Dose determination method

Abstract

The invention relates to a method for determining the dose for an X-ray scan.

Description

[0001] Dose determination procedure

[0002] The invention relates to a method for determining the dose in an X-ray image.

[0003] Under German radiation protection law, the operation of X-ray equipment for use on humans is only possible if a number of regulations are observed. Among other things, it must be ensured that the X-ray equipment displays parameters for determining the exposure that the person being examined or treated received during the application. These parameters are recorded during the examination or treatment. Traditionally, the parameters used in X-ray examinations – the dose area product and, particularly in the dental field, the tube voltage (kV), the tube current (mA), and the exposure time (s) – are recorded in an X-ray control book along with other information about the examination or treatment. Based on the recorded parameters and information, the exposure of the person being treated or examined can be determined.

[0004] However, this procedure is problematic in several respects for determining the exposure of the person being examined or treated. When displaying or recording the values ​​set on the X-ray tube, errors can occur that are not immediately noticeable in the X-ray image. Furthermore, it is not always certain that the exposure to the person calculated using the settings on the X-ray tube occurs as expected. Last but not least, with existing X-ray equipment, it is not easy to digitally and ideally automatically archive the parameters set on the X-ray tube. Due to the necessary radiation protection equipment in the room in which such X-ray equipment is operated, transmitting the X-ray parameters via cable or wirelessly is difficult or only possible with considerable effort - especially on-site.

[0005] Apart from a C-arm setup, smaller, flexible X-ray tubes are often used in dentistry or materials testing. To determine the dose area product, the distance from the X-ray tube to the patient, as well as the aperture cone and the image plate size, must be taken into account using distance sensors or a tape measure to accurately determine the dose.

[0006] It is an object of the present invention to provide a method for dose determination which at least partially eliminates the problems mentioned and in particular can also be subsequently implemented without structural changes in existing X-ray devices, but in particular for flexible X-ray tubes.

[0007] This object is achieved by a method according to independent claim 1. The inventive method for dose determination comprises raw data from a digital X-ray image created using an image detector. Determining gray values ​​corresponding to minimum X-ray absorption and / or from the raw data of a digital X-ray image created using an image detector. Additionally determining gray values ​​of one or more regions with expected X-ray absorption values; calculating a dose based on the gray values ​​using a calibration.

[0008] The method according to the invention therefore provides for an analysis of the raw data of a digital X-ray image and, using a calibration, enables a calculation of the dose reaching the image receptor on the basis of this analysis.

[0009] An image detector within the meaning of the present application can be a storage plate that latently stores an X-ray image in a phosphor layer. The latent image is read out using photoluminescence by a scanner as a digital image generation device. Alternatively, the image detector can be an electronic detector, such as a solid-state detector, or another storage medium that generates an X-ray image directly or indirectly via an intermediary device, such as a monitor or a PC.

[0010] Raw data refers to the data from a digital X-ray image that is largely unprocessed immediately after the physical creation of the image. In particular, no scaling or other further processing of the intensity values ​​of the image material may have taken place.

[0011] In this context, grey values ​​are understood to be the part of the raw data which, for each image pixel, represents the X-ray absorption that has occurred at this image pixel as an intensity value or

[0012] The gray value corresponds to a measured value of the image sensor. In the case of image plates, this measured value is obtained when the image plate is read out by the output signal of the photomultiplier, which amplifies the weak photoluminescence light and outputs it as counting pulses. These counts can be represented as gray values.

[0013] It is a finding of the invention that the gray values ​​of the digital X-ray image can be interpreted as dose values, i.e., as a unit of measurement for the energy dose caused by ionizing radiation. A gray value of a single pixel thus describes the energy absorbed per unit mass. It can therefore serve as a measure for determining the dose deposited in the image receptor and thus ultimately also for the exposure of the person being treated or examined (patient dose).

[0014] The method according to the invention provides for an analysis of the raw data in two ways. According to a first alternative, the gray values ​​of the raw data are analyzed, and those gray values ​​are determined which correspond to minimal X-ray absorption in the structure under investigation and, conversely, to maximum absorption in the image receptor. Depending on the type of determination, these can be the brightest or darkest gray values. In the present case, it is assumed that with increasing X-ray absorption, the gray value decreases, i.e., the brightest gray values ​​have experienced the least X-ray absorption in the structure under investigation. For this analysis, representation in a histogram, for example, can be helpful.

