JP6433220B2 - X-ray image processing apparatus, method, and program - Google Patents

Description

  The present invention relates to a technique related to a sensor that detects X-rays, and particularly relates to a technique for performing a constancy test of a sensor having a function of automatically detecting X-rays.

  In a medical X-ray imaging apparatus, a diagnostic action is performed based on an output image. Therefore, it is desirable to always output an image with stable image quality without causing the apparatus to fail. Therefore, it is indispensable to inspect the entire apparatus on a daily basis and manage its quality. This test for quality control is called constancy test. In the invariance test, the X-ray generator, the X-ray sensor, and the geometric characteristics of both of them are inspected to check whether the image performance meets the application criteria. It is determined whether it is unchanged or not. In this specification, the constancy test for the X-ray sensor, particularly the constancy test in the state of irradiation with X-rays will be described below as a scope.

  In the constancy test for X-ray sensors, when the X-ray sensor is evenly irradiated with X-rays in the absence of a subject, there is no deterioration in the output from the sensor, or there is in-plane variation in the output from the sensor. Check if there is any. Specifically, a specific evaluation index representing the degree of deterioration or variation is calculated from the output from the sensor, and when the evaluation index is within the threshold, the image performance of the sensor output image is unchanged. Judge.

  On the other hand, in order to acquire an X-ray image with an X-ray sensor, the sensor needs to know that X-ray irradiation has started. There are a plurality of methods (hereinafter, detection modes) in which the sensor detects the start of X-ray irradiation. For example, there is a method in which the X-ray sensor acquires the timing at which X-ray irradiation starts from the X-ray generator, and starts acquiring the current in the sensor at that timing, that is, starts imaging (hereinafter, synchronous detection mode). . In addition, there is a method in which the X-ray sensor acquires the timing at which X-ray irradiation starts from the control software of the X-ray image apparatus instead of the X-ray generation apparatus (hereinafter, manual synchronization detection mode). In this method, the user presses a switch for starting irradiation at a timing designated by the control software. In addition, there is a method in which the X-ray sensor automatically detects that X-ray irradiation has started without using an X-ray generator or control software, and starts acquiring current in the sensor ( Hereinafter, automatic detection mode).

  In Patent Documents 1 and 2, as a method for automatically detecting the start of X-ray irradiation, a radiographic imaging apparatus waits for irradiation while sequentially selecting each scanning line of the sensor and switching between on and off states. A method for detecting the start of X-ray irradiation based on the detection of a change in the current flowing inside is disclosed.

JP 2011-247605 JP JP 2011-249891 A Japanese Patent No. 3445258

  In the synchronous detection mode, since it is necessary to link an X-ray sensor that acquires information from the X-ray generator and the X-ray generator in advance, user convenience is poor. Also in the manual synchronization detection mode, X-rays can be emitted only at timings specified by the user, and the user's usability is similarly poor. On the other hand, in the automatic detection mode, it is not necessary to link the X-ray sensor, the X-ray generator, and the control software, so the user convenience is good. However, X-rays are lost until the X-ray sensor detects X-ray irradiation, and this loss occurs as artifacts and defects on the X-ray image.

  In the invariance test, it is assumed that X-rays are evenly irradiated to the X-ray sensor, but the assumption is not satisfied when artifacts and defects occur. As a result, the specific evaluation index indicating the degree of deterioration or variation becomes an index value including artifacts and defects, and the reliability of the index value is reduced. As a result, the reliability of the constancy test is lowered. On the other hand, there is a limit to the ability to remove the influence of artifacts, and when the amount of artifacts exceeds a predetermined amount, the influence of artifacts cannot be removed, leading to a decrease in the reliability of the index value.

  Accordingly, an object of the present invention is to perform a reliable test for quality control on an X-ray image apparatus.

  As a means for achieving the above object, an X-ray image processing apparatus of the present invention has the following configuration. That is, an acquisition unit that acquires an image generated by a sensor that detects X-rays emitted from a radiation source, and an evaluation value for evaluating the quality of the image based on the image acquired by the acquisition unit A generating unit for generating, a determining unit for determining an evaluation result for the quality of the image based on the evaluation value generated by the generating unit, and a display for displaying the evaluation result determined by the determining unit Corresponding to a plurality of different detection modes for detecting the start of the X-ray irradiation set in the sensor, each of the generation means or the determination means performs different processing. It is characterized by.

