DE102009023892B4 - Processing method for an X-ray image of an examination object detected by means of an X-ray detector - Google Patents

Abstract

Processing method for an X-ray image (B) of an examination subject (5) captured by means of an X-ray detector (3), - with a mask area (14) which is smaller than the X-ray image (B), within which the X-ray image (B) is to be processed, - wherein for each location (P) lying within the mask area (14) in which fluctuations in the image data values (I) of the X-ray image (B) within a neighborhood (U) of the respective location (P), which is small compared to the mask area (14), is below a predefined threshold (S) remain, a preliminary correction factor (g ') is determined, - a respective final correction factor (g) is determined based on the determined preliminary correction factors (g') for each location (P) lying within the mask area (14) - wherein for each location (P) lying within the mask area (14) the image data value (I) of the respective location (P) on the basis of the mask value (M) at the same location and the respective final correction rfactor (g) is corrected, - whereby the respective preliminary correction factor (g ') is corrected using the two ...

Description

The present invention relates to a processing method for an X-ray image of an examination subject detected by means of an X-ray detector.

The present invention furthermore relates to a computer program which comprises machine code which can be processed directly by a computer and whose execution by the computer causes the computer to carry out such a processing method.

The present invention further relates to a data carrier on which such a computer program is stored in machine-readable form.

Finally, the present invention relates to a computer having a mass storage, wherein in the mass memory in machine-readable form such a computer program is stored, wherein the computer program is executable by the computer.

The above objects are well known.

In order to generate X-ray images, digital imaging X-ray systems generally use imaging arrangements which have an X-ray source and, as X-ray detectors, a flat detector. The examination subject (often a human) is placed between the x-ray source and the flat detector. The flat detector extends flat across the X-ray direction seen. Mostly it is rectangular. The flat detector is capable of detecting a two-dimensional spatially resolved X-ray image by means of a single image.

In order to capture a large X-ray image, the X-ray detector must also have the appropriate size. This is often problematic in the prior art. For this reason, in the prior art often two or four sub-detectors are used, which are arranged adjacent to each other, so that the adjoining sub-detectors each form a joint edge.

Although the use of multiple sub-detectors leads to be able to detect a large X-ray image in a relatively simple manner. However, in the region of the abutment edge - or in the case of more than two subdetectors of the abutting edges - it can happen that the individual detector elements of the subdetectors exhibit an amplification behavior which deviates from the amplification behavior of the other detector elements of the subdetectors. Even with other X-ray detectors - ie with flat detectors that have no abutting edge - such effects can occur. Also, the effects may occur outside the vicinity of the abutting edge. As a rule, however, the effects are limited to a relatively small area of the X-ray detector.

Within the range in which variations of the amplification behavior occur, the extent of the deviations - generally speaking - depends on the irradiation history of the detector elements and possibly on other circumstances, which are still unknown. However, the range in which a deviation of the amplification behavior occurs is as such independent of the X-ray dose. An exception can only apply if very high doses occur in which the X-ray detector does not become linear.

The deviations in the amplification behavior can lead to a faulty image analysis, which in extreme cases can also lead to faulty medical diagnoses. It is therefore necessary to take into account the different amplification behavior.

From the US 2009/0080756 A1 a processing method for an X-ray image of an examination object detected by means of an X-ray detector is known, which is also provided in particular for the correction of artifacts resulting from butting in flat-panel detectors.

From the publication US 4,611,283 A For example, there is known a method of operating radiographic imaging apparatus that allows, among other things, the correction of non-linearities and nonuniformities.

From the US 6,919,568 B2 For example, a method of identifying a composite defect pixel card is known.

The object of the present invention is to provide possibilities by means of which the deviations in the amplification characteristic can be corrected.

The object is achieved by a treatment process with the features of claim 1. Advantageous embodiments of the treatment process according to the invention are the subject of the dependent claims 2 to 7.