[0015] Alternatively or additionally, the raw data is analyzed with regard to the image content so that one or more areas with expected X-ray absorption can be determined. These can, for example, be the areas with minimal X-ray absorption already mentioned. In the X-ray image, these are, for example, areas through which only thin soft tissue was irradiated or areas in which no significant X-ray absorption has obviously occurred. Additionally or alternatively, areas can also be analyzed which are subject to significant X-ray absorption but for which the X-ray absorption characteristics are known. For example, artificial structures used in dentistry to hold the image receptor and which are clearly visible in the X-ray image can be useful for this purpose due to their known nature and dimensions.

[0016] After determining the relevant gray values, either by analyzing all gray values ​​for their frequency to identify those corresponding to minimal X-ray absorption or / and by analyzing the image content to determine gray values ​​with expected X-ray absorption, a calibration can be applied to these identified gray values ​​to calculate the X-ray dose received by the patient. Calibration assigns a corresponding dose to each gray value occurring in the X-ray image. Among other things, calibration takes into account the image receptor's conversion characteristics from X-ray absorption to an intensity value or gray value.

[0017] In this context, it is preferred if the gray values ​​of one or more regions with expected X-ray absorption are determined based on the gray values ​​that correspond to minimal X-ray absorption. Thus, if the gray values ​​for which minimal X-ray absorption is recognizable in the raw data during the X-ray process are identified in a first step, these can be identified as regions in the digital X-ray image in a second step. This allows for subsequent corrections if necessary and thus offers greater accuracy in dose calculation.

[0018] It can advantageously be provided that the determination of the gray values ​​of one or more areas with expected X-ray absorption takes into account the homogeneity of the raw data within the respective area. The aim here is for the homogeneity of the gray values ​​to be as high as possible in the areas to be used for determining the dose. Conversely, a high variation in the gray values, i.e. low homogeneity, indicates an area that is less suitable for calculating the dose. If the gray values ​​of the raw data are sufficiently homogeneous, however, all gray values ​​of the determined areas can be taken into account when calculating the dose. Advantageously, an average is calculated across all gray values ​​of the areas. Alternatively or additionally, the median or other statistical parameters such as variance, standard deviation, skewness, excess, etc. can be used for evaluation.

[0019] Calibration advantageously includes calibrating the X-ray image generation chain. This can also be done using a factory-provided calibration.

[0020] In a preferred embodiment, the X-ray image generation chain comprises an X-ray source, the image detector, and a digital image generation device for generating the raw data of the digital X-ray image. As already mentioned, the image detector can be a storage plate or a solid-state sensor. In the case of a storage plate, the digital image generation device is, for example, the scanner that reads the latent X-ray image from the storage plate. In the case of electronic detectors such as solid-state detectors, this can be, for example, the evaluation electronics.

[0021] A calibration phantom is preferably used for calibration. The calibration phantom can preferably be designed such that it has absorption materials with appropriate thicknesses for the maximum X-ray absorption or the expected X-ray absorption.

[0022] In a preferred embodiment, the calibration comprises a calibration factor and / or a calibration curve. If there is a linear relationship between the X-ray absorption and the gray value, the calibration curve is a straight line. If the image detector outputs a gray value of 0 in the absence of X-ray absorption, the calibration curve can be represented as a calibration factor. In a preferred embodiment, the calibration factor and / or the calibration curve take into account the resolution of the X-ray image detector, the sensitivity of the X-ray image detector and / or the nature of the calibration phantom. After calibration has been completed, a change in the X-ray image generation chain can be mapped into the calibration curve or the calibration factor without the need for recalibration.For example, a change in the resolution or sensitivity of the digital imaging device can be taken into account in the calibration factor.

[0023] It is particularly preferred if the calculated dose corresponds to the exposure of a person being examined or treated. This allows the requirements of the Radiation Protection Ordinance to be met without the need for complex data transmission options from an X-ray machine.

