Method and apparatus for generating result image based on tomosynthesis image dataset of chest

Abstract

The invention relates to a method for generating a result image (E) on the basis of a tomosynthesis image data set (T) of a chest (O), the chest being fastened to a probe (5) by means of a compression plate (6), the contact surface of the compression plate (6) and / or of the probe (5) facing the chest (O) having a concave arch, the method comprises the following steps: defining an arched layer (S) between the compression plate (6) and the detector (5), the arched layer following the arches of the compression plate (6) and / or the detector (5) or transitions between the arches in terms of its shape; reconstructing a layer image (B), said layer image being on the curved layer (S); generating a result image (E) of the layer stack as the layer image (B); and outputting the result image (E). The invention also relates to a corresponding device and a mammography system.

Description

Technical Field

[0001] The invention relates to a method and an apparatus for generating a result image based on a tomosynthesis image data set of a breast and a corresponding mammography system. The invention particularly concerns non-planar post-processing of spectral X-ray images. Background Art

[0002] When studying the female breast, tomosynthesis is a favorable technique for creating three-dimensional images. Typically, in digital breast tomosynthesis, a cubic volume in which the breast is located is reconstructed above the detector. The height of the volume is typically defined by the compressed thickness of the breast.

[0003] The cubic volume contains the planar sections through the chest. In this case, a reconstruction can be performed so that the planar sections within the volume are reconstructed into a stereoscopic coordinate system. The stereoscopic coordinate system helps in this case to scan along the spatially varying direction of the Z resolution, which is a key feature of DBT reconstruction. In this scan, each beam can typically be parameterized by means of two angles α and φ relative to the normal vector of the detector (Z axis).

[0004] In case of curved compression paddles or alternative compression or fixation methods that preserve the natural curved shape of the breast, such reconstruction with planar cross sections results in a non-optimal display of the DBT cross sections that only show a small part of the breast in the cross sections that are located outside. Summary of the invention

[0005] The object of the present invention is to specify an improved method and an improved device for generating result images based on tomosynthesis image data sets of a breast, by which the above-mentioned disadvantages are avoided and which in particular enable non-planar post-processing of spectral X-ray images.

[0006] This object is achieved by a method according to the invention, by a device according to the invention, and by a mammography system according to the invention.

[0007] The method according to the invention is used to generate a result image based on a tomosynthesis image data set of a breast which is fixed on a detector by means of a compression plate. In this case, the contact surface of the compression plate and / or the detector facing the breast has a concave curvature. As a result, the breast is not held between two flat plates and thus compressed at the top and bottom relative to a flat surface, but has a curvature at the top and / or bottom. Since the plates are curved concavely, the breast located therebetween is curved convexly.

[0008] The method comprises the following steps:

[0009] a curved layer is defined between the compression plate and the detector, said curved layer following in its shape the curvature of the compression plate and / or the detector or being a transition between said curvatures,

[0010] - reconstructing a (curved) slice image, said slice image being located on a curved layer,

[0011] - generating a resulting image as a layer stack of (curved) layer images,

[0012] - Output the resulting image.

[0013] The chest is thus held against the detector by the compression plate during the recording. At least one of the two surfaces compressing the chest is concavely curved. This means that the surface allows a convex curvature of the chest. The concave curvature can be achieved, for example, by the compression plate being curved downward at its edge, like a shell, or by the compression plate being a flat plate with a convex recess at its underside. However, the surface above the detector can also have depressions. Thus, the upper side and / or the lower side of the recorded chest is convexly curved.

[0014] Now, corresponding to the curvature of the breast, a curved layer is defined between the compression plate and the detector, the curvature of the breast corresponding in its shape to the curvature of the compression plate and / or the detector. The curved layer follows the curvature of the breast or is a transition between different curvatures. Preferably, the outermost layer substantially follows the curvature of the breast, wherein "substantially" corresponds to up to at least 80%, in particular up to at least 90%. The layers in between should be (in particular regular) transitions between the curvatures of the outermost layer.