• – dass ein im Vergleich zum Röntgenbild kleiner Maskenbereich gegeben ist, innerhalb dessen das Röntgenbild aufbereitet werden soll,
• – dass für jeden innerhalb des Maskenbereichs liegenden Ort, bei dem Schwankungen der Bilddatenwerte des Röntgenbildes innerhalb einer im Vergleich zum Maskenbereich kleinen Umgebung des jeweiligen Ortes unterhalb einer vordefinierten Schwelle bleiben, ein vorläufiger Korrekturfaktor ermittelt wird,
• – dass anhand der ermittelten vorläufigen Korrekturfaktoren für jeden innerhalb des Maskenbereichs liegenden Ort ein jeweiliger endgültiger Korrekturfaktor ermittelt wird,
• – dass für jeden innerhalb des Maskenbereichs liegenden Ort der Bilddatenwert des jeweiligen Ortes anhand des ortsgleichen Maskenwertes und des jeweiligen endgültigen Korrekturfaktors korrigiert wird,
• – dass der jeweilige vorläufige Korrekturfaktor anhand der zwei Punkten zugeordneten Maskenwerte eines Maskenbildes, das nur im Maskenbereich von Null verschiedene Maskenwerte aufweist, und der ortsgleichen Bilddatenwerte des Röntgenbildes ermittelt wird,
• – dass für jeden Ort die zwei Punkte dadurch bestimmt sind, dass sie innerhalb des Maskenbereichs und innerhalb der im Vergleich zum Maskenbereich kleinen Umgebung des jeweiligen Ortes liegen und voneinander verschiedene Maskenwerte aufweisen.

According to the invention, it is provided

• That a small mask area is given in comparison to the X-ray image, within which the X-ray image is to be processed,
• That for each location within the mask area where variations in the Image data values of the X-ray image remain within a small area of the respective location, which is smaller than the mask area, below a predefined threshold, a provisional correction factor is determined,
• That a respective final correction factor is determined for each location located within the mask area on the basis of the determined provisional correction factors,
• For each location within the mask area, the image data value of the respective location is corrected on the basis of the locally identical mask value and the respective final correction factor,
• The respective provisional correction factor is determined on the basis of the mask points associated with two points of a mask image, which has mask values which differ only from zero in the mask area, and the spatially identical image data values of the x-ray image,
• For each location, the two points are determined by being within the mask area and within the small area of the respective location in comparison to the mask area, and having different mask values from each other.

In a preferred embodiment of the present invention, it is provided that the image data values in accordance with the relationship I '= I / 1 + g * M where I 'is the respective corrected image data, I is the respective image data, g is the respective final correction factor, and M is the respective mask value. This procedure has given good results in tests.

ermittelt wird, wobei g' der jeweilige vorläufige Korrekturfaktor, M₁ der einem der jeweiligen beiden Punkte zugeordnete Maskenwert, I₁ der bezüglich des einen Punktes ortsgleiche Bilddatenwert des Röntgenbildes, M₂ der dem anderen der jeweiligen beiden Punkte zugeordnete Maskenwert und I₂ der bezüglich des anderen Punktes ortsgleiche korrespondierende Bilddatenwert des Röntgenbildes sind. Auch diese Ausgestaltung hat in Versuchen zu guten Ergebnissen geführt.In a further preferred embodiment of the present invention, it is provided that for each pair of points the respective provisional correction factor according to the relationship

where g 'is the respective preliminary correction factor, M _{1 is} the mask value assigned to one of the two points, I _{1 is} the image data value of the X-ray image which is the same with respect to the one point, M _{2 is} the mask value assigned to the other of the respective two points, and I ₂ the other point are the same corresponding image data value of the X-ray image. This embodiment has led to good results in experiments.

It is possible that the mask image is predetermined. However, it is preferred that the mask image is determined on the basis of a calibration image.

To test the X-ray image for homogeneity, it is possible, for example, for a gradient image to be determined on the basis of the X-ray image and for the examination of whether the fluctuations in the image data values of the X-ray image within the environment of the respective location remain below the predefined threshold on the basis of the gradient image.

In order to optimize the procedure according to the invention, it is advantageous if the final correction factors are determined by smoothing, filtering and / or interpolating the preliminary correction factors.

As a rule, both the mask area as such and the mask values occurring within the mask area are constants. However, it may be expedient for the x-ray image to be assigned a detector temperature which the x-ray detector had at the time of capturing the x-ray image, and for the mask area to be a function of the detector temperature and / or for the mask values to be functions of the detector temperature.

The object is further achieved by a computer program whose execution by a computer causes the computer to carry out a processing method according to the invention. The object is also achieved by a data carrier on which such a computer program is stored in machine-readable form. Finally, the object is achieved by a computer having a mass memory, wherein in the mass memory in machine-readable form such a computer program is stored, which is executable by the computer.

Further advantages and details will become apparent from the following description of exemplary embodiments in conjunction with the drawings. In a schematic representation:

1 a block diagram of a treatment arrangement,

2 in perspective an X-ray detector,

3 a plan view of a portion of the X-ray detector and

4 to 6 Flowcharts.