[0024] The problem is also solved by a method for calibrating an X-ray image detector. The method comprises the following steps:

[0025] Irradiating an X-ray image receiver with a known dose; reading the raw data from the X-ray image receiver and creating a digital X-ray image; assigning the gray values ​​of the X-ray image to the dose. Thus, the calibration allows the corresponding dose to be extracted from the X-ray image.

[0026] In one embodiment, irradiating the X-ray image receiver advantageously involves passing radiation through a calibration phantom. This allows for a particularly precise determination of the calibration.

[0027] Preferably, the assignment of the gray values ​​comprises determining a homogeneous region of the X-ray image.

[0028] The invention can advantageously be implemented as a computer program or computer program product, in particular on a scanner or an image output unit. Exemplary embodiments of the invention are explained in more detail below with reference to the drawings. These show:

[0029] Figure 1A shows a schematically illustrated measuring setup for an embodiment of a calibration method;

[0030] Figures 1 B-1 E are diagrams explaining the calibration procedure; and

[0031] Figures 2A-2D are diagrams illustrating an embodiment of a dose determination method.

[0032] Figure 1 illustrates a highly schematic representation of a measurement setup for calibrating an X-ray image generation chain. The X-ray image generation chain comprises an X-ray source 10, an X-ray image receiver 12, and a digital image generation device 14. In the embodiment shown in Figure 1A, a storage plate 13 serves as the image receiver, and a so-called scanner 15 serves as the digital image generation device. Alternatively, the image receiver 12 could also be an electronic detector / sensor, for example, a solid-state detector with corresponding evaluation electronics as the digital image generation device 14.

[0033] For the calibration process, the X-ray beam 16 emitted by the X-ray source 10 is limited by a diaphragm 18 so that it completely covers the X-ray image receiver 12. A calibration exposure is then performed. Before, during, or after the calibration exposure, the actual dose delivered by the X-ray source 10 can be measured using a dosemeter 20, which is irradiated instead of the X-ray image receiver 12 or simultaneously with it. Simultaneously or alternatively, the dose delivered by the X-ray source 10 can be determined using the X-ray parameters set on the X-ray source 10, namely the X-ray voltage, the X-ray current, and the exposure time.

[0034] Furthermore, when determining the calibration, it can be advantageous if an absorber 22, for example in the form of a calibration phantom 23, is arranged in the beam path between the X-ray source 10 and the X-ray image receiver 12. Depending on its properties, the absorber 22 can have locally different absorption coefficients and thus cause an image to be recorded with different, but defined, doses in the X-ray image receiver 12. In addition, the absorption by the absorber 22 can be adapted to the actual recording conditions, thus causing the calibration to be close to the later operating point. This enables precise calibration and thus also a subsequent precise calculation of the dose at the image receiver 12.

[0035] Advantageously, a slot for a dose measuring device can be provided in the calibration phantom 23, which can carry out a dose measurement during the irradiation.

[0036] Particularly preferred is that calibration can be carried out at the factory or subsequently for products that have already been delivered.

[0037] The procedure for performing a calibration is described below with reference to Figure 1 E.

[0038] Based on extensive measurements, it was determined that the two technologies, imaging plate and electronic detector, exhibit a linear relationship between the dose delivered to a pixel and the corresponding gray value displayed in the raw X-ray image data. It would also be possible to create a characteristic curve for a nonlinear relationship. This would require the creation of several calibration points.

[0039] Assuming that no X-ray absorption in the structure under investigation results in a gray value of zero, the calibration or the calibration curve can be represented as a calibration factor.

[0040] The calibration factor is as follows:

[0041] Ap = PosiSKal

[0042] Gray value Kai This calibration factor can be adapted to different circumstances of the X-ray process or the generation of the X-ray image.

[0043] The embodiment shown in Figure 1A uses imaging plate technology to create the X-ray image, in which the latently stored image in the imaging plate 13 is read out by the scanner 15. The reading of the imaging plate 13 can be performed on the scanner 15 using various parameters that can be included in the calibration factor. If one or more of the parameters are changed, the following formulas can be used individually or in combination to adapt the calibration factor to the changed parameters: r " " P Cor — 1 + 0.125

[0044] For the size LP KorThis is a correction factor that takes into account the readout of imaging plates with different resolutions (line pairs). LP Kor represents a correction factor which, at usual resolutions of 5-50, particularly preferably 10-40, especially preferably 20-30 line pairs per millimetre, takes into account the fact that reading the storage plate with higher resolution results in a lower light yield for the same dose.