[0015] Here, two situations can be fundamentally distinguished: one in which the breast is arched on one side and flat on the other side (wherein the flat layer on the flat side corresponds there to the "arch" of the breast), and the other in which the breast is arched at the top and bottom and the upper arch bulges upwards, while the lower arch bulges downwards.

[0016] The intervening layers should then form the transition from curvature to flat surface or between upward curvature and downward curvature (optionally with a flat layer in the middle).

[0017] Image reconstruction is basically known in the prior art. However, in the method described here, instead of reconstructing flat (planar) slice images, slice images are reconstructed on curved slices (in which case one of the slices can also be flat, for example the middle slice of a biconvex breast or the outermost slice of a monoconvex breast). This results in essentially curved slice images.

[0018] It should be noted here that a slice is a theoretical plane, by means of which reconstruction is performed, and a slice image is a finally reconstructed image.

[0019] A result image is then generated from the slice images and output as a slice stack, for example the slice images are simply placed one above the other.

[0020] Thus, the (curved) slice images are directly reconstructed in the recorded volume. The curvature can be based on the known curvature of the compression plate or the surface above the detector. This direct reconstruction has the advantage that no (planar) slices that are too short are produced.

[0021] The device according to the invention is used to generate a result image based on a tomosynthesis image data set of a breast, which is fixed on the detector by means of a compression plate, wherein the contact surface of the compression plate and / or the detector facing the breast has a concave curvature. The device is preferably designed to carry out the method according to the invention and comprises the following components:

[0022] a layer unit which is designed to define a curved layer between the compression plate and the detector, the curved layer following in its shape the curvature of the compression plate and / or the detector or forming a transition between the curvatures,

[0023] a reconstruction unit designed to reconstruct a (curved) slice image located on a curved slice,

[0024] a result unit which is designed to generate a result image of the layer stack as a (curved) layer image,

[0025] - A data interface designed to output the resulting image.

[0026] The functions of the units have been described above in the context of the method.

[0027] The mammography system according to the invention comprises the device according to the invention and / or is designed to carry out the method according to the invention.

[0028] The invention can be implemented in particular in the form of a computing unit with suitable software, in particular in a control device for a mammography system. The computing unit can, for example, have one or more microprocessors or the like working together for this purpose. The microprocessor can be implemented in particular in the form of a suitable software program part in the computing unit. The large-scale software implementation has the advantage that the computing units that have been used so far can also be retrofitted in a simple manner by software updates or firmware updates in order to work in the manner according to the invention. In this regard, the object is also achieved by a corresponding computer program product with a computer program, which can be directly loaded into the storage device of the computing unit, and the computer program product has program segments so that when the program is executed in the computing unit, all steps of the method according to the invention are performed. In addition to the computer program, such a computer program product may possibly include additional components, such as file compilations and / or additional components, as well as hardware components, such as hardware keys (software dongles, etc.) for using the software. A computer-readable medium, such as a memory stick, a hard disk or other transportable or permanently installed data carrier, on which program segments of a computer program that can be read in and executed by the computing unit are stored, can be used for transport to the computing unit and / or for storage at or in the computing unit.

[0029] Further particularly advantageous designs and improvements of the invention are described below, wherein the implementation of the method can also be improved similarly to the implementation and description of the device, and individual features of different embodiments or variants can also be combined to form new embodiments or variants.

[0030] According to a preferred method, the curvature of the first layer adjacent to the compression plate substantially corresponds to the shape of the contact surface of the compression plate. As mentioned above, "substantially" here means that the curvature preferably corresponds to up to 80%, in particular up to 90% (e.g. the curvature radius) of the curvature of the compression plate. The curvature can be identical to the curvature of the compression plate, but in the case of a first layer slightly away from the compression plate, the curvature of the layer is preferably already slightly adapted to the opposite curvature.

