EP0456322B1 - Arrangement for producing X-ray photographs - Google Patents

Description

The invention relates to an arrangement for generating x-ray images with an x-ray emitter for generating an x-ray beam, a photoconductor for converting x-radiation into a charge pattern, which is applied to a carrier that is rotationally symmetrical with respect to an axis of rotation, a drive unit for driving the carrier about the axis of rotation, and a Readout unit, which converts the charge pattern on the surface of the rotating photoconductor into electrical image values after an X-ray exposure.

Such an arrangement is known from DE-A-35 34 768. The exposure of the photoconductor takes place during the exposure through a slit diaphragm, the main direction of expansion of which runs parallel to the axis of rotation and which suppresses a narrow, fan-shaped x-ray beam which passes through the examination area and exposes a narrow strip on the surface of the photoconductor. During the recording, the carrier rotates about the axis of rotation and is displaced synchronously with it perpendicular to the axis of rotation, so that the examination area is sequentially imaged on the surface of the photoconductor.

The charge pattern generated in this way is read out immediately after the X-ray exposure. The carrier rotates at a much higher speed than during the X-ray exposure, and the readout device reads the charge with one or more probes on a practically circular track on the surface of the photoconductor out. In order to be able to read out the entire surface, the reading unit is moved parallel to the axis of rotation at a comparatively low average speed.

With such an arrangement, the reading can take place much faster, more precisely and more precisely than is possible with a flat photoconductor plate, as is known from US Pat. No. 4,134,137. However, the fast readout is absolutely necessary because the photoconductor is not only discharged by the X-ray exposure, but also by dark currents. This is offset by the disadvantage that the total exposure time is comparatively long and that the power of the X-ray tube is poorly used because only a thin fan of rays is used to expose the photoconductor.

It is an object of the present invention to provide an arrangement which enables the photoconductor to be read out quickly with short exposure times.

Starting from an arrangement of the type mentioned at the outset, this object is achieved according to the invention in that the drive unit is controlled in such a way that the photoconductor does not rotate during an X-ray exposure.

While in the known arrangement the carrier rotates both during the X-ray exposure and during the readout process, this is only the case with the invention during the readout. During the X-ray exposure, the support with the photoconductor does not rotate, and therefore the area of the photoconductor intended for the X-ray exposure can be exposed in all areas at the same time, so that there is a short total exposure time and good utilization of the performance of the X-ray tube. However, with the same maximum format, the X-ray (measured in the circumferential direction of the carrier or photoconductor), the diameter of the carrier may be substantially larger than in an arrangement of the type mentioned. While in the latter only the circumference of the carrier must be larger than the recording format, the diameter of the Carrier can be larger than the recording format.

At this point it should be mentioned that EP-A-94 843, in particular FIG. 8, already discloses an arrangement for generating X-ray images in which a storage phosphor is applied to a cylindrical support which is stationary during an X-ray image. Such storage phosphors lose their image information much more slowly than photoconductors, so that fast reading is not necessary. After each X-ray exposure, the carrier is rotated once around the axis of rotation in three steps, the X-ray exposure being read out with a two-dimensionally guided laser beam in the first step and the storage phosphor being erased in the next step. An additional recording can be made at each step. Here, the carrier stands both during the X-ray exposure and during the readout process; reading out cannot be faster than with a flat record carrier.

A preferred further development provides that means for geometrical image transformation are provided which compensate for the image distortions caused by the curvature of the surface of the photoconductor. This avoids distortions that are unavoidable due to the curvature of the rotationally symmetrical, preferably cylindrical, carrier, unless the recording format is small in comparison to the drum diameter. With this image transformation, the pixels on the drum surface Assigned to virtual pixels in an image plane located in the X-ray beam in such a way that the connecting straight lines intersect through assigned pixels in the starting point of the X-ray beam. The image plane can be tangential to the surface of the photoconductor and perpendicular to the plane defined by the focus of the x-ray emitter for generating the x-ray beam and the axis of rotation. However, a different position within the X-ray beam is also possible.

In a further embodiment of the invention it is provided that the drive unit of the carrier is controlled in such a way that, prior to reading out an X-ray image, the carrier is rotated to such an extent that an area of the photoconductor which was not exposed during the previous X-ray image enters the beam path. This makes it possible to take two pictures - in smaller formats even more than two - in close succession without having to read out the photoconductor in between. However, this embodiment assumes that the image memory has enough capacity to store two or more X-ray images at the same time.

• Fig. 1 ein Röntgengerät nach der Erfindung in schematischer Darstellung,
• Fig. 2a und 2b die geometrischen Verhältnisse bei dieser Ausführungsform und
• Fig. 3 eine Einheit zur Verarbeitung der von der Ausleseeinheit gelieferten Bildwerte.

The invention is explained below with reference to the drawing. Show it

• 1 shows an X-ray device according to the invention in a schematic representation,
• 2a and 2b, the geometric relationships in this embodiment and
• 3 shows a unit for processing the image values supplied by the readout unit.