According to 1 has an X-ray arrangement 1 including an X-ray source 2 and one X-ray detector 3 on. The x-ray detector 3 is an area detector. He extends according to 2 in two dimensions, hereinafter referred to as x and y direction. The x-ray detector 3 has a variety of pixels 4 on. As a rule, the pixels form 4 a rectangular grid. Every pixel 4 can therefore be defined by two indices i, j. For example, the index i can be used for the x-direction and the index j for the y-direction, where each index i, j (of course independent of the other index j, i) the values 0, 1, 2, ... up to a maximum value of z. B. 511, 1023, 2047 or another value can go through.

By means of the X-ray arrangement 1 becomes a two-dimensional X-ray image B of an examination subject 5 captured and a calculator 6 fed. If appropriate, a plurality of such images B can also be acquired, for example an angiographic sequence or projections from different angulations, and later a three-dimensional reconstruction of the examination subject 5 to be able to determine. In the context of the present invention, however, only the correction of detector-specific errors is important, which is independent of the other X-ray images B for all X-ray images B. In the context of the present invention, it is therefore sufficient to consider a single X-ray image B.

The x-ray detector 3 can according to 2 for example, two sub-detectors 7 . 8th consist. Each of the subdetectors 7 . 8th is flat. The part detectors 7 . 8th adjoin one another. They thereby form a joint edge 9 , According to the presentation of 2 runs the abutting edge 9 parallel to the x-direction. Alternatively, the abutting edge could 9 parallel to the y-direction. In principle, it is even possible that the abutting edge 9 neither parallel to the x-direction nor parallel to the y-direction, but has any orientation. It is also possible that the x-ray detector 3 more than two subdividors 7 . 8th having. For example, the X-ray detector could 3 four sub-detectors 7 . 8th have in a 2 × 2 arrangement adjacent to each other and thereby two orthogonally intersecting abutting edges 9 form. If several abutting edges 9 they can be considered independently. In the context of the present invention, it is therefore sufficient only in 2 illustrated abutting edge 9 consider.

In the vicinity of the abutting edge 9 (in general: in a small area of the X-ray image B) occur according to 3 in the acquired X-ray image B gain deviations. The correct consideration of these deviations is the subject of the present invention.

The captured X-ray image B, as already mentioned, the computer 6 fed. The computer 6 according to 1 a mass storage 10 on, for example, a hard disk. In the mass storage 10 is in machine-readable form - usually electronic or magnetic, in rare cases also optically - a computer program 11 saved. The computer program 11 is from the calculator 6 immediately executable.

The computer program 11 can the calculator 6 be supplied in various ways. For example, the computer program 11 the calculator 6 via a - not shown - computer-computer connection has been supplied. Examples of suitable computer-computer connections are a local area network (LAN) or the World Wide Web. Alternatively, it is possible that the computer program 11 on a (mobile) disk 12 is stored and the calculator 6 over the disk 12 is supplied. Is shown in 1 an embodiment of the data carrier 12 as a USB memory stick. The disk 12 However, could be configured as desired, for example as a CD-ROM, as an SD memory card, etc ..

The computer program 11 includes machine code 13 , The machine code 13 is from the calculator 6 directly abarbeitbar, the computer program 11 So from the computer 6 executable. The execution of the machine code 13 through the computer 6 causes the calculator 6 performs the step of a treatment process, which is subsequently described in connection with 4 and later also the 5 and 6 is explained in more detail.

According to 4 takes the calculator 6 in a step S1, the X-ray image B opposite. In a step S2, the computer selects 6 within a mask area 14 a place P. The mask area 14 is given by a mask image B '. The mask image B 'points within the mask area 14 Mask values M on, which at individual, singular places of the mask area 14 may have the value zero, but are usually different from zero. The mask values M decrease within the mask area 14 a value within a value continuum. For example, for all mask values M, the relation 0 ≤ M ≤ 1 be valid. Outside the mask area 14 are the mask values M of the corresponding pixels 4 always zero. The mask area 14 extends according to 3 on both sides of the abutting edge 9 , It corresponds to the small area in which gain deviations occur.

In a step S3, the computer determines 6 a logical variable OK. The logical variable OK assumes the value "TRUE" if and only if fluctuations of the image data values I of the X-ray image B within a small neighborhood U of the selected location P remain below a predefined threshold S. Otherwise, the logical variable OK assumes the value "FALSE". The environment U is how out 3 is recognizable, not only small with respect to the entire X-ray image B, but also small with respect to the mask area 14 ,

For the implementation of step S3, the calculator 6 , as in 4 shown, using the X-ray image B determine a gradient image. In this case, the calculator 6 the check as to whether the fluctuations of the image data values I of the X-ray image B remain within the environment of the respective location P below the predefined threshold S, on the basis of the gradient image.

In step S4, the computer checks 6 the value of the logical variable OK. If the logical variable OK is TRUE, the calculator executes 6 Steps S5 and S6 off. Otherwise, steps S5 and S6 are skipped.