[0031] According to a preferred method, the curvature of the second layer adjacent to the detector substantially corresponds to the shape of the contact surface of the detector. Here, “substantially” also means that the curvature preferably corresponds to up to 80%, in particular up to 90% (e.g. the curvature radius) of the curvature of the corresponding surface. The curvature can be identical to the curvature of the surface, but in the case of a second layer slightly removed from the surface, the curvature of the layer is preferably already slightly adapted to the opposite curvature.

[0032] According to a preferred method, the curvature of the intervening layer follows a regular transition between the curvature of the first layer and the second layer. Such a regular transition is, for example, a regular transition of the radius of curvature.

[0033] Preferably, the contact surface of the detector is flat and the second layer is also flat. The curvature of the further layers then preferably increases continuously starting from the second layer up to the first layer.

[0034] According to a preferred method, the curvature of the first layer and / or the second layer is determined based on the measurement of the respectively adjacent contact surface. This can be performed simply because not only the surface of the compression plate but also the surface above the detector is known (or can be well measured). It should be noted that the compressed breast follows this shape so that the shape of the breast is known or can be easily determined.

[0035] According to a preferred method, the curvature of the first layer and / or the second layer is determined based on data from a sensor measuring the shape of the surface of the breast. This approach has the advantage that errors in adapting the breast to the surface pressing on it are avoided.

[0036] According to a preferred method, a coordinate system (x n ,y n ,z n ), where (x n ,y n ) are in the same layer and z n On the surface (x n ,y n ) defines the Z position (ie the height above the detector) of a point of the curved layer. Thus, the coordinate system directly defines the curved layer.

[0037] Specifically, the coordinate (x n ,y n ,z n ) can be calculated in Cartesian coordinates with the help of the function F. In a simple example, the (x n ,y n ) can correspond to the (x,y) of the flat layer of the Cartesian coordinate system, that is, (x n ,y n )=(x,y). In contrast to the flat layer, the z of the curved layer n It can be calculated from a defined function F which defines the curvature. In this case, the layers can be calculated, for example, as follows:

[0038] - Individualized function F for each layer n , z n =F n (x,y),

[0039] - a common, n-dependent function F for all layers, z n =(x,y,n),

[0040] - for a selected value z, for example a flat layer, the z-dependent function F of all layers,

[0041] z n =F(x,y,z).

[0042] According to a preferred method, the reconstruction of the slice image is based on the calculation of the path of the X-rays penetrating the slice of the arch. For a preferred projection of the slice, the coordinate system (x n ,y n ,z n ) and a coordinate pair between a coordinate system (x, y, z), in which multiple rays having angles (α, φ) between their intersections with the detector and the object surface are scanned at equidistant steps.

[0043] For the purpose of illustration, it is conceivable that when reconstructing a slice image from a tomosynthesis image data set (e.g., consisting of a plurality of projection images), one starts from each voxel from which a voxel value is calculated from the tomosynthesis image data set using known calculation operations (in a Cartesian coordinate system). The position (x n ,y n ,z n ). Therefore, we can also n ,y n ,z n ) calculates the voxel value for each voxel at . Thus, the required voxel values ​​for all curved layers are obtained.

[0044] It should be noted that X-rays are usually cone-shaped during recording. This should be taken into account during reconstruction. The cone shape has the effect that, on the one hand, multiple beams with different angles (α, φ) are emitted in each projection image of the tomosynthesis image data set, and that the beam intensity in the voxels of the upper layer is greater than the beam intensity in the voxels of the lower layer. However, this can be compensated by a correction calculation. The correction calculation is known in the Cartesian coordinate system. Here, the position (x ) of each point of the curved layer can now also be determined in the Cartesian coordinate system. n ,y n ,z n ). Therefore, we can also n ,y n ,z n) is calculated for each voxel at the curved layer. The required compensation for all curved layers is thus obtained. According to a preferred method, the cone shape of the recorded X-rays is reconstructed. In this case, the positions (x ) of a plurality of points of the curved layer are determined (in the Cartesian coordinate system). n ,y n ,z n ) and correction calculations are performed on these points to compensate for the conical shape of the X-rays.

