METHOD FOR DETERMINING THE QUALITY OF A STORAGE FILM AND STORAGE FILM SCANNERS FOR THIS PURPOSE

Description

BACKGROUND OF THE INVENTION 1. Field of the invention

[0001] The invention relates to a method for determining the quality of a storage phosphor plate and to a storage phosphor plate for this purpose. 2. Description of the state of the art

[0002] In X-ray technology, particularly in dental X-ray technology, storage phosphor plates are used today to capture X-ray images. These plates consist of a phosphor material embedded in a transparent matrix. This creates so-called storage centers that can be excited into metastable states by incident X-rays. When such a storage phosphor plate is exposed in an X-ray machine, for example to image a patient's bitewing, the plate contains a latent X-ray image in the form of excited and unexcited storage centers.

[0003] To read out the storage phosphor plate, it is scanned point by point with readout light in a scanning device. This causes the metastable states of the excited storage centers to be brought into a state that rapidly relaxes, emitting fluorescent light. This fluorescent light can be detected by a detector unit, allowing the X-ray image to be visualized with appropriate evaluation electronics. For the readout process, drum scanners are used, for example, which guide the storage phosphor plate along a cylindrical surface over a readout slit.

[0004] A major advantage of storage phosphor technology lies in the reusability of the phosphor plates. After being read out, during which the image information stored on the plate is erased anyway, a storage phosphor plate can generally be used for many further recording and storage operations. However, various aging processes limit its reusability. One such limiting factor is the formation of local storage centers over time, which can no longer be stimulated and therefore remain dark in the stored image. In practical use, mechanical stress can also occur, which, for example, can lead to scratches or pinpoint damage on the surface of the storage phosphor plate due to improper handling.

[0005] In practical operation, it is desirable to know the current storage quality of a storage film. Conventionally, one can visually inspect the film and assess the impact of scratches on the recording and readout process. Furthermore, one can count the number of images taken with the film or determine its lifespan based on its production date. However, none of these measures provide a reliable indication of the actual storage capacity of the film.

[0006] US 5,420,441 A describes an automated procedure for analyzing the photometric calibration and image quality characteristics of a high-resolution storage phosphor reader. A specially designed test target with a lead mask is mounted on a storage phosphor and exposed to a standard medical X-ray source. A storage phosphor reader to be calibrated reads the exposed storage phosphor and generates a digital X-ray image. The digital X-ray image is analyzed by a computer algorithm that presents the analysis results in text and graphical form to isolate reader problems and quantify the reader's calibration status.

[0007] US Patent 2010 / 266187 A1 describes a system, a method, and computer-readable media for generating a corrected image from image information extracted from an X-ray plate. First, defect map image information is extracted from the X-ray plate to identify any defects (e.g., scratches) on the plate. The plate is then exposed to acquire physical image information (e.g., anatomical information), and this physical image information is extracted from the plate. The defect map information is used to identify corresponding defects in the extracted physical image information. Image processing is then performed to correct the defects (e.g., due to scratches on the X-ray plate) in the physical image information and generate a corrected physical image. SUMMARY OF THE INVENTION

[0008] It is an object of the invention to provide a method for determining the quality of a storage phosphor plate that avoids the aforementioned disadvantages and, in particular, makes it possible to make an accurate statement about the storage capacity and quality of the storage phosphor plate.

[0009] The problem is solved by a method according to the independent claim. The method according to the invention comprises the steps of: exposing the storage phosphor plate, scanning the storage phosphor plate to determine an image, determining the signal-to-noise ratio of the image, performing edge detection on the image, and calculating a quality value of the storage phosphor plate based on the signal-to-noise ratio of the image and on the basis of the detected edge structure.

[0010] The inventive method incorporates two quality features that are used in combination. One feature is the aforementioned signal-to-noise ratio. After exposure of the storage phosphor plate, the signal-to-noise ratio of the image provides an indication of whether the phosphor plate has defects in the form of non-functional storage centers. If storage centers do not emit fluorescent light despite uniform exposure of the storage phosphor plate during the readout process, the signal-to-noise ratio deteriorates.