  It is possible to perform a reliable test for quality control on the X-ray imaging apparatus.

1 is a configuration diagram of an X-ray image processing apparatus according to Embodiment 1. FIG. 1 is a hardware configuration diagram according to Embodiment 1. FIG. 5 is a flowchart of processing of the X-ray image processing apparatus according to Embodiment 1-1. The figure explaining the estimation of an artifact area | region. The example of the display to the display device in Embodiment 1-1. 6 is a flowchart of processing of the X-ray image processing apparatus according to Embodiment 1-2. The example of the display to the display device in Embodiment 1-2. The figure which shows the relationship between ROI and an artifact area | region. 7 is a flowchart of processing of the X-ray image processing apparatus according to Embodiment 1-3. 10 is a flowchart of processing of the X-ray image processing apparatus according to Embodiment 1-4. 7 is a flowchart of processing of the X-ray image processing apparatus according to Embodiment 1-6. FIG. 3 is a configuration diagram of an X-ray image processing apparatus according to a second embodiment (part 1). 10 is a flowchart of processing of the X-ray image processing apparatus according to Embodiment 2-1. 10 is a flowchart of processing of the X-ray image processing apparatus according to Embodiment 2-2. Configuration of the X-ray image processing apparatus according to the second embodiment (part 2). 10 is a flowchart of processing of the X-ray image processing apparatus according to Embodiment 2-4. The example of the display to the display device in Embodiment 2-4.

  Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the drawings. First, an embodiment in which quality control evaluation is performed while reducing the influence of artifacts on an image in the automatic detection mode will be described as the first embodiment. Further, as the second embodiment, an embodiment in which quality control evaluation is performed according to the amount of artifact on an image in the automatic detection mode will be described.

[Embodiment 1]
FIG. 1 shows a configuration diagram of the X-ray image processing apparatus according to the present embodiment. The X-ray image processing apparatus according to this embodiment includes an image acquisition unit 101, a switching unit 102, image analysis units 103 and 104, and a display unit 105. The image acquisition unit 101 receives an X-ray image acquired by an X-ray sensor, generates an image subjected to predetermined preprocessing (hereinafter, preprocessed image), and outputs the generated image. The switching unit 102 receives an X-ray detection mode and selects one unit from among a plurality of image analysis units corresponding to the detection mode. In FIG. 1, the switching unit 102 selects either the image analysis unit 103 or the image analysis unit 104.

  The image analysis unit 103 receives a preprocessed image (invariance test image) as an input, analyzes the preprocessed image, calculates an evaluation index for the invariance test, and outputs an evaluation result. Similarly to the image processing unit 103, the image analysis unit 104 receives the preprocessed image, analyzes the preprocessed image, calculates an evaluation index for the invariance test, and outputs an evaluation result. However, the image analysis units 103 and 104 have different image analysis methods. In FIG. 1, two types of image analysis units 103 and 104 are prepared as different image analysis means. However, the number is not limited to two, and any number of two or more may be used. The display unit 105 receives the evaluation result calculated by the image analysis unit 103 or 104 and outputs the evaluation result to a display device such as a monitor.

  A configuration example in the case of realizing the configuration of FIG. 1 using a PC will be described with reference to FIG. FIG. 2 is a hardware configuration diagram including the X-ray image processing apparatus according to the present embodiment. The control PC 201, the X-ray sensor 202, and the X-ray generator (radiation source) 210 are connected by a gigabit ether 222. In FIG. 2, the signal line is the Gigabit Ethernet 222, but it may be a CAN (Controller Area Network) or an optical fiber. A display unit 209, a storage unit 211, and a network interface (I / F) unit 212 are connected to the gigabit ether 222. In the control PC 201, for example, a CPU (Central Processing Unit) 203, a RAM (Random Access Memory) 204, a ROM (Read Only Memory) 205, an input unit 206, a display unit 207, and a storage unit 208 are connected to the bus 221. The A command is sent from the control PC 201 to the X-ray sensor 202, the display unit 209, and the like.