• – innerhalb des Maskenbereichs 14 liegen,
• – innerhalb der Umgebung U des Ortes P liegen und
• – voneinander verschiedene Maskenwerte M₁, M₂ aufweisen.

In step S5, the computer determines 6 for the selected location P two points P ₁ , P ₂ . According to 3 the calculator determines 6 the two points P ₁ , P ₂ such that the two points P ₁ , P ₂

• - within the mask area 14 lie,
• - lie within the environment U of the place P and
• Have mutually different mask values M ₁ , M ₂ .

In the case that the place P is near the abutting edge 9 As a rule, a fourth condition is still satisfied, namely that the two points P ₁ , P ₂ lie on the same side of the mask area 14 lie. This condition is only deviated in rare exceptional cases.

In step 56 the calculator determines 6 for the selected location P a preliminary correction factor g '. The computer 6 determines the provisional correction factor g 'on the basis of the mask values M ₁ , M _{2 of} the two points P ₁ , P ₂ as well as the corresponding image data values I ₁ , I ₂ (ie assigned to the same locations). Preferably, the computer determines 6 as indicated in step S6, the provisional correction factor g 'according to the relationship

The above formula provides a more accurate result, the larger the difference of the mask values M ₁ , M ₂ . It is therefore preferred that the preliminary correction factor g 'for the selected location P is determined on the basis of the two points P ₁ , P ₂ at which the difference between their mask values M ₁ , M _{2 is} maximum. To determine the maximum difference, for example, the distance and / or - at a constant distance - the position of at least one of the two points P ₁ , P _{2 are} varied.

In step S7, the computer checks 6 whether he already steps S3 to S6 for all locations P of the mask area 14 has finished. If this is not the case, the computer goes to a step S8. In step S8, the computer selects 6 another, not yet checked place P. Sodann he goes back to step S3.

If the calculator 6 on the other hand, if the steps S3 to S6 have already been executed for all locations P, the computer proceeds to a step S9.

In step S9, the computer determines 6 for everyone within the mask area 14 each location P has a final correction factor g. The computer 6 determines the final correction factors g from the provisional correction factors g '. For example, the calculator 6 , as in 4 illustrated, the final correction factors g by smoothing, filtering and / or a - for example, linear - interpolation of the provisional correction factors g 'determine.

In a step S10, the computer corrects 6 within the mask area 14 the image data values I of the locations P of the X-ray image B. For each location P, the computer corrects 6 its image data value I on the basis of the respectively corresponding (ie location-identical) mask value M and the final correction factor g determined for the respective location P.

In the context of investigations on the basis of the present invention, it has proved plausible that the image data values I of the X-ray image B are within the mask area 14 the relationship I = I '(1 + g * M) suffice. I means the acquired image data I of a respective pixel 4 , I 'the unadulterated and actually sought image data value, by the object under investigation 5 is caused. g is the one of the respective operating parameters of the X-ray device 1 dependent sought final correction factor g. M is the respective mask value M of the mask image B '. It therefore makes sense to set the image data values I in step S10 according to the relationship I '= I / 1 + g * M to correct.

The formula reproduced above represents a simplified representation. For, strictly speaking, for each variable I, I ', g, M, the indexing (i and j) should additionally be indicated in each case.

In a step S11, the computer performs 6 a further processing of the X-ray image B, which is no longer the subject of the present invention.

The mask image B ', whose mask values M are required both for the determination of the provisional correction factors g' of the step S5 and S6 and for the final correction of the image data values I of the step S10, can be sent to the computer 6 be fixed. However, it is preferred that the step S1 of 4 corresponding 5 Steps S21 and S22 are upstream. In step S21, the calculator takes 6 a calibration image B '' opposite. The calibration image B "corresponds in terms of the approach to the X-ray image B. The difference from the X-ray image B is that the calibration image B" is detected, without that in the beam path between the X-ray source 2 and the X-ray detector 3 the examination object 5 is arranged. So it is a "blank picture" recorded. In step S22, the computer determines 6 on the basis of the calibration image B '' the mask image B '.

As a rule, the mask image B 'is constant, that is to say independent of the acquisition parameters of the X-ray image B. In individual cases, however, it is possible for the mask image B' to be a function of the temperature T of the X-ray detector 2 when detecting the X-ray image B must be varied. According to 6 Therefore, it is possible to modify the step S1 such that the computer 6 in addition to the X-ray image B, the temperature T of the X-ray detector 3 at the time of capturing the X-ray image B receives. This temperature T, referred to below as the detector temperature T, is consequently assigned to the X-ray image B. In this case, the mask image B 'is a function of the detector temperature T. The mask area 14 and / or the mask values within the mask area 14 can therefore be functions of the detector temperature T. More specifically, in this case, in a step S26, the mask area becomes 14 determined accordingly and / or the mask values M are determined accordingly.