[0011] The other characteristic is edge detection. Edge detection on the image provides similar indications of possible defects. With uniform exposure of the storage phosphor plate, the resulting image should show uniform illumination – that is, a uniform gray value. If, however, structures are detected using edge detection, this indicates possible defects in the storage phosphor plate.

[0012] Preferably, the storage phosphor plate is exposed to a specific dose. This simplifies determining the signal-to-noise ratio or performing edge detection. However, the exposure and subsequent calculations can also be performed with an unknown dose. The dose used should, however, be within a dose range that prevents the storage phosphor plate from being underexposed or overexposed. The dose can be adjusted, for example, by changing the exposure time, the X-ray voltage, and the anode voltage.

[0013] In an advantageous embodiment of the invention, the exposure step includes setting a specific distance between the film and a recording device. This facilitates the determination of the expected absolute gray value of the image and thus the reproducibility of the calculation of the storage film's quality value, but is not a necessary condition for the successful execution of the method. It is only necessary to ensure that the storage film is neither over- nor underexposed. In principle, the dose-distance pair should be selected accordingly. For example, a greater distance can be compensated for by a greater dose.

[0014] In one embodiment of the invention, the exposure of the storage phosphor plate is performed as part of a routine examination. This means that the defect detection method can also be carried out on storage phosphor plates that have been exposed to, for example, patients and therefore contain an image. The evaluation method—that is, determining the signal-to-noise ratio and / or performing edge detection—records detected defects and assigns them to the storage phosphor plate. After repeated use of the storage phosphor plate and another defect detection procedure, any recurring defects can be registered and marked as such. Any artifacts that may occur, such as those caused by radiopaque objects (e.g., metal fillings) that appear on the image, can be detected, evaluated, and marked accordingly.

[0015] A further development of the invention provides that determining the signal-to-noise ratio includes the application of a mask filter. The mask filter can, for example, exclude an edge of the storage plate—e.g., one or two millimeters—from the calculation of the signal-to-noise ratio. This edge of the storage plate can be predefined or entered by an operator. Alternatively or additionally, the mask filter can exclude other markings on the storage plate besides the edge. These markings can also be detected using morphological operations and added to the mask filter.

[0016] In a particularly preferred embodiment of the invention, determining the signal-to-noise ratio includes determining a local signal-to-noise ratio for one or more sections of the image. For example, a field measuring 21 by 21 pixels can serve as the basis for calculating the signal-to-noise ratio. For such a field size, a local signal-to-noise ratio can be calculated for the entire image. This results in a corresponding signal-to-noise ratio value for each pixel of the image (with the exception of a border corresponding to the 21 by 21 pixels). Areas with a low local signal-to-noise ratio correspond to areas that are presumed to have a defect. A threshold value can be defined for this purpose.The threshold can be determined, for example, using the maximum and minimum values ​​of the locally measured signal-to-noise ratio. This can be done by calculating the ratio of the signal-to-noise value (minus the minimum value) to the difference between the maximum and minimum values. A pixel can be marked as defective, for instance, if the ratio described above is 0.2 or less, meaning that the local signal-to-noise ratio is 20% or less of the possible range.

[0017] Alternatively or additionally, in one embodiment, damage to the storage phosphor plate can be determined based on a detected edge structure. Edge detection allows defects – for example, non-functional storage centers – to be identified as structures in the otherwise uniformly exposed phosphor plate.

[0018] In a preferred embodiment, performing edge detection includes applying the Canny algorithm. Preferably, blurring or the addition of a fuzzy layer can be performed before edge detection. This reduces the production of artifacts due to pixels with a high noise level. Furthermore, a thresholding operation can be performed after edge detection, preferably to obtain a binary image.

[0019] In a further development of the invention, it can be provided that, after performing edge detection, particularly after performing the threshold operation, a morphological dilation and / or a closing operation is carried out. By means of the morphological dilation and the closing operation, the visibility of individual areas with irregularities / damage can be improved, since, in principle, values ​​with a high gray value are enlarged and values ​​with a lower gray value are reduced. At the same time, small gaps between such areas can be closed.