  In the control PC 201, the processing content for each shooting mode is stored as a software module in the storage unit 208, read into the RAM 204 by an instruction unit (not shown), and executed. The image processing unit 101, the switching unit 102, and the image analysis units 103 and 104 illustrated in FIG. 1 are stored in the storage unit 208 as software modules. Of course, the image processing unit 101, the switching unit 102, and the image analysis units 103 and 104 shown in FIG. 1 may be mounted as a dedicated image processing board. In other words, an optimal mounting may be performed according to the purpose. The display unit 105 illustrated in FIG. 1 corresponds to the display unit 209 via the Gigabit Ethernet 222. The display unit 105 shown in FIG. 1 may be the display unit 207 in the control PC.

  Hereinafter, the details of the processing of the X-ray image processing apparatus according to the present embodiment will be described along the six embodiments 1-1 to 1-6. There are various evaluation indexes in the constancy test. In the embodiments 1-1 to 1-6 described below, the following two examples are given as examples. The first is a pixel average value calculated by a plurality of ROIs (Region Of Interest) in the image. In this case, the pixel average value itself is checked and whether the pixel average value is uniform in the image is checked. The second is a variance value calculated by a plurality of ROIs in the image. In this case, the variance value itself is checked and whether the variance value is uniform is checked. For this reason, the imaging condition for the constancy test is to expose X-rays without a subject. Further, when shooting is performed in a state where there is a grid, the value of the evaluation index to be calculated varies depending on the type of grid (material, number, grid ratio, etc.), and correct evaluation cannot be performed. Therefore, it is assumed that shooting is performed with the grid removed. In addition, there are many X-ray automatic detection methods, but in this embodiment, a detection method is adopted for each scanning line as in Patent Documents 1 and 2.

[Embodiment 1-1]
Embodiment 1-1 will be described along the flow of processing using the configuration diagram of FIG. 1 and the overall flow of FIG. FIG. 3 is a flowchart of processing of the X-ray image processing apparatus according to the present embodiment. The image acquisition unit 101 acquires an image with the X-ray sensor 202 and performs preprocessing (steps 301 and 302). The preprocessing is processing for correcting sensor characteristics such as offset correction and gain correction. By such processing, the image acquisition unit 101 brings the acquired image into a state in which the correlation with the surrounding pixels is maintained.

  Next, the switching unit 102 acquires the detection mode from information set by the user (step 303). If the acquired detection mode is the automatic detection mode, the switching unit 102 selects the image analysis unit 103 (step 304). If the acquired detection mode is other than the automatic detection mode, the image analysis unit 104 is selected (step 304).

  The image analysis unit 103 corrects the artifact in the automatic detection mode from the preprocessed image (step 305). Here, with reference to FIG. 4, the artifact correction in the automatic detection mode will be described. FIG. 4 is a diagram for explaining the estimation of the artifact region. In this embodiment, an X-ray sensor used for imaging is used as a method for automatically detecting the start of X-ray irradiation. And since it is set as the method of detecting for every scanning line like patent document 1 or 2, an artifact generate | occur | produces in the area | region as shown in FIG. In other words, all the scanning lines that have been released until the X-ray is detected become artifact lines. Therefore, when the scanning lines are continuously driven, an artifact area 410 is generated as one area as shown in FIG. When scanning lines are driven in an interlaced manner, an artifact area 420 is generated as an area in which artifact lines and normal lines are alternately mixed as shown in FIG. 4B.

  In order to correct the artifact, the image analysis unit 103 first extracts a region where the artifact is generated (artifact region). The artifact region is a region from when X-ray exposure is started until the X-ray sensor detects X-ray exposure. The line where the X-ray is detected (hereinafter referred to as the stop line) can be obtained as information from the X-ray sensor, but the line where the X-ray exposure has started (start line) is obtained as information from the X-ray sensor. I can't get it. Therefore, in order to extract the artifact region, it is first necessary to estimate the start line. The method for estimating the start line is shown below.