[0045] Then, in the backprojection loop, specific calculations or searches can be performed based on the coordinate pairs (this should be optimized with regard to performance). The calculations are performed in the coordinate system (x n ,y n ,z n ) into a point in the coordinate system (x, y, z). In a Cartesian coordinate system, this can preferably be done via a simple linear transformation.

[0046] Therefore, according to a preferred method, a three-dimensional image of the breast is created as a result image in Cartesian coordinates. Preferably, the image coordinates of the slice image are converted into Cartesian coordinates by means of a linear transformation.

[0047] According to a preferred method, the slices are oriented orthogonally to the surface normal of the contact surface of the detector. The slice images are then stacked one on top of the other in the direction of the surface normal to form a slice stack. This enables a very simple construction of the resulting image.

[0048] The present invention describes an advantageous sampling method which takes into account the shape of the breast and enables maximum utilization of the content from the layer images. It is also advantageous that non-planar sampling geometries can be used directly for reconstruction. This avoids additional interpolation steps which, for example, can make visibility of micro-classification difficult. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The present invention will be described in detail again according to the embodiments with reference to the accompanying drawings. In this case, the same components are provided with the same reference numerals in different figures. The accompanying drawings are usually not to scale. The accompanying drawings show:

[0050] Figure 1 A rough schematic diagram showing a preferred mammography system with preferred equipment,

[0051] Figure 2 A flow chart showing a possible sequence of steps of the method according to the invention,

[0052] Figure 3 shows an example of a compressed breast with reconstruction using flat layers according to the prior art,

[0053] Figure 4 shows an example of a curved compressed breast with reconstruction using flat layers according to the prior art,

[0054] Figure 5 shows an example of a curved compressed breast with reconstruction of layers by means of curves according to the method according to the invention,

[0055] Figure 6 An example of a curved compressed breast with reconstruction of layers by means of curves according to the method according to the invention is shown. DETAILED DESCRIPTION

[0056] exist Figure 1 , a mammography system 1 in the form of a tomosynthesis system 1 is shown by way of example and roughly schematically. Relative directional indications such as “above”, “below”, etc. relate to the tomosynthesis system 1 set up for operation as specified. The tomosynthesis system 1 comprises a tomosynthesis device 2 and a control device 9 .

[0057] The tomosynthesis device 2 has a column 7 and a source-detector arrangement 3, which in turn comprises an X-ray emitter 4 and a detector 5 with a detector surface 5.1. The column 7 stands on a foundation during operation. The source-detector arrangement 3 is movably connected to the column so that the height of the detector surface 5.1, i.e. the distance from the foundation, can be set to the chest height of the patient.

[0058] The patient's chest O (shown schematically here) is placed on the upper side on the detector surface 5.1 as an examination object O for examination. A compression plate 6 is arranged above the chest O and the detector surface 5.1, and the compression plate 6 is movably connected to the source detector device 3. For the examination, the chest O is compressed and fixed at the same time in that the compression plate 6 is lowered onto the chest so that pressure is applied to the chest O between the compression plate 6 and the detector surface 5.1. The contact surface of the compression plate 6 facing the chest has a concave curvature, so that the chest is convexly curved there, as for example in Figure 4 , Figure 5 and Figure 6 As shown in .

[0059] The X-ray emitter 4 is arranged opposite the detector 5 and is designed in such a way that the detector 5 detects the X-ray radiation R emitted by the X-ray emitter after at least part of the X-ray radiation R has penetrated the patient's chest O. The X-ray emitter 4 can be pivoted relative to the detector 5 by means of a swivel arm 8 in a range of ±50° about a basic position in which the swivel arm stands vertically above the detector surface 5.1. The section to be recorded can be defined or limited by means of a collimator C.