[0020] The image obtained from the signal-to-noise conversion and the image resulting from edge detection are combined using a logical OR operation. The quality of the storage phosphor plate is determined based on the total number of marked pixels in this process. For example, the number of marked pixels can be used to determine the quality of the storage phosphor plate. N markedThe following formula can be used to convert the data into a number representing the quality of the storage phosphor: f markiert = − ln N markiert N gesamt − N Maske

[0021] This presents f marks the quality assessment of the storage foil, N marked the number of marked image pixels, N total represents the total number of pixels in the viewed image and N mask This represents the number of pixels masked by the exclusion mask at the film's edge and the markings. Thus, a storage film with an image that has a very low number of marked pixels receives a high rating, while storage films whose images have a high number of marked, i.e., defective, pixels receive a low rating.

[0022] In order to scale the resulting rating of a storage foil into a range between 0 and 1 and to allow an adjustment of the weighting of the individual determined / marked features, the rating can be f marks can be transformed as follows: f final = w 0 + w 1 f markiert

[0023] This allows for a comparison of the resulting assessment with other evaluation criteria, such as those obtained through visual inspection and subsequent evaluation of a storage phosphor plate. For example, it may be stipulated that the parameters w 0 and w 1. The value is chosen by linear regression on a training data set such that the resulting final value corresponds as closely as possible to an expert assessment of the slide quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Exemplary embodiments of the invention are explained in more detail below with reference to the drawings. These show: Figure 1 shows a flowchart of a first embodiment of a method according to the invention; Figure 2 shows a flowchart of a first further development of the method. Figure 1 Figure 3 shows a second embodiment of the method in a flowchart. Figure 1Figure 4 shows a second alternative or additional embodiment of the method in a flowchart. Figure 3 ; and Figure 5 shows a schematic representation of a storage phosphor scanner according to the invention. DESCRIPTION OF PREFERRED EXAMPLES

[0025] Figure 1 A simplified schematic flowchart illustrates an embodiment of a method according to the invention. The method comprises the step of exposing a storage phosphor plate (S1). When exposing the storage phosphor plate, it is advantageous if it is exposed to a specific dose. It has proven advantageous if the storage phosphor plate is positioned within a predefined distance of the imaging device. In practice, a distance of 12 cm, which can be set to the diameter of a CD, has proven effective. The setting for the X-ray unit can, for example, be the setting for modular teeth.

[0026] In the next step, the storage phosphor plate is read out and a digital image is generated (S2). After or during the readout process, the exposure dose to the phosphor plate can be determined. An output can be sent to the operator indicating whether the dose was too high or too low. If necessary, the operator must re-expose the phosphor plate, as the signal-to-noise ratio can change with the dose.

[0027] In the next step, the image is preprocessed to determine the quality of the storage phosphor plate (S3). The preprocessing step (S3) is explained in detail below.

[0028] After preprocessing (S3), two process blocks can be processed simultaneously in parallel, sequentially, or alternatively to each other.

[0029] The preprocessed image can be transformed into a mask in which pixels with a signal-to-noise ratio below a certain threshold are marked (S4). Alternatively or additionally, edge detection can be performed on the preprocessed image (S5), which can then be used to generate a mask in which the detected pixels are marked.

[0030] Using these two masks, a total number of marked pixels can be determined (S6). By relating this number of marked pixels to the total number of all pixels (reduced by the number of pixels in the exclusion mask), the quality of the storage phosphor plate can be determined based on this ratio (S7).

[0031] The following section discusses in more detail the preprocessing (S3) of the image obtained from the readout process. Figure 2Figure 3 shows a possible embodiment of the preprocessing process. A border mask is provided for the read image (S31). This border mask can, for example, be a space around the edge of the image, outside of which the image data is discarded or not used for processing. For common storage film sizes, for example, for the oral cavity (from 2×3 cm to approximately 6×8 cm), one to two millimeters should be left as a border. This distance can be fixed or freely selectable by the operator.

[0032] In a further step, possible markings on the storage phosphor plate that do not belong to the actual image content are detected (p. 32). This detection can be performed, for example, using a morphological operation and can search for known structures within the image content. Detected markings or similar structures are added to the boundary mask, which then becomes an exclusion mask (p. 33).

[0033] Optionally, the margin or exclusion mask created in this way can be displayed to an operator for control and / or information purposes (S34).

[0034] The exclusion mask or border mask thus created is applied to the extracted image (p. 35), resulting in a preprocessed image. This is then used as the basis for one of the next steps (p. 4, p. 5).

[0035] Figure 3A flowchart illustrates some aspects of an embodiment of a method for determining the number of non-operable pixels of a storage phosphor plate by calculating a signal-to-noise ratio (S4). The method (S4) first comprises performing (S41) a local signal-to-noise ratio (SNR) calculation. An averaging field is defined, within which the SNR is determined as a single value for all pixels contained therein. This value is then assigned to one of the pixels within the averaging field. This calculation step is performed at all possible positions in the image. The averaging field is, in effect, shifted pixel by pixel, and a local SNR is calculated at each new position. Thus, an SNR is obtained for almost every pixel in the image (with the exception of a border strip corresponding to the chosen size of the averaging field).These values ​​can be interpreted and represented as the SNR image of the preprocessed image (p. 42). In practice, a mean square of 21 × 21 pixels has proven advantageous for the aforementioned storage phosphor plate sizes and corresponding image sizes. Other mean square sizes are also conceivable, for example, half or twice as large.

[0036] In a further step (S43), an SNR range is determined as an additional parameter (S43). The local SNR values ​​can be used to calculate the SNR range. For example, a minimum and a maximum value can be determined. Alternatively, a global SNR value can be calculated from the local SNR values ​​or solely based on the preprocessed image.

[0037] In a further step (S44), an SNR threshold is determined. This can be based on the SNR range. Alternatively or additionally, a local SNR value can be included in the determination of the SNR threshold. The SNR threshold can be calculated, for example, as follows: SNR lokal x y − SNR min SNR max − SNR min , where SNR local ( x, y ) the local SNR value, SNR min the minimal and SNR max This represents the maximum SNR value. If this value is 20% or below, the corresponding pixel is marked as defective. However, other percentage values ​​such as 10% or 30% are also conceivable.

[0038] Using the SNR threshold determined in this way, the pixels in the SNR image whose SNR is below the SNR threshold can now be identified (S45).

[0039] In a further step (S46), the pixels marked in this way are combined into a mask which marks the areas of the storage phosphor that are no longer operable - determined on the basis of the SNR.

[0040] Figure 4 A flowchart illustrates an embodiment of a method (S5) for determining non-functional areas of a storage phosphor plate using an edge detection method. This embodiment is used simultaneously with the method that provides for determining the signal-to-noise ratio.

[0041] In a first step, a blur is added to the preprocessed image (p. 51). The purpose of adding the blur is to prevent the edge detection algorithm from reacting too strongly to individual, highly noisy pixels.

[0042] Next, edge detection is performed on the prepared image (S52). The Canny algorithm, for example, can be used for this purpose. However, other edge detection algorithms are also conceivable. The result of the edge detection algorithm is again an image, which is converted into a binary image (where only 0 or 1 exist as image values) using a threshold operation (S53). The choice of threshold depends, for example, on the general image quality, the noise level, the quality of the readout process, etc.

[0043] To improve the visibility of the detected structures, a morphological dilation and a closing operation can be performed on the edge detection image in order to fill small gaps or holes in the edges (p. 54). This results in a second mask, which also marks areas in the image, and thus indirectly on the storage phosphor plate, that are no longer functioning correctly.

[0044] The image masks generated in these two ways can be combined, for example, using a logical OR operation. The resulting number of damaged pixels can then be converted into an evaluation of the storage phosphor plate, as explained in detail above, according to the formulas. f markiert = − ln N markiert N gesamt − N Maske and f final = w 0 + w 1 f markiert

[0045] The weighting factors w 0 and w 1 can be used to adapt the quality control procedure to the desired scaling of the quality rating of the respective storage foil and can, for example, be determined by means of a linear regression on a training data set so that the resulting final value corresponds as closely as possible to an expert assessment of the foil quality.