The present invention has many advantages. In particular, it is easy to implement, works reliably and shows good results in practice. It can also be retrofitted to existing systems by updating the operating software accordingly.

The above description is only for explanation of the present invention. The scope of the present invention, however, is intended to be determined solely by the appended claims.

LIST OF REFERENCE NUMBERS

1

X arrangement

2

X-ray source

3

X-ray detector

4

pixel

5

object of investigation

6

computer

7, 8

part detectors

9

impact edge

10

mass storage

11

computer program

12

disk

13

machine code

14

mask area

B

X-ray photograph

B '

makeup

B '

calibration image

g, g '

correction factors

I, I ₁ , I ₂

Image data values

i, j

indices

M, M ₁ , M ₂

mask values

OK

logical variable

P

places

P ₁ , P ₂

Points

S

threshold

S1 to S26

steps

T

detector temperature

U

Surroundings

x, y

directions

Claims (10)

Preparation process for an X-ray detector ( 3 ) recorded X-ray image (B) of an examination subject ( 5 ), Wherein a small mask area (B) is smaller than the X-ray image (B) 14 ) within which the X-ray image (B) is to be processed, - for each within the mask area ( 14 ) in which fluctuations of the image data values (I) of the X-ray image (B) within one in comparison to the mask region (FIG. 14 ) remain below a predefined threshold (S), a provisional correction factor (g ') is determined, - based on the determined provisional correction factors (g') for each within the mask area ( 14 ) (P) a respective final correction factor (g) is determined, - wherein for each within the mask area ( 14 ) (P) the image data value (I) of the respective location (P) is corrected on the basis of the locally identical mask value (M) and the respective final correction factor (G), In which the respective provisional correction factor (g ') is based on the mask values (M ₁ , M ₂ ) of a mask image (B') assigned to two points (P ₁ , P ₂ ) which are only displayed in the mask region ( 14 ) and non-zero mask values (M) are obtained, and the same image data values (I ₁ , I ₂ ) of the X-ray image (B) are determined, - wherein for each location (P) the two points (P ₁ , P ₂ ) are determined thereby in that they are within the mask area ( 14 ) and within the compared to the mask area ( 14 ) are small environment (U) of the respective location (P) and have mutually different mask values (M ₁ , M ₂ ). Processing method according to claim 1, characterized in that the image data values (I) according to the relationship I '= I / 1 + g * M where I 'is the respective corrected image data, I is the respective image data, g is the respective final correction factor, and M is the respective mask value. Processing method according to claim 1 or 2, characterized in that for each location (P) the respective provisional correction factor (g ') according to the relationship

where g 'is the respective provisional correction factor, M _{1 is} the mask value assigned to one of the respective two points (P ₁ , P ₂ ), I _{1 is} the image data value of the X-ray image (B), M _{2 which is the} same with respect to the one point (P ₁ ) the mask value associated with the other of the respective two points (P ₁ , P ₂ ) and I _{2 are} the image data value of the X-ray image (B) which is the same with respect to the other point (P ₂ ). Processing method according to claim 1, 2 or 3, characterized in that the mask image (B ') is determined on the basis of a calibration image (B' '). Preparation method according to one of the above claims, characterized in that a gradient image is determined on the basis of the X-ray image (B) and the check as to whether the fluctuations of the image data values (I) of the X-ray image (B) within the environment (U) of the respective location (P) remain below the predefined threshold (S), based on the gradient image. Processing method according to one of the above claims, characterized in that the final correction factors (g) are determined by smoothing, filtering and / or interpolating the provisional correction factors (g '). Processing method according to one of the above claims, characterized in that the X-ray image (B) is assigned a detector temperature (T) which the X-ray detector ( 3 ) at the time of acquiring the X-ray image (B), and that the mask area (FIG. 14 ) is a function of the detector temperature (T) and / or that the mask values (M) are functions of the detector temperature (T). Computer program, the machine code ( 13 ) generated by a computer ( 6 ) is immediately executable and its execution by the computer ( 6 ) causes the computer ( 6 ) carries out a treatment process according to one of the above claims. Data carrier on which in machine-readable form a computer program ( 11 ) is stored according to claim 8. Computer that has a mass storage ( 10 ), wherein in the mass memory ( 10 ) in computer readable form a computer program ( 11 ) is stored according to claim 8, wherein the computer program ( 11 ) is executable by the computer.

Images