[0060] The control device 9 obtains the measured raw data RD and sends control data SD to the tomosynthesis device 2 via a data interface. The data interface is connected to a terminal 20, via which a user can communicate instructions to the tomosynthesis system 1 or retrieve measurement results. The control device 9 can be arranged in the same space as the tomosynthesis device 2, but can also be in an adjacent control space or at a greater spatial distance.

[0061] The device 10 according to the invention for generating a result image based on a tomosynthesis image data set of a breast is in this case part of the control device 9 and comprises a slice unit 11, a reconstruction unit 12, a result unit 13 and a data interface 14 (see also according to Figure 2 method).

[0062] The layer unit 11 defines a curved layer S between the compression plate 6 and the detector 5, the curved layer S following the curvature of the compression plate 6 and / or the detector 5 or being a transition between the curvatures in its shape. Figure 5 and Figure 6 Shown in.

[0063] The reconstruction unit 12 is used to reconstruct a slice image B, which is located on a curved slice S. Therefore, the slice image B is substantially equally curved and results in a section through the breast O along one of the slices S.

[0064] Then, the result unit 13 generates a result image E which is a layer stack of the layer images B. For example, the result unit arranges the layer images B in a stacked form.

[0065] Then, the result image E is output via the data interface 14 .

[0066] Figure 2 A flow chart showing a possible sequence of steps of a method according to the invention for generating a result image E based on a tomosynthesis image data set T of a breast O, which has been acquired, for example, by means of a Figure 1 On the left is shown how a compressed breast is recorded and how a tomographic data set T is provided to the method.

[0067] In step I, a curved layer S is defined between the compression plate 6 and the detector 5, which follows the curvature of the compression plate 6 and / or the detector 5 or is a transition between the curvatures in its shape. The region in which the chest O is located during the recording is of particular interest for the method.

[0068] Then, in step II, a layer image B is reconstructed, which is located on the curved layer S. The reconstruction of layer-dependent layer images is well known. A feature of this is that at the location of the flat layer (see, for example, Figure 3 and Figure 4 ) starts from the arched layer S.

[0069] In step III, a result image E is created as a slice stack of slice images B. The slices S are oriented, for example, orthogonally (ie generally horizontally) to the surface normal of the contact surface of the detector 5. The slice images B are stacked one on top of the other in the direction of the surface normal (ie here along a vertical line) to form a slice stack.

[0070] Then, in step IV, the result image E is output.

[0071] Figure 3 and Figure 4 An example of a compressed chest O with reconstruction using flattened layers according to the prior art is shown.

[0072] Figure 3 Here, a chest O is shown which is compressed by two flat surfaces. In this case, a flat layer (dashed line) is advantageous because it divides the space between the compression plate 6 and the detector 5 evenly.

[0073] In contrast, flat layers Figure 6 is disadvantageous in Figure 6 The middle chest O is compressed by means of the curved compression plate 6 , since the flat slice at the top (where the curve is located) only covers a small part of the chest O. Therefore, the reconstructed slice image B shows only a small slice of the chest O there.

[0074] Figure 5 and Figure 6 An example of a compressed chest O with a reconstruction of a curved layer S according to the method according to the invention is shown. The curvature of the first layer S (upper part) adjacent to the compression plate 6 corresponds essentially to the shape of the contact surface of the compression plate 6. The curvature of the second layer S (lower part) adjacent to the detector 5 corresponds essentially to the shape of the contact surface of the detector 5. The curvature of the layer S located in between follows a regular transition between the curvature of the first layer S and the second layer S. In this case, the contact surface of the detector 5 is flat and thus the second layer S is also flat.

[0075] Finally, it is pointed out once again that the method described in detail above is only an embodiment, which can be modified by a person skilled in the art in completely different ways without departing from the scope of the invention. In addition, the use of the indefinite article "a" or "an" does not exclude that the features involved can also be present in multiple forms. Similarly, the term "unit" does not exclude that the components involved are composed of multiple cooperating subcomponents, which can also be distributed spatially if necessary. The term "multiple" should be interpreted as "at least one".