First, when the artifact pixel is projected in the scanning line direction (x direction), the value of the artifact line PV (projection data 411 and 421) is expressed by Equation 1. Here, V _D is a residual dark current component, V ′ is a pixel value (hereinafter, true value) in a normal operation, and R is a value proportional to the charge release time.
(Formula 1)

  The constancy test in this embodiment is performed in the absence of a subject. For this reason, the sensor output is substantially uniform. Since the dark current component is also substantially uniform, it can be thought of as a linear expression. Therefore, the projection data (PV (y)) is fitted with a linear expression. Then, the linear expression is estimated as a start line (start line estimation 412, 422) that takes the average of the lines where no artifact has occurred. Thereby, an artifact area can be extracted.

The image analysis unit 103 corrects the artifact for the extracted artifact region. For example, correction is performed so that the true value average of each line is obtained. Assuming that the average of the lines in which the artifacts of each line are not generated is V _m , the corrected artifact pixel value V ″ is obtained as shown in Equation 2. w is the number of pixels in the line direction.
(Formula 2)

画像解析部１０３は、アーチファクト補正した画像から、不変性試験用の評価指標を算出する（ステップ３０６）。本実施形態による画像解析部１０３は、画像内に複数のROIを設定し、評価指標として、ROI毎に平均値や分散値を算出する。画像解析部１０３は、各ROIにおいて、平均値や分散値が所定の閾値範囲内であることを調べることで、出力値の不変性を確認することができる。また、画像解析部１０３は、ROI毎の平均値や分散値を比較し、それらが所定の閾値範囲内であることを調べることで、面内一様の不変性を確認することができる。 In the case of interlaced driving as shown in FIG. 4B, the average value of adjacent normal lines may be used as V _m in Equation 2. Or you may correct | amend using the pixel of the normal line which adjoins every artifact pixel like defect pixel correction | amendment. The corrected artifact pixel value V ″ in this case is shown in Equation 3.
(Formula 3)

The image analysis unit 103 calculates an evaluation index for the invariance test from the artifact-corrected image (step 306). The image analysis unit 103 according to the present embodiment sets a plurality of ROIs in an image and calculates an average value and a variance value for each ROI as an evaluation index. The image analysis unit 103 can confirm the invariance of the output value by checking that the average value and the variance value are within a predetermined threshold range in each ROI. Further, the image analysis unit 103 can confirm the in-plane invariance by comparing the average value and the variance value for each ROI and checking that they are within a predetermined threshold range.

  The image analysis unit 103 compares the evaluation value calculated in step 306 with a threshold value, and obtains the result of the invariance test (step 307). The image analysis unit 103 passes the invariance test if all the evaluation items are within the threshold, and rejects if there are items outside the threshold. At this time, the image analysis unit 103 uses a threshold for the automatic detection mode. In the automatic detection mode, since an image in which artifacts are corrected is used, a threshold different from the threshold for the synchronous detection mode is used in accordance with the correction method. In particular, the dispersion value is corrected as follows. Note that, as described above, the synchronization detection mode is a detection mode in which the X-ray sensor acquires the timing at which X-ray irradiation starts from the X-ray generator and starts imaging at that timing.

In other words, when the attenuation is amplified and corrected as shown in Equation 2, the noise is also amplified together, so that the threshold value of the dispersion value is made _lower than the threshold value Th _VarS of the synchronization detection mode. For example, in Equation 2, the noise is increased by an amplification factor (V _m / PV) times. When the ratio of the artifact pixels included in the ROI is r (0 ≦ r ≦ 1), the threshold value Th _VarA in the variance value is _expressed by Equation 4.
(Formula 4)

On the other hand, when the correction is performed using the average value as in Expression 3, the noise is attenuated on the contrary, and therefore, the threshold value of the dispersion value is made stricter than the threshold value Th _VarS of the synchronization detection mode. For example, in Expression 3, in the case of correction from two adjacent pixels, the noise is 1 / √2 times. As in Equation 4, when the ratio of artifact pixels included in the ROI is r (0 ≦ r ≦ 1), the threshold value Th _VarA in the variance value is as _shown in Equation 5.
(Formula 5)

Thus, prepared by the threshold value Th _vars for synchronization detection mode, the threshold for automatic detection mode may be used to calculate Th _VarA using Equation 4 and Equation 5, previously, for the automatic detection mode Alternatively, the threshold may be prepared and switched. Further, if the used artifact correction method is a correction method that does not change the magnitude of noise, there is no problem even if the same threshold value for the synchronization detection mode is used. Also, the threshold value for the automatic detection mode with respect to the average value can be calculated from the threshold value for the synchronization detection mode.