The X-ray beam 10 emanating from the focus of an X-ray emitter 1 passes through a patient 2 lying on a table top 3 and an anti-scatter grid 8 before it hits a cylindrical carrier (drum) 4, the cylinder axis 7 of which is perpendicular to the plane of FIG. 1 is simultaneously its axis of rotation . The carrier 4 can be driven about the axis of rotation 7 by means of a motor drive 9. The carrier 4 is coated on its lateral surfaces with a photoconductor, preferably a 0.5 mm thick selenium layer 41.

Outside the beam path there is a charging device 6, which charges the rotating carrier before an X-ray exposure, so that a voltage of, for example, between the surface of the electrically conductive carrier and the outer surface of the selenium layer 1,500 volts is present. Also outside the beam path is a readout device 5, which reads out the charge density on one or more tracks after an X-ray exposure with one or more probes. In order to be able to read the entire surface, the reading device 5 is displaced relative to the carrier 4 by means of a further drive unit 11 parallel to the axis of rotation 7 at an average speed which is low in comparison to the peripheral speed of the carrier. The structure and function of the readout device 5 and the charging unit 6 are described in more detail in DE-A-35 34 768, to which express reference is made.

Dabei ist L der Abstand des Fokus 1 von der Bildebene.In an X-ray exposure, the drive 9 for the carrier 4 is switched off, so that the photoconductor does not rotate. Therefore, the outer radius r of the photoconductor layer 41 on the carrier 4 must be large enough so that the part of the patient 2 to be photographed can be completely imaged on the surface of the photoconductor 41. For the Maximum recording format B (measured in an image plane 12 tangent to the surface of the photoconductor 41 in the apex line - this is the line in which the plane defined by the focus 1 and the axis of rotation 7 intersects the surface of the photoconductor 41) $${\text{B = 2 r * (1 + 2r / L)}}^{\text{-1/2}}$$

L is the distance of focus 1 from the image plane.

In practice, the recording format should be smaller than this limit and preferably 0.95 B - or less. Thus, for a recording format of e.g. 40 cm with a value L = 180 cm a radius r of at least 23.7 cm is required.

The geometric distortions caused by the curvature of the carrier 4 would only be negligible if the diameter (2r) of the carrier 4 were large in comparison to the dimensions of the recording format. Since this is not possible due to lack of space, it is necessary to compensate for the distortions. For this purpose, a point in the image plane 12 is assigned to each point of the part of the photoconductor layer 41 irradiated during the recording such that the connecting straight line of the points assigned to one another passes through the focus of the X-ray emitter 1. The transformation required for this is explained in more detail with reference to FIGS. 2a and 2b.

FIG. 2a shows the carrier 4 in a perspective representation, while FIG. 2b shows it in the same representation as FIG. 1, ie with the axis of rotation 7 perpendicular to the plane of the drawing. The coordinates on the surface of the photoconductor are denoted by x, y, the y axis being identical to the apex line already mentioned, in which the image plane 12 touches the photoconductor. The x coordinate of a point is the length of the arc, that connects this point on the surface of the photoconductor to the y-axis. The coordinates of the assigned image point in the image plane are denoted by x _v and y _v . The origin of the x _v , y _v coordinate system is identical to the origin of the xy coordinate system and the y _v axis coincides with the y axis. The auxiliary variable z denotes the distance of a pixel from the image plane 12.

For z: $$\text{z = r * (1-cos (x / r))}$$

For x _v : $${\text{x}}_{\text{v}}{\text{= (r²- (rz) ²)}}^{\text{1/2}}\text{) / (1 + z / L)}$$

y _{v is} calculated according to the relationship: $${\text{y}}_{\text{v}}\text{= y * 1 / (1 + z / L)}$$

In this way, a pixel x _v , y _v in the image plane 12 can be assigned to each pixel x, y on the surface of the photoconductor hit by X-radiation.

It is not absolutely necessary for the image plane to touch the apex line of the wearer; an image can also be calculated in a plane parallel to 12. In this case, equations 3 and 4 must weight x _v and y _v with a constant factor.

Likewise, the image plane can also run at an angle different from 90 ° to the plane formed by the axis of rotation 7 and the focus 1. The transformation equations then become more complicated. Such an oblique image plane can arise, for example, in the case of oblique images in which the patient 2 or the tabletop 3 is irradiated obliquely and in which an X-ray image is nevertheless made in a plane 3 relative to the tabletop parallel plane should take place. However, in the case of such an inclined image, it can also be expedient to have the image plane perpendicular to the plane (inclined in the case of an inclined image), which is defined by the axis of rotation 7 and the focus 1. In this case, the image plane would run obliquely to the table top 3 and one could avoid the distortions that occur with conventional oblique photographs.