Claims

1. A method for generating a result image (E) based on a tomosynthesis image data set (T) of a breast (O), the breast being fixed to a detector (5) by means of a compression plate (6), wherein the contact surface of the compression plate (6) and / or the detector (5) facing the breast (O) has a concave curvature, the method comprising the following steps: - a curved layer (S) is defined between the compression plate (6) and the detector (5), said curved layer (S) following in its shape the curvature of the compression plate (6) and / or the detector (5) or being a transition between said curvatures, - reconstructing a layer image (B), said layer image (B) being located on said curved layer (S), - generating said result image (E) as a layer stack of said layer images (B), - Outputting the resulting image (E).

2. A method according to claim 1, wherein the curvature of the first layer (S) adjacent to the compression plate (6) substantially corresponds to the shape of the contact surface of the compression plate (6) and the curvature of the second layer (S) adjacent to the detector (5) substantially corresponds to the shape of the contact surface of the detector (5), and the curvature of the layer (S) in between follows a regular transition between the curvatures of the first layer (S) and the second layer (S), Preferably, the contact surface of the detector (5) is flat and the second layer (S) is also flat, and the curvature of the further layer (S) increases continuously starting from the second layer (S) to the first layer (S).

3. The method according to claim 2, wherein the curvature of the first layer (S) and / or the second layer (S) is determined based on measurements of respectively adjacent contact surfaces or based on data of a sensor measuring the surface shape of the breast (O).

4. The method according to any one of the preceding claims, wherein a coordinate system (x n ,y n ,z n ), where (x n ,y n ) are in one layer (S) respectively for the same n and z n On the surface (x n ,y n ) defines the position of the point of the layer (S) of the arch, Preferably, by means of a defined function F, particularly preferably a plurality of functions F for the individual layers (S) n , functions related to n or functions related to z, calculate the coordinates (x n ,y n ,z n ), wherein for z in particular the coordinates of a point on a flat layer are selected.

5. The method according to claim 4, wherein the reconstruction of the layer image (B) is based on the calculation of the path of the X-rays penetrating the layer (S) of the arch, preferably wherein the projection for the layer (S) is determined in the coordinate system (x n ,y n ,z n ) and a coordinate pair between a coordinate system (x, y, z), in which a plurality of rays having angles (α, φ) between their intersection points with the detector and the object surface are scanned at equidistant steps.

6. The method according to claim 4 or 5, wherein the reconstruction takes into account the cone shape of the recorded X-rays, wherein a plurality of points (x n ,y n ,z n ) and a correction calculation is performed on the point to compensate for the conical shape of the X-ray.

7. The method according to any of the preceding claims, wherein a three-dimensional image of the breast is created as a result image (E) in Cartesian coordinates, preferably wherein the image coordinates of the layer image (B) are converted into Cartesian coordinates by means of a linear transformation.

8. A method according to any one of the above claims, wherein the layer (S) is oriented orthogonal to the surface normal of the contact surface of the detector (5) and the layer image (B) is stacked one on top of the other in the direction of the surface normal to form the layer stack.

9. A device (10) for generating a result image (E) based on a tomosynthesis image data set (T) of a breast (O), the breast being fixed on a detector (5) by means of a compression plate (6), wherein the contact surface of the compression plate (6) and / or the detector (5) facing the breast (O) has a concave curvature, the device (10) comprising: a layer unit (11) which is designed to define a curved layer (S) between the compression plate (6) and the detector (5), the curved layer (S) following the curvature of the compression plate (6) and / or the detector (5) or being a transition between the curvatures in terms of its shape, a reconstruction unit (12) designed to reconstruct a layer image (B) located on the curved layer (S), a result unit (13) which is designed to generate the result image (E) as a layer stack of the layer images (B), A data interface (14) which is designed to output the result image (E).

10. A mammography system (1) comprising an apparatus (10) according to claim 9 and / or designed to perform a method according to any one of claims 1 to 8.

11. A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to perform the steps of the method according to any one of claims 1 to 8.

12. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method according to any one of claims 1 to 8.

Images