrangement has an overall 2 designated X-ray source with one of these upstream source near the aperture 3 ( 4 ) and an areal array of several rows and columns of detector elements - one of these is in 3 With 4 designated - trained detector array 5 with an upstream detector aperture 6 ( 4 ) on. The X-ray source 2 with the aperture 3 on the one hand and the detector array 5 with the aperture 6 on the other hand are in from the 4 clearly on a rotating frame also referred to below as a gantry 7 each other so ge attached opposite that one in operation of the CT device from the X-ray source 2 outgoing, through the adjustable radiation diaphragm 3 superimposed, pyramid-shaped X-ray beam, the edge rays of which 8th are labeled on the detector array 5 incident. Here is the radiation diaphragm 6 by means of the radiation diaphragm 3 set cross-section of the X-ray beam correspondingly set so that only that area of the detector array 5 is released, which can be hit directly by the X-ray beam. These are in the in the 3 and 4 illustrated operating state four rows of detector elements. That more, from the aperture 6 covered rows of detector elements are present in 4 dotted hinted.

The x-ray beam has the cone angle φ, which is the opening angle of the x-ray beam projected in a plane perpendicular to the system axis. The cone angle φ corresponds to the fan angle with the individual lines of the detector array 5 interacting portions of the x-ray beam.

The gantry 7 can be set in rotation about a system axis designated Z by means of a drive device, not shown. The system axis Z is identical to the z axis of an in 1 spatial rectangular coordinate system shown. The circular opening of the gantry 7 has a radius R _M that corresponds to the radius of the measuring field or the object cylinder. The radius on which the focus F moves is designated R _f .

The columns of the detector array 5 likewise run in the direction of the z-axis, while the lines, whose width S is measured in the direction of the z-axis and is, for example, 1 mm, run transversely to the system axis Z or the z-axis.

In order to be able to bring an examination object, for example a patient, into the beam path of the X-ray beam, there is a positioning device 9 provided, which can be moved parallel to the system axis Z, that is to say in the direction of the z axis.

For recording volume data on the storage device 9 located examination object, for example a patient, the examination object is scanned by moving the measuring unit 1 A large number of projections from different projection directions α are recorded around the system axis Z. The one from the detector array 5 The data supplied therefore contain a large number of projections for each active detector line.

During the continuous rotation of the measuring unit 1 around the system axis Z is the storage device at the same time 9 in the direction of the system axis Z relative to the measuring unit 1 continuously shifted, with synchronization between the rotational movement of the rotating frame 7 and the translational movement of the bearing device 9 in the sense that the ratio of the translation speed to the rotational speed is constant and this constant ratio can be set by a value for the advance h of the storage device ensuring a complete scanning of the volume of interest of the examination object 9 per revolution of the rotating frame 7 is chosen.

The ratio of the feed h to the width S of a detector line is referred to as pitch p; the maximum pitch p _max , which just barely guarantees a seamless scanning of an examination object, results on the assumption that all lines of the detector array 5 have the same width S, where n is the number of active lines of the detector system 5 is.

The focus F of the X-ray source 2 moves from the examination object on one in 1 helical spiral path denoted by H around the system axis Z, which is why the described type of recording of volume data is also referred to as spiral scanning or spiral scan. The detector elements of each row of the detector array 5 delivered volume data, each of which is a specific row of the detector array 5 and projections assigned to a specific position with respect to the system axis Z are read out in parallel in a sequencer 10 serialized and to an image calculator 11 transfer.

After preprocessing the volume data in a preprocessing unit 12 the image calculator 11 the resulting data stream reaches a memory 14 in which the volume data corresponding to the data stream are stored.

The image calculator 11 contains a reconstruction unit 13 which reconstructs image data from the volume data, for example in the form of sectional images of desired layers of the examination object, according to methods known per se to the person skilled in the art. The one from the reconstruction unit 13 reconstructed image data are stored in a memory 14 saved and can on a to the image calculator 11 connected display unit 16 , such as a video monitor.