  Finally, the display unit 105 displays the invariance test result obtained in step 307 on a display device such as a monitor. FIG. 5 shows an example of display on the display device in the present embodiment. If the constancy test result is unacceptable, the display unit 105 displays a warning message such as a message prompting the constancy test to be performed again or a message urging contact with a service person (repair person) as shown in FIG. Is displayed. In the automatic detection mode, since the artifact correction may have failed and the invariance test may have failed, the display unit 105 displays a message indicating the possibility or a message for performing the invariance test in the synchronous detection mode. Is displayed. The display unit 105 may display an artifact correction image so that the user can check the image. At that time, the display unit 105 may display an artifact area, a stop line, a start line, or the like, or display the image so that the image can be enlarged or adjusted, so that the user can check the image.

When the detection mode is other than the automatic detection mode, the switching unit 102 selects the image analysis unit 104 (step 304). The image analysis unit 104 calculates an evaluation index for the constancy test from the preprocessed image (step 306). The evaluation index to be calculated may be the same as the automatic detection mode described above. The image analysis unit 104 determines the result of the invariance test by comparing the calculated evaluation index with a threshold value (step 307). As the threshold value, the above-described threshold value Th _VarS for the synchronization detection mode is used. Finally, the display unit 105 displays the result of the constancy test. If the constancy test result is unacceptable, the display unit 105 displays a warning message such as a message prompting the constancy test to be performed again or a message urging contact with the service person.

[Embodiment 1-2]
As Embodiment 1-2, another embodiment regarding a method for avoiding the influence of artifact lines will be described along the flow of processing using the configuration diagram of FIG. 1 and the entire flow of FIG. FIG. 6 is a flowchart of processing of the X-ray image processing apparatus according to the present embodiment. Since the processing from step 601 to 604 is the same as the processing from step 301 to 304 in the embodiment 1-1, description thereof will be omitted.

  The image analysis unit 103 identifies the amount of artifact included in the specific ROI from the preprocessed image (step 605). The method of estimating the artifact line is performed by the method shown in the embodiment 1-1. Here, ROI is an area for obtaining the evaluation index calculated in step 607. At this time, if the content rate of the artifact line in the ROI is larger than a specific threshold value, the accuracy as the evaluation index cannot be maintained, so the display unit 104 displays a message prompting re-shooting on the display device (in step 606). Yes, step 610). For example, if the threshold value of the content rate is 0%, re-photographing is prompted if there is an artifact line even with one line.

  If the artifact is not included in the ROI including the artifact in the previous imaging due to the re-imaging, the one extracted from the preprocessed image obtained in the re-imaging is used as the ROI for calculating the evaluation index in step 607. . If no artifact is included in the ROI for both the previous shooting and the re-shooting, the ROI extracted from either preprocessed image in step 607 may be used. In this way, the evaluation index is calculated based on the image from which the influence of the artifact is removed by re-photographing (step 607). Note that if the artifact line content rate becomes larger than a specific threshold value with the same ROI even after re-photographing, re-photographing is prompted via the display unit 105. In addition, when the photographing is repeated a specific number of times, the constancy test in the automatic detection mode is stopped, and the constancy test in the synchronous detection mode is promoted via the display unit 105.

  The display unit 105 performs display so that all of the acquired plurality of preprocessed images (invariance test images) can be confirmed, including the case where re-photographing is performed. FIG. 7 shows an example of display on the display device in the present embodiment. The display unit 105 switches and displays images as shown in FIG. 7A or displays a list of images as shown in FIG. 7B. In the case of a list display, the selected item may be enlarged and displayed. Further, as shown in FIG. 7C, the display unit 105 displays the position of the overlapping artifact line, so that the user can check which part of the data is insufficient.