[0045] The HV factor KorThis takes into account the fact that very small amounts of light must be used to read the latent image in the imaging plate, which is why photomultipliers are used. These are operated with a voltage HV. A change in this voltage has a significant impact on the resulting gray value of the X-ray image in the raw data. The exponent X represents the typical design of the photomultiplier used in the scanner 15 of the exemplary embodiment. The exponent can vary depending on the design of the scanner 15, which is indicated by X. A typical value for X can be, for example, between 4 and 10, in particular between 5 and 7, for example, 1 / 0.1677. If a phantom is used for calibration, the influence of the calibration phantom can be represented using a phantom factor PF. The overall calibration factor for a single image is as follows:

[0046] KF BUd = AP * LPKor * AGM Kor * PF

[0047] This calibration factor KF B iid can be stored in the header data of an X-ray image and enables dose determination accordingly.

[0048] Figures 1B-1D illustrate the calibration process. Figure 1B shows an X-ray image 30 of an exposed imaging plate 13 obtained with a scanner 15, without the use of a calibration phantom 23. The numbers along the X and Y axes shown in Figures 1B and 1C represent a coordinate system for identifying individual pixels.

[0049] The X-ray image 30 shown in Figure 1B shows the essentially unprocessed raw data of the X-ray image as grayscale values. A homogeneous region 32 is defined in the X-ray image 30, within which the displayed grayscale values ​​are subject to slight fluctuations. The selected homogeneous region 32 is shown enlarged in Figure 1C.

[0050] Figure 1D illustrates the gray value distribution of the homogeneous area 32 as a histogram. The x-axis, or abscissa, represents the gray value of a pixel. The ordinate, or y-axis, shows the corresponding number of pixels, normalized here to a distribution with the value 1.

[0051] As can be seen in Figure 1D, the gray values ​​exhibit a Gaussian distribution. Accordingly, the mean of the Gaussian distribution is determined. The dose measured during the acquisition of X-ray image 30 can then be assigned to this mean.

[0052] Figure 1E illustrates an alternative embodiment in which a calibration phantom 23 was used to acquire the X-ray image 34, as shown in Figure 1. The calibration phantom 23 can simultaneously serve as an acceptance phantom during the initial acceptance of an X-ray system and shows various structures that exhibit different absorption coefficients during fluoroscopy. Accordingly, Figure 1E shows three regions 36, 38, and 40 where different material thicknesses of the acceptance or calibration phantom 23 are present and which can therefore be used for calibration.

[0053] Figures 2A-2D illustrate a method for determining the dose of a digital x-ray image 50. Figure 2A illustrates the x-ray image 50 directly after its raw data creation. The black borders created when reading a storage plate have already been masked off. Before further processing and analysis of the image 50, a Gaussian filter can be applied. This can reduce image noise and is equivalent to smoothing with a low-pass filter. This makes the algorithms for determining the dose-relevant pixels more robust.

[0054] Figure 2B shows a histogram of the gray values ​​of the X-ray image 50 shown in Figure 2A. For dose determination, it is important to identify the gray values ​​with the lowest X-ray absorption in the image receptor. These can be assumed to have reached the image receptor largely unattenuated and therefore provide a good measure for dose calculation.

[0055] For example, a specific percentage of the histogram can be selected, such as the 10% of gray values ​​with the highest gray value. Alternatively, peak detection and analysis, or other selection of pixels with the desired gray value, can be performed. This is shown in Figure 2B as field 52. After determining the areas for the dose-relevant gray values, the determined area mask is applied back to the raw data image in order to use the pixel values ​​of the raw data for dose determination.

[0056] In principle, the dose can be calculated simply from selecting these pixels. To improve the calculation result, a further analysis of the dose-relevant pixels can be performed, specifically in the graphical representation. This is shown in Figure 2C. The brightest 10% of the pixels selected in the step described with reference to Figure 2B are shown in Figure 2C as contiguous regions 54, 56.