The X-ray source 2 , for example an x-ray tube, is generated by a generator unit 17 supplied with the necessary voltages and currents. In order to be able to set these to the respectively necessary values, the generator unit is 17 a control unit 18 with keyboard 19 and mouse 20 assigned that allows the necessary settings.

The other operation and control of the CT device is also carried out by means of the control unit 18 and the keyboard 19 as well as the mouse 20 , which is illustrated by the fact that the control unit 18 with the calculator 11 connected is.

In a first operating mode, the the usual The procedure for spiral scans corresponds to that in the course of a spiral scan of volume data recorded transverse sectional images, So sectional images, the image plane perpendicular to the system axis Z runs reconstructed according to procedures known per se and in the literature described as 180 LI reconstruction and 360 LI reconstruction are.

However, there is also the possibility in a second operating mode from the volume data, at least as Intermediate step to reconstruct sectional images, their image planes are inclined to the system axis Z.

In this case, according to the invention, in contrast to that from the US 5 802 134 known procedure proceeded so that the image plane with respect to the system axis Z both around a first axis intersecting the system axis Z at a right angle, namely the x-axis, about an angle of inclination γ and the tilt angle δ and also about a second, both the first axis (x Axis) and the system axis Z intersecting at right angles, namely the y axis, are inclined by a tilt angle δ with respect to the system axis, as is the case in FIG 5 can be seen.

In a first mode of the second operating mode, output data for a spiral segment of length [−α _max , + α _max ] are used for a given pitch p and a given z position z _ima , where α _max = Mπ / p and M the number of the detector lines, the z position indicating the position of the image plane on the z axis. This total segment is divided into a number n _ima of overlapping spiral segments, each of which has the length of 180 ° plus cone angle. A separate image with an inclined image plane is reconstructed at the point z _ima for each of the spiral segments. By reconstructing an image with an inclined image plane for each of the spiral segments, it is possible by an appropriate choice of the inclination angle γ and the tilt angle δ of the image plane of the image for each of these spiral segments to be optimally adapted to the corresponding section of the spiral path and both that of the radiation diaphragm 6 released area of the detector array 5 as well as to use the radiation dose in this area theoretically completely and as far as possible in practice.

In an alternative second mode of the second operating mode, a spiral segment of length 180 ° plus cone angle φ centered with respect to the reference projection angle α _r = 0 is used, and on the basis of this spiral segment a number of n _ima images with differently inclined image planes for different z positions. In this mode, too, it is possible to optimally adapt the image plane of the image for each of these z positions, and both, by reconstructing several images with a differently inclined image plane for different z positions and by appropriately selecting the inclination angle γ and the tilt angle δ of the image segment for each of these z positions to use the detector array and the dose theoretically completely and in practice as far as possible. According to a preferred embodiment of the invention, the plurality of inclined image planes intersect in a straight line that is tangent to the spiral.

The second mode is as follows explained in more detail.

For the sake of simplicity, a single spiral segment is considered which is centered with respect to the reference projection angle α _r = 0. Since the image planes of the n _ima images are inclined both with respect to the x-axis by the angle of inclination γ and with respect to the y-axis by the angle of inclination δ, the normal vector of an image plane is given by:

The distance d (α, δ, γ) that any point (x _f , y _f , z _f ) on the spiral path in the z direction from the image plane inclined by the angle of inclination γ and the angle of inclination δ is given by

It is assumed that the position (–R _f , 0, 0) of the focus F for the reference projection angle α _r = 0 lies in the image plane.

The angle of inclination γ and the angle of tilt δ of the inclined image plane must be selected in such a way that the root mean square of all points on the spiral segment is minimal.

Assuming that bt is the coordinate system x - y rotated about the z-axis by an angle α - π / 2, then bt is the local coordinate system for a projection with the projection angle α. x = bsinα + tcosα y = –bcosα + tsinα (4)

If you imagine a virtual detector array before that the projection of the detector array into the system axis plane containing z, the so-called virtual detector plane, so for the detector level t = 0.