If the pixels in both the image generated on the photo surface and in the image derived therefrom each have the same dimensions, e.g. 0.2 mm x 0.2 mm, then it follows from the geometric relationships that the image value of a pixel at the edge of the image plane 12 is wholly or partly composed of the image values of several pixels on the surface of the photoconductor. The weighted sum of the image values mentioned must therefore be formed, the weighting factors being between 0 and 1.

dI_v ist dabei der Beitrag des Bildwertes I(x,y) zu dem Bildwert I_v für den Bildpunkt x_v, y_v in der Bildebene. k ist der Korrekturfaktor, der mit zunehmendem Betrag von x abnimmt und der die erwähnte Gewichtung berücksichtigt. Die Änderung des Faktors k als Funktion von x ist umso ausgeprägter, je härter die Röntgenstrahlung ist, d.h. je größer die Spannung an der Röntgenröhre während der Aufnahme ist. Bei sehr weicher Strahlung verschwindet diese Abhängigkeit praktisch.As a rule, the X-rays are not completely absorbed within the photoconductor layer (eg 0.5 mm selenium). The result of this is that an X-ray beam striking obliquely (at the edge) changes the charge density more than an X-ray beam striking perpendicularly (in the middle). A homogeneous object would therefore lead to an X-ray exposure exposed to different locations. This can be compensated for by multiplying the image values I (x, y) assigned to the individual pixels on the surface of the photoconductor - preferably in connection with the equalization transformation - by a correction factor k so that the relationship applies $${\text{dI}}_{\text{v}}{\text{(x}}_{\text{v}}{\text{, y}}_{\text{v}}\text{) = k * I (x, y)}$$

dI _v is the contribution of the image value I (x, y) to the image value I _v for the pixel x _v , y _v in the image plane. k is the correction factor, which decreases with increasing amount of x and which takes into account the weighting mentioned. The change in the factor k as a function of x is more pronounced the harder the x-ray radiation, ie the greater the voltage on the x-ray tube during the exposure. With very soft radiation, this dependency practically disappears.

3 schematically shows the processing of the values supplied by the read-out unit 5. They are first fed to an analog-digital converter 20 and stored in a memory 22 by an image processing unit 21. The image processing unit 21 calculates the image values I _v (x _v , y _v ) of the image transformed into the image plane from the image values contained in the memory 22 according to equations 2 to 5 and stores them in a further image memory 23 corrected image can be displayed on a monitor 24.

It is not necessary for an image memory to be available for the image values I on the surface of the photoconductor and the image values I _v in the image plane, as indicated in FIG. 3. If the image values I have been used for the calculation, they are no longer required, so that the calculated image values I _{v can be transferred} to the memory 22. All that is required is a buffer memory for a small part of the image.

In the case of X-ray images which are produced using a photoconductor, further processing steps are required, for example low-pass filtering or as known from DE-A-38 42 525, a correction of the self-discharge of the photoconductor which takes place after an X-ray exposure. These processing steps are carried out in the image processing unit 21 before the transformation described above.

In the case of various examinations, it is necessary to have two X-ray recordings in succession at a time interval which is smaller than the time period required for reading an image by means of the reading device 5. Such an operation is possible with the arrangement shown in FIG. 1 because each x-ray exposure does not even expose the drum to half its circumference. For this purpose, only the drive 9 has to be controlled such that, after a picture has been taken, the carrier 4 is rotated through 180 °, so that a part of the photoconductor which has not yet been exposed reaches the beam path in the subsequent X-ray picture. With a smaller format of the X-ray image, a smaller rotation of e.g. 120 or 90 ° possible, so that three or four X-rays could be taken in succession without having to read out in between. However, it goes without saying that in these cases the storage capacity of the image memory 22 must be sufficient to store two or three or four images.

Claims (5)

1. A device for forming X-ray images comprising an X-ray source (1) for producing an X-ray beam, a photoconductor (41) for converting X-rays into a charge pattern provided on a carrier (4) which is constructed so as to be rotationally symmetrical with respect to an axis of rotation (7), a drive unit (9) for driving the carrier about the axis of rotation, and a reading unit (5) which, after an X-ray exposure, converts the charge pattern on the surface of the rotating photoconductor into electric image values, characterized in that the drive unit is controlled so that the photoconductor (41) does not rotate during an X-ray exposure.
2. A device as claimed in Claim 1, characterized in that there are provided means (21) for geometric image transformation which compensate for the image distortions caused by the curvature (r) of the surface of the photoconductor (41).
3. A device as claimed in any one of the preceding Claims, characterized in that the drive unit (9) of the carrier (4) is controlled so that before the reading of an X-ray image the carrier is rotated so far that an area of the photoconductor which has not been exposed during the preceding X-ray exposure enters the beam path.
4. A device as claimed in any one of the preceding Claims, characterized in that there are provided means (21) for multiplying the image values by a correction factor (k) which is larger for image values at the edge of the image than for those in the centre of the image.
5. A device as claimed in any one of the preceding Claims, characterized in that there is provided a further drive unit (11) for moving the reading unit (5), in a plane containing the axis of rotation (7), along the surface of the photoconductor (41).

Images