[0046] A comparison of the quality classification using the procedure described above and the classification results of experts (dentists, technicians) revealed a largely high correlation between the two classifications.

[0047] Figure 5Figure 10 shows a scanning device 10 for reading a storage phosphor plate 12, which carries a latent X-ray image in the form of metastable storage centers excited by X-rays. The scanning device 10 has a support device 14 for the storage phosphor plate 12. For example, the storage phosphor plate 12 can be attached to the support device 14 by means of a vacuum so that the storage phosphor plate 12, which is generally flexible, conforms flat to the support surface 14. The scanning device 10 further includes a laser 16 as a readout light source, which generates a readout light beam 18 with a wavelength in the red region, with which the metastable storage centers of the storage phosphor plate 12 can be excited to fluorescence light. This fluorescence light 20 is typically in the blue region.

[0048] In the present embodiment of the scanning device 10, the laser 16 is arranged such that it directs the readout light beam 18 onto a controllable deflection unit. In this case, the controllable deflection unit is designed as a mirror 22. Other deflection units, such as optics or the like, are also conceivable. The mirror 22 can be designed as a micromirror, in particular as a MEMS component, thus enabling scanning of the surface of the storage phosphor plate 12 with no or only minimal relative movement between the mirror 22 and the support device 14. Alternatively, the mirror 22 can also be designed as a rotating mirror for a drum scanner. In this case, relative movement between the support device 14 and the mirror 22 is achieved by means of a transport mechanism (not shown). Another technique for reading the storage phosphor plate 12 involves the use of a rotating pentaprism.

[0049] The scanning device 10 can further comprise a reflector 24, indicated by dashed lines in the drawing, which completely encloses the measuring space around the storage plate 12 in a light-tight manner, so that the fluorescence light 20 emanating from the storage plate 12 is reflected to a photodetector 26. To prevent scattered readout light 18 from entering the photodetector 26, suitable measures such as a dichroic filter material can be provided. For controlling the readout process, the scanning device 10 includes a control unit 28, which, in addition to the control function, can also perform evaluation or correction functions. In the embodiment shown here, the control unit 28 is configured to carry out one of the aforementioned methods for determining the quality of a storage plate.

Claims

1. A method for determining the quality of an imaging plate, comprising the steps of a) carrying out an exposure of the imaging plate (S1); b) carrying out a scan of the imaging plate in order to determine an image (S2); c) determining a signal-to-noise ratio of the image (S4) and carrying out an edge recognition on the image (S5); and d) calculating a quality value of the imaging plate on the basis of the signal-to-noise ratio of the image and on the basis of the recognized edge structure (S6, S7), wherein a number of instance of damage is determined in order to ascertain the quality of an imaging plate on the basis of an image which a logic OR-link between an image which identifies instances of damage on the basis of a low signal-to-noise ratio and an image which identifies instances of damage on the basis of an edge recognition.

2. The method as claimed in claim 1, wherein the step of carrying out an exposure comprises setting a specific distance between the plate and a recording device.

3. The method as claimed in either of the preceding claims, wherein determining a signal-to-noise ratio comprises applying a mask filter (S31).

4. The method as claimed in any of the preceding claims, wherein determining the signal- to-noise ratio comprises determining a local signal-to-noise ratio for one or more segments of the image (S41, S42).

5. The method as claimed in any of the preceding claims, comprising the step of ascertaining a number or / and a size of instances of damage on the basis of a locally low signal-to-noise ratio.

6. The method as claimed in any of the preceding claims, wherein an instance of damage is ascertained on the basis of a recognized edge structure (S5).

7. The method as claimed in any of the preceding claims, wherein carrying out an edge recognition comprises using the Canny algorithm.

8. The method as claimed in any of the preceding claims, wherein after carrying out an edge recognition, a morphological dilatation or / and a closing operation is performed (S54).

9. An imaging plate scanner configured to carry out one of the methods mentioned above