Each point (x, y, z) on the image plane is identified by

If you insert (4) with t = 0 in (5), you get the intersection line of the virtual detector plane with the image plane

The z coordinate on the virtual detector level is given by

wobei ⁀ der Winkel ist, bei dem die Spiralbahn die Bildebene durchstößt. Es hat sich gezeigt, dass ⁀ = π/3 ein günstiger, wenn nicht optimaler Wert für diesen Parameter ist.The angle of inclination γ is initially in the same way as in the case of US 5 802 134 optimized, ie for the tilt angle δ = 0. The result is

where ⁀ is the angle at which the spiral path penetrates the image plane. It has been shown that ⁀ = π / 3 is a favorable, if not optimal, value for this parameter.

For the according to (8) with ⁀ = π / 3 obtained tilt angle y ₀ , the tilt angle δ is optimized. The optimization criterion for the tilt angle δ is that the z-coordinate according to (7) for the lines –RFOV ≤ b ≤ RFOV, which limit the area of the examination object covered by the radiation in the z-direction backwards or forwards, not only within the active detector area, ie within that of the beam diaphragm 6 released area of the detector array hit by the radiation 5 , must lie, but must also use the detector area as well as possible.

wobei M die Anzahl der Detektorzeilen und S die in z-Richtung gemessenen Breite einer Detektorzeile ist.For the maximum possible tilt angle ± δ _max , the lines given by the z coordinate according to (7) for b = ± RFOV reach the front or rear end of the detector surface in the z direction. If this applies to the respective spiral segment for the projections at the beginning and end of the spiral segment, ie for the outermost projection angles α _r = ± 120 °.

where M is the number of detector lines and S is the width of a detector line measured in the z direction.

By using (6) for α = α _l and γ = γ ₀ in (8) and solving for δ _max , the result is

A new γ _min is determined for the corresponding δ _max by re-iteration, specifically by minimizing the quadratic mean of the distances d (α, δ _max , γ) measured in the z direction of all points of the spiral segment from the image plane according to (3) ,

gilt.The available range [−α _max , + α _max ] of the tilt angle is now subdivided evenly in accordance with the number n _{ima of} the images to be reconstructed with an inclined image plane, as in the case of the exemplary embodiment described. That is, in the case of a uniform subdivision, each image plane 0 i i n n _ima − 1 is characterized by the angle of inclination γ _min (which is preferably the same for all image planes as in the case of the exemplary embodiment described) and the respective tilt angle δ _(i) , where for the respective tilt angle

applies.

The number n _{ima of} the images with inclined image planes to be reconstructed for the spiral segment is given by

The effect of the method and CT device according to the invention is described below using the example of a CT device with M = 12 detector lines of width S, which is operated with a pitch of p = 12. For each z-position z _ima a spiral segment of length [–α _max , + α _max ] with α _max = π is recorded.

In 6 The distances of all points of this spiral segment from the image plane, which are measured in the direction of the z-axis and are related to the width S of a detector array, are shown for different inclination angles γ above the sine of the projection angle α multiplied by the radius R _f , specifically for the tilt angle δ = 0.

7 shows based on 6 assuming that the spiral segment under consideration contributes to the respective image in its entirety, the square roots of the square mean values of the square mean values of the distances of all points of the measured in the direction of the z-axis, which are plotted over the quotient γ / γ ₀ and refer to the width S of a detector line considered spiral segment from the image plane, which will be referred to as SMSD (Squareroot Mean Square Distance) in the following.

Out 7 it becomes clear that by optimizing γ from 3.5 S in the case of a completely uninclined image plane, ie γ = 0, SMSD is reduced to 2.2 S. It is assumed that those with the method according to US 5 802 134 achievable improvement in image quality is due to this reduction in SMSD. Incidentally, the value of the angle of inclination γ, for which SSMD is minimal, differs only slightly for the value of pitch p under consideration from the value of γ ₀ determined according to (8).

If not a single image is reconstructed with respect to the entire spiral segment, but instead the required number n _ima determined according to (12) is reconstructed from images with an inclined image plane, then a number n _ima results for the values M = 12 and p = 12 selected for the present example = 2. That is to say, for a total segment of length 2α _max = 2π = 360 °, two images are made for two spiral segments of length 180 ° plus cone angle, for example length 240 °, which are offset by 120 ° and thus together comprise the entire segment reconstructed with an inclined image plane. The image planes of the images have different z positions and thus, according to (11), different tilt angles δ, namely −δ _max and δ _max .

In 8th are for one of the spiral segments of length 240 °, which relate to the width S of a row of detector elements, measured in the direction of the z-axis distances of all points of this spiral segment from the inclined by −δ _max or + δ _max and by γ _{min in} each case Image planes of the two images belonging to this spiral segment are shown above the sine of the projection angle α multiplied by the radius R _f . First δ _max and γ _{0 were determined} on the basis of (8) and (10); For the purpose of optimization, the inclination angle γ was then rationed on the basis of δ _max , for which purpose SMSD was determined separately for the spiral segments considered at both image planes, an overall SMSD was then formed as the square root of the separately determined SMSDs and finally the inclination angle γ was reiterated was, which leads to γ _min = 1.26 · γ ₀ and a total SMSD of 0.8 S.

This corresponds to a reduction in SMSD compared to that from the US 5 802 134 known methods - tilt angle with δ = 0 and reconstruction of a single image from the overall segment - by a factor of more than 3 and promises a gain in image quality.

The image planes belonging to one of the two spiral segments, each with a length of 240 °, are exemplary in 9 and 10 shown in perspective from different angles. Especially from 10 it can be seen that the two inclined image planes intersect in a straight line tangent to the spiral, as mentioned.

The detector and thus dose usage for that from the US 5 802 134 known method is in 11 illustrated using the virtual detector for M = 12 and p = 8. The area delimited by the bold parallelogram-shaped line shows that area of the virtual detector surface onto which the inclined image plane belonging to the spiral segment is projected during the movement of the focus along the spiral segment.

It becomes clear that large parts of the detector area remain unused and accordingly the dose usage is low. A theoretically optimal detector and dose usage is only possible for the maximum pitch p _{m a x} = 12; as the pitch p decreases, detector and dose usage become worse and worse.

For the invention, the detector and thus dose usage is in 12 also shown for M = 12 and p = 8 using the virtual detector. The area delimited by the bold parallelogram-shaped line shows that area of the virtual detector surface onto which the image planes belonging to the spiral segment according to (12) n _ima = 3 are projected during the movement of the focus along the spiral segment.

It is clear that in the case of Invention most of it the virtual detector area - just two small ones triangular areas remain unused - is used and the dose usage is correspondingly high.

13 shows analog to 12 likewise for the invention the ratios for M = 12 and p = 12. Accordingly, the area delimited by the bold parallelogram-shaped line again indicates that area of the virtual detector surface on which the image planes belonging to the spiral segment according to (12) n _ima = 2 are inclined are projected along the spiral segment during the movement of the focus.

Like the comparison of 11 and 12 shows in the case of the invention in practice there is only a slight dependence of the detector and dose usage on the pitch p, to the extent that the two small unused triangular areas of the virtual detector surface gradually grow with decreasing pitch p.

So it becomes clear that contrary to that from the US 5 802 134 known method in the case of the invention the detector and thus the dose usage is largely independent of the pitch p and almost optimal.

The invention is also for investigations of the heart.

14 shows the maximum tilt angle δ _max determined according to (10) as a function of the pitch p for a CT device with M = 12. It can be seen that the invention for p = 16, δ _max = 0 in the from the US 5 802 134 known algorithm passes.

Under the use of Δz = R f tans Max (13) the angle of inclination δ can be transformed into a z-shift of the corresponding image. This corresponds to a shift in the reference projection angle α _r .

Accordingly, at any z position Images of spiral segments of 240 ° long are computed with respect to a range from [-Δα, Δα] Reference projection angles are centered.

Für T_rot = 0.5 s ist das für eine beliebige z-Position abgedeckte Zeitintervall in 15 als Funktion des Pitchs p dargestellt.If the gantry rotates in time T _red , this area corresponds to a time interval of length

For T _rot = 0.5 s, this is the time interval in for any z position 15 shown as a function of pitch p.