[0057] Additional logical safeguards can be implemented here. For example, it can be verified that a minimum number of pixels is used for evaluation. Alternatively or additionally, the boundaries of the relevant regions can be eroded, i.e., shifted to the interior of the region to prevent peripheral areas, for example, in a tooth structure or dense tissue, from being included in the calculation.

[0058] After this step, the number of dose-relevant pixels and their corresponding gray values ​​are determined. This is shown in Figure 2D. Now, the edge clipping (masking) and Gaussian filtering can be removed to use the actual raw data for the dose calculation. The mean value is then calculated according to the following formula and multiplied by the calibration factor to calculate the dose of the X-ray image:

[0059] The method for determining the dose is based on the assumption that real patient images contain image portions where the incident radiation is attenuated only by soft tissue—for example, the cheeks in the dental area. Experiments have shown that image regions in real images that show only thin cheek tissue without jawbone, teeth, implants, bite plates, or other highly absorbent or thick materials are particularly suitable for dose evaluation, as they appear as homogeneous areas in the image receptor. With appropriate calibration, the soft tissue factor can be recalculated accordingly, and dose values ​​approximately equivalent to unshielded radiation can be calculated.

[0060] While the calibration method shown in Figures 1A-1D selected a central area for calibration, a further development of the invention could allow the image detector to be calibrated specifically across the entire detector surface to compensate for local sensitivity differences. The calibration factors stored for the respective dose-relevant areas could then be used for dose determination. During dose determination, care should be taken to ensure that no saturation occurs at the image detector during radiation absorption.

Claims

PATENT CLAIMS 1. Methods for dose determination, comprising: Determine, from the raw data of a digital X-ray image created using an image receptor, grayscale values ​​corresponding to maximum X-ray absorption and / or Determine, from the raw data of a digital X-ray image created using an image receptor, the grayscale values ​​of one or more areas with expected X-ray absorption; Calculating a dose based on grayscale values ​​using a calibration.

2. Method according to claim 1, wherein the determination of the grey values ​​of one or more areas with expected X-ray absorption is based on the grey values ​​corresponding to a minimum X-ray absorption.

3. Method according to claim 1-2, wherein the determination of the grey values ​​of one or more areas with expected X-ray absorption takes into account the homogeneity of the raw data within the respective area.

4. Method according to claim 3, wherein all grey values ​​of the specified areas are taken into account when calculating the dose.

5. Method according to any one of claims 1-3, wherein the calibration comprises calibrating the X-ray image generation chain.

6. Method according to claim 5, wherein the X-ray image generation chain comprises an X-ray source, the X-ray image receiver and a digital image generation device for generating the raw data of the digital X-ray image.

7. Method according to one of claims 1-6, wherein a calibration phantom is used for calibration.

8. Method according to any one of claims 1-7, wherein the calibration comprises a calibration factor and / or a calibration characteristic curve.

9. Method according to claim 8, wherein the calibration factor and / or the calibration characteristic curve takes into account a resolution of the X-ray image receiver, a sensitivity of the X-ray image receiver and / or the nature of the calibration phantom.

10. Method according to any one of claims 1-9, wherein the calculated dose corresponds to the exposure of a person examined or treated.

11. Procedure for calibrating an X-ray image receiver, comprising the following steps: Irradiation of an X-ray image receptor with a known dose; Reading the raw data from the X-ray image receiver and creating a digital X-ray image; Assigning the gray values ​​of the X-ray image to the dose.

12. Method according to claim 11, wherein the irradiation of the X-ray image receiver comprises irradiating a calibration phantom.

13. Method according to claim 11 or 12, wherein the assignment of the gray values ​​comprises determining a homogeneous area of ​​the X-ray image.

14. Computer program product comprising instructions which, when the program is executed by a computer, cause it to execute the method according to any one of claims 1-10.

15. Computer program product, comprising instructions which, when executed by a computer, cause the program to perform the following steps: Reading the raw data from the X-ray image receiver and creating a digital X-ray image; Assigning the gray values ​​of the X-ray image to a dose.

16. Computer program product according to claim 15, wherein the assignment of the gray values ​​particularly comprises determining a homogeneous area of ​​the X-ray image.

17. Scanner comprising means for carrying out the method according to any one of claims 1-10.

18. Scanner comprising means for carrying out the computer program product according to any one of claims 15-16.

Images