  The image analysis unit 103 calculates an evaluation index for the constancy test (step 607). The image analysis unit 103 according to the present embodiment sets a plurality of ROIs in an image and calculates an average value and a variance value for each ROI as an evaluation index. FIG. 8 shows the relationship between the ROI and the artifact area. When the artifact line is included in the ROI as shown in FIG. 8, the image analysis unit 103 sets the area from which the artifact line is removed as the ROI for calculating the evaluation index, and calculates the average value or only from the area where the artifact is not generated. Calculate the variance value. If the threshold of the artifact line content rate is set to 0% in step 606, the artifact line is not included, so there is no need to consider such processing.

  The image analysis unit 103 compares the evaluation value calculated in step 607 with a threshold value, and obtains the result of the invariance test (step 608). Since the evaluation value is calculated in a state in which the influence of the artifact is removed, the image analysis unit 103 does not need to prepare a threshold value for the automatic detection mode as in Embodiment 1-1. The result of the constancy test is obtained using the threshold value as it is. Since the process of step 609 is the same as the process of step 308 of Embodiment 1-1, a description thereof is omitted.

[Embodiment 1-3]
As Embodiment 1-3, another embodiment regarding the method for avoiding the influence of artifact lines will be described along the flow of processing using the configuration diagram of FIG. 1 and the entire flow of FIG. FIG. 9 is a flowchart of processing of the X-ray image processing apparatus according to the present embodiment. Since the processing from step 901 to 904 is the same as the processing from step 301 to 304 in the embodiment 1-1, description thereof will be omitted.

  The image analysis unit 103 estimates an artifact line from the preprocessed image (step 905). The method of estimating the artifact line is performed by the method shown in the embodiment 1-1. The estimated artifact line is replaced with a line at the same position in the preprocessed image of the previous shooting (step 906). An evaluation index is calculated based on the image from which the influence of the artifact has been removed by the replacement, that is, the image after the replacement (step 908). However, if there is an artifact line in the previous captured image, the display unit 104 displays it on the display device so as to prompt re-imaging without replacement (No in step 907, step 911). In addition, when the photographing is repeated a specific number of times, the constancy test in the automatic detection mode is stopped and the constancy test in the synchronous detection mode is prompted.

  The image analysis unit 103 calculates an evaluation index for the constancy test (step 908). The image analysis unit 103 according to the present embodiment sets a plurality of ROIs in an image and calculates an average value and a variance value for each ROI as an evaluation index. The image analysis unit 103 compares the evaluation value calculated in step 907 with a threshold value, and obtains the result of the invariance test (step 909). Since the evaluation value is calculated in a state in which the influence of the artifact is removed, the image analysis unit 103 does not need to prepare a threshold value for the automatic detection mode, and uses the threshold value in the synchronous detection mode as it is. Get the result. Since the process of step 910 is the same as the process of step 308 of Embodiment 1-1, a description thereof is omitted.

[Embodiment 1-4]
As Embodiment 1-4, another embodiment regarding result determination will be described along the flow of processing using the configuration diagram of FIG. 1 and the entire flow of FIG. FIG. 10 is a flowchart of processing of the X-ray image processing apparatus according to the present embodiment. The processing from steps 1001 to 1004 is the same as the processing from steps 901 to 904 in the embodiment 1-3, and thus the description thereof is omitted.

  The image analysis unit 103 calculates an evaluation index for the constancy test (step 1004). The image analysis unit 103 according to the present embodiment sets a plurality of ROIs in an image and calculates an average value and a variance value for each ROI as an evaluation index. The image analysis unit 103 compares the evaluation value calculated in step 1004 with a threshold value, and obtains the result of the invariance test (step 1005). Since the evaluation value is calculated using the preprocessed image, the image analysis unit 103 does not need to prepare a threshold value for the automatic detection mode, and uses the threshold value in the synchronous detection mode as it is. Get.

  If the result of the constancy test is unacceptable, the image analysis unit 103 checks whether or not the mode in which the constancy test is being performed is the automatic detection mode (No in step 1006, step 1007). When the constancy test is passed in step 1006, the display unit 104 displays a message indicating that it is passed on the display device (Yes in step 1006, step 1011). If the constancy test fails in step 1006 and the automatic detection mode is not set in step 1007, the display unit 104 prompts the service person to contact the customer because the constancy of the X-ray sensor is impaired. The message is displayed on the display device (No in step 1006, No in step 1007, and step 1011).