If a spiral segment of length 240 ° in its Whatever the case, fit the whole on the virtual detector surface a maximum pitch of p = 3 is available to to cover a time interval of one second, which is a complete cardiac cycle at a pulse rate of 60 beats per minute (60 bpm).

The image planes of the images which are obtained for the same reference projection angle α _r , ie from the same cardiac cycle, are shown in 16 illustrated. Reformatting is required to obtain transverse sectional images.

As mentioned, reformatting is required to obtain transverse sectional images, which is the case with conventional CT devices do not is required.

The multi-slice CT devices currently available have some few, e.g. 4, rows of detector elements. For this number of lines, the slope X-ray beam path neglected become. For such a CT device therefore became conventional Algorithms for the reconstruction of transverse sectional images from spiral data extended. After performing a spiral weighting with a suitable weighting function to determine a one-line data record is available for the reconstruction layer thickness, from that with a convolution-back-projection algorithm Transverse sectional image is reconstructed. The reconstruction layer thickness, i.e. the thickness of those in the reconstructed transverse sectional image detected layer of the object to be examined is selected by the Width of the weighting function used in spiral weighting established. A change of the reconstruction layer thickness is only by renewed reconstruction with changed Weight function possible.

As with the invention, the procedure according to the invention is especially not suitable for CT devices with too big Number of lines (M ≤ 40) there is an adaptation to the oblique beam path of the X-rays in the reconstruction, in which, as already explained, images for in their Inclination of image planes adapted to the spiral-like scanning geometry be reconstructed. Due to the inclination of the picture planes is after of the reconstruction, hereinafter referred to as reformatting Conversion of these images with respect to the System axis inclined image planes in transverse sectional images required. If this does not happen, it is especially in secondary views a reconstructed image volume (e.g. sagittal or coronal) to calculate geometric designations.

The reformatting happens with Interpolation functions of selectable width, which makes the Layer sensitivity profile and the image noise in the resulting Allow transversal sectional image to be influenced.

It is advantageous that the setting the desired one Reconstruction layer thickness retrospectively in the course of reformatting he follows.

und mit dem Nullpunkt im Punkt (-R_f, 0, z_R), indem (x, y, Δz_R) in die Ebenengleichung einsetzt wird n (δ, γ)·x = 0.The number of images with an inclined image plane required for the reformatting to be obtained in order to obtain a transverse sectional image at the z position z = z _R is obtained as follows:
At the edge of the object cylinder parameterized by (x, y) = (R _M cos (Φ), R _M sin (Φ)), the distance Δz _{R of} an image plane inclined by the angle of inclination and the angle of inclination with the normal vector is obtained

and with the zero point in the point (-R _f , 0, z _R ) by inserting (x, y, Δz _R ) in the plane equation n (δ, γ) x = 0 ,

It then follows:

rekonstruierten Bilder mit geneigter Bildebene zur Verfügung stehen, d.h. im Speicher 14 gespeichert werden.For the reformatting of a transverse sectional image with the image plane in z _R , all must therefore be in the interval

reconstructed images with an inclined image plane are available, ie in the memory 14 get saved.

If an interpolation function when reformatting is used, its length z * exceeds the limit values set by the above interval, so is the number of reconstructed required for reformatting Images with an inclined image plane due to the length of the interpolation filter certainly.

In general, the number N _M of the reconstructed images with an inclined image plane required for reformatting a transverse sectional image applies

N _{S is} the number of images reconstructed per width S of a row of detector elements with an inclined image plane.

For example, for a detector array with 16 rows of detector elements for a pitch of p = 16 and the number N _S = 4 of the images reconstructed per width S with an inclined image plane, the number N _M of the reconstructed images with an inclined image plane N required for reformatting a transverse sectional image is obtained _M = 10, provided that a triangular interpolation function of the half-width S is used.

Due to the fact that the reconstruction layer thickness a desired one Transversal sectional image is set retrospectively, the Reconstruction of the images with an inclined image plane preferably by Selection of a correspondingly narrow weighting function for spiral reconstruction with the least possible Reconstruction layer thickness. This not only ensures maximum sharpness in the z direction of the pictures with inclined picture planes, but also of that through the Reformat of the transverse sectional image obtained.

• – Die Rekonstruktionsschichtdicke kann retrospektiv gewählt werden, ohne dass eine erneute Rekonstruktion erforderlich ist,
• – die Rekonstruktionsschichtdicke ist frei wählbar, und
• – für die Reformatierung steht eine Vielzahl von geeigneten Interpolationsfunktionen frei wählbarer Breite zur Verfügung.

In addition to this advantage, further advantages of the reformatting described are:

• - The reconstruction layer thickness can be selected retrospectively without the need for a new reconstruction,
• - The reconstruction layer thickness is freely selectable, and
• - A large number of suitable interpolation functions of freely selectable width are available for reformatting.

If in the in 17 illustrated operation with inclined gantry 7 the axis of rotation Z ', about which the focus F rotates about the system axis Z, is not identical to the system axis Z, but intersects it at a so-called gantry angle ρ, results from the geometry according to 5 a according to 18 tilted coordinate system with the z'-axis corresponding to the central axis of the spiral path H, which is tilted with respect to the z-axis by the gantry angle ρ, the y'-axis, which is also tilted with respect to the y-axis by the gantry angle ρ, and the unchanged x-axis.

The following applies to spiral path H in this coordinate system:

The procedure described above for determining the maximum tilt angle δ _max can be applied to the case of the tilted gantry, the equation (7) being used instead.

However, the inclination angle γ 'in the coordinate system (x, y', z ') must now be used in the determination equation for the maximum tilt angle δ _max , ie in equation (10), in the case of the inclined gantry.

wobei s die Bogenlänge der Spiralbahn H für das jeweils betrachtete Spiralsegment ist.The following applies to the angle of inclination γ 'in the case of the inclined gantry:

where s is the arc length of the spiral path H for the spiral segment under consideration.

As the 19 shows, the angle of inclination γ 'for the case of the tilted gantry is almost independent of the reference projection angle α _r . It shows 19 the situation for a line number M = 16, a pitch p = 16 and a gantry angle of ρ = 30 °.

The maximum tilt angle δ _max is, as from the 20 can be seen, almost independently of the reference projection angle a _r , with the 20 shows the situation for the number of lines M = 16, the pitch p = 16 and the gantry angle ρ = 30 °. A shows the course of the maximum tilt angle δ _max for + RFOV and ρ = 30 °, while C shows the course of the maximum tilt angle for –RFOV and ρ = 30 °.

For comparison, in 20 the corresponding courses of the maximum tilt angle δ _{max are entered} for a gantry angle of ρ = 0 °, where B applies to + RFOV and ρ = 0 °, while D applies to –RFOV and ρ = 0 °.

To illustrate the effect of the invention is in 21 analogous to 19 the course of the angle of inclination γ as a function of the reference projection angle also for a number of M = 16, a pitch p = 16 and a gantry angle of ρ = 30 ° for that from the US 5 802 134 known methods shown. It is clear that there is a strong dependence of the angle of inclination γ on the reference projection angle α _r .

The 19 to 21 each show a full rotation (360 °) of focus F.

Also in the case of the inclined gantry, there is the possibility, for a given amount of the maximum value of the tilt angle | δ _max |, which is obtained, for example, from (10) based on the slope of the spiral path H according to (21), the associated optimal value of the angle of inclination To determine γ 'in such a way that an error criterion, for example a minimum mean value of the distances of all points of the spiral segment from the image plane measured in the z direction, is met.

The structure of the image calculator 11 is described in the case of the above embodiment in a manner as if the preprocessing unit 12 and the reconstruction unit 13 Hardware components. Indeed, this can be so. As a rule, however, the components mentioned are implemented by software modules which run on a universal computer provided with the required interfaces, which differs from that 1 also the function of the then superfluous control unit 18 can take over.

The CT device in the case of the exemplary embodiment described has a detector array 5 with lines whose width measured in the z direction is the same and is, for example, 1 mm. Deviating from this, a detector array can also be provided within the scope of the invention, the rows of which are of different widths. For example, two inner lines, each with a width of 1 mm, and a line with a width of 2 mm can be provided on either side of them.

In the case of the exemplary embodiments described, the relative movement between the measuring unit 1 and the storage device 9 each generated by the storage device 9 is moved. However, there is also the possibility within the scope of the invention of the storage device 9 to leave it stationary and instead the measuring unit 1 to postpone. In addition, there is the possibility within the scope of the invention of the necessary relative movement by moving both the measuring unit 1 as well as the storage device 9 to create.

In connection with the above described embodiments find CT machines 3rd generation use, i.e. the x-ray source and that Detector array are during the image generation together around the system axis. The invention can also be used in connection with CT devices the 4th generation, where only the X-ray source around the System axis is shifted and with a fixed detector ring interacts, find use if it is in the detector array for an areal Array of detector elements.

Also with 5th generation CT devices, i.e. CT scanners, where the x-rays not just from one focus, but from multiple foken one or several x-ray sources shifted around the system axis can the inventive method Find use if the detector array has a large area Has array of detector elements.

The related to the above described embodiments used CT devices have a detector array with an orthogonal matrix arranged detector elements. The invention can also in connection with CT devices Find their detector array in a way other than area-explanatory Array arranged detector elements.

The exemplary embodiments described above relate to the medical application of the method according to the invention. However, the invention can also be used outside of medicine, for example at the baggage check or used in material testing.

Claims (34)

Procedure for computer tomography, showing the process steps a) for scanning an object with a focus conical ray beam and with a matrix-like detector array for detecting the beam the focus relative to the object on a spiral path around a system axis moves, the detector array corresponding to the received radiation Provides output data, and b) it will be made each during the Movement of focus on a spiral segment provided output data Reconstructed images with an inclined image plane, the image planes both around a first axis that intersects the system axis at right angles by an angle of inclination γ as also around a second one, both the first axis and the system axis right-angled axis by a tilt angle δ with respect to System axis are inclined. Method according to Claim 1, in which images with an inclined image plane are reconstructed for a number n _{ima of} successive spiral segments, the image planes having the same z position z _ima , and immediately successive spiral segments being offset from one another by at most 180 ° and an overall segment of length [ –Α _max , + α _max ], where α _max = Mπ / p and M is the number of detector lines. Method according to Claim 1, in which each spiral segment has a length of 180 ° plus cone angle and in which images with an inclined image plane are reconstructed for a number n _ima of inclined image planes for each spiral segment, the image planes having the different z positions z _ima . The method of claim 3, wherein the plurality of inclined Image planes in a straight line tangent to the spiral to cut. Method according to Claim 3 or 4, in which the following applies to the extreme values + δ _max and −δ _{max of} the tilt angle δ of the inclined image planes belonging to a spiral segment:

for the tilt angle δ = 0 is the value of the inclination angle γ. Method according to one of claims 1 to 5, wherein the focus rotates about a rotation axis around the system axis, that of the system axis equivalent. Method according to one of Claims 1 to 5, in which the axis of rotation about which the focus rotates about the system axis intersects the system axis at a gantry angle ρ, the following being valid for the angle of inclination γ 'to be selected:

Method according to one of claims 5 to 7, in which for a given amount | δ _max | of the maximum value of the tilt angle δ, the associated optimum value γ _{min of} the tilt angle γ is determined in such a way that an error criterion is met Method according to one of Claims 3 to 8, in which the following applies to the number n _{ima of} the inclined image planes for which images with an inclined image plane are generated for each spiral segment:

The method of claim 9, wherein the tilt angle δ of the inclined image planes after

be determined. Method according to one of claims 1 to 10, comprising the a further method step that a transverse sectional image of a the system axis generates a transverse layer intersecting at right angles is grouped together by several images with an inclined image plane become. The method of claim 11, wherein the summarizing the several images with an inclined image plane to form a transverse sectional image done by interpolation. The method of claim 12, wherein the summarizing the several images with an inclined image plane to form a transverse sectional image done by averaging. The method of claim 13, wherein the summarizing the several images with an inclined image plane to form a transverse sectional image by weighted averaging. A method according to any one of claims 11 to 14, which further comprises the Method step has that the number of images inclined Image plane, which is combined to generate a transverse sectional image are, according to the desired layer thickness of the transverse layer chosen becomes. The method of claim 15, wherein the images are tilted Image plane with the lowest possible layer thickness be reconstructed. The method of claim 15 or 16, wherein the number of images with an inclined image plane, which are combined to produce a transverse sectional image, after

is chosen. Computed tomography (CT) device having a radiation source, the focus of which emits a conical beam of rays, a matrix-like detector array for detecting the beam, wherein the detector array corresponding to the received radiation output data provides means for generating a relative movement between the radiation source and detector array on the one hand and an object on the other hand and one Computer to which the output data are supplied, the means to generate a relative movement for scanning the object with the bundle of rays and the two-dimensional detector array a relative movement of the Cause focus to a system axis such that the focus relative to the system axis on a spiral path around the system axis moves, and taking the calculator off each while moving the focus output data supplied on a spiral segment images with inclined Reconstructed image plane, the image planes both around a first, the system axis intersecting right angle by the angle of inclination γ as well around a second, both the first axis and the system axis right-angled axis about the tilt angle δ with respect to System axis are inclined. The CT apparatus of claim 18, wherein the computer reconstructs images with an inclined image plane for a number n _{ima of} successive spiral segments, the image planes having the same z position z _{i ma} , and immediately successive spiral segments being offset from one another by at most 180 ° and one Total segment of length [–α _max , + α _max ] result, where α _max = Mπ / p and M is the number of detector lines. A CT device according to claim 18, in which each spiral segment has a length of 180 ° plus a cone angle and in which the computer reconstructs images with an inclined image plane for a number n _ima of inclined image planes for each spiral segment, the image planes representing the different z positions z exhibit _{in a} . CT machine 21. The claim 20, wherein the plurality of inclined image planes cut in a straight line tangent to the spiral. CT device according to Claim 20 or 21, in which the following applies to the extreme values + δ _max and −δ _{max of} the tilt angle δ of the inclined image planes belonging to a spiral segment:

for the tilt angle δ = 0 is the value of the inclination angle γ. CT machine, according to one of the claims 18 to 22, the focus of which around a rotation axis around the system axis rotates, which corresponds to the system axis. A CT device according to any one of claims 18 to 22, wherein the axis of rotation about which the focus rotates about the system axis intersects the system axis at a gantry angle ρ, the computer according to the angle of inclination y '

chooses. A CT device according to any one of claims 22 to 24, wherein the computer for a given amount | δ _max | of the maximum value of the tilt angle δ the associated optimum value γ _{min of} the inclination angle γ is determined in such a way that an error criterion is fulfilled. CT device according to one of Claims 20 to 25, in which the following applies to the number n _{ima of} the inclined image planes for which the computer generates images with an inclined image plane for each spiral segment:

A CT device according to claim 26, wherein the computer reads the tilt angle δ of the inclined image planes

determined. CT machine according to one of the claims 18 to 27, in which the computer is a transverse sectional image of a System axis creates a transverse layer intersecting at right angles, by combining several images with an inclined image plane. CT machine The method of claim 28, wherein the computer combines the plurality Images with an inclined image plane to a transverse sectional image Interpolation causes. CT machine The method of claim 29, wherein the calculator combines the plurality Images with an inclined image plane to a transverse sectional image Averaging causes. CT machine 31. The method of claim 30, wherein the computer combines the plurality Images with an inclined image plane to a transverse sectional image weighted averaging. CT machine according to one of the claims 28 to 31, in which the computer also includes the number of images inclined image plane, which he used to create a transverse sectional image summarizes, according to the desired layer thickness of the transverse layer chooses. The method of claim 32, wherein the computer includes the images reconstructed inclined image plane with the lowest possible layer thickness. The method of claim 32 or 33, wherein the computer after the number of images with an inclined image plane, which it combines to generate a transverse sectional image, after

chooses.