  If the constancy test fails in step 1006 and the automatic detection mode is set in step 1007, the image analysis unit 103 estimates an artifact line from the preprocessed image (No in step 1006, Yes in step 1007). Step 1008). The method of estimating the artifact line is performed by the method shown in the embodiment 1-1. The image analysis unit 103 calculates the evaluation index again (step 1009). The image analysis unit 103 sets the same ROI in step 1004 in the image, and recalculates the average value and the variance value for each ROI. At that time, when the artifact line is included in the ROI as shown in FIG. 8, the image analysis unit 103 sets the area from which the artifact line is removed as the ROI for evaluation index calculation, and only from the area where the artifact does not occur. Calculate the mean and variance.

  The image analysis unit 103 compares the evaluation value calculated again in step 1009 with a threshold value, and obtains the result of the invariance test (step 1010). As the threshold value, the threshold value in the synchronization detection mode is used as it is. If the determination result is acceptable, the determination result in step 1005 is likely to be caused by an artifact line. Therefore, the display unit 104 displays on the display device so as to stop the constancy test in the automatic detection mode and prompt the constancy test in the synchronous detection mode. On the other hand, if the determination result is rejected, the determination result in step 1005 is not likely due to the artifact line, and there is a high possibility that the invariance of the X-ray sensor is impaired. Therefore, the display unit 104 displays a message for prompting contact with the service person (step 1011).

Embodiment 1-5
In the embodiment 1-4, in step 1009 of FIG. 10, the image analysis unit 103 calculates the evaluation index for the preprocessed image even in the automatic detection mode, but for the image subjected to artifact correction. An evaluation index may be calculated. The correction method uses the method shown in the embodiment 1-1. After performing the artifact correction, the same steps as in the embodiment 1-4 are performed. Even if the artifact correction is performed, the correction is insufficient, and therefore, the determination result of the invariance test may be rejected due to the influence of the artifact line. In the same manner as described above, determination is made at step 1009 and step 1010.

Embodiment 1-6
As an embodiment 1-6, an embodiment in which image processing using a peripheral image such as grid suppression processing is performed after pre-processing is performed using the configuration diagram of FIG. 1 and the entire flow of FIG. I will explain that. FIG. 11 is a flowchart of processing of the X-ray image processing apparatus according to the present embodiment. The constancy tests in the above-described Embodiments 1-1 to 1-5 are performed in a state where there is no subject and the grid is removed. However, in the case of a unit in which the grid and the X-ray sensor are integrated, it is difficult to remove the grid. In this case, the X-ray image processing apparatus performs imaging with the grid attached, calculates the evaluation index after applying grid stripe suppression processing. Since the processing from step 1101 to 1102 is the same as the processing from step 301 to 302 in the embodiment 1-1, description thereof is omitted.

  The image acquisition unit 101 performs grid stripe suppression processing on the preprocessed image (step 1103). The image acquisition unit 101 performs, for example, a method described in Patent Document 3 as grid stripe suppression processing. The image acquisition unit 101 extracts approximate grid stripe components from the preprocessed image, generates a grid stripe image from the grid stripe components, and suppresses the grid stripes by removing the grid stripe image from the preprocessed image. . Since the processing from Step 1104 to Step 1105 is the same as the processing from Step 303 to Step 304 in Embodiment 1-1, the description thereof is omitted.

  When the detection mode acquired in step 1104 is the automatic detection mode, the image analysis unit 103 confirms from the grid-suppressed image whether the grid stripe suppression process has failed due to the artifact line (in step 1105). Yes, step 1106). When the grid stripe suppression process fails due to an artifact line, a new artifact is also generated in a normal region near the artifact line. Therefore, the image analysis unit 103 sets a plurality of ROIs in the image, and compares the average value variation for each ROI before and after the grid stripe suppression process.

  Before grid stripe suppression processing, the grid is uniformly included in all ROIs, so the average value variation for each ROI is small. Even when the processing is successful after the grid stripe suppression processing, since the grid is removed in all ROIs, the average value variation for each ROI is small. On the other hand, if the processing fails after grid stripe suppression processing, artifacts are newly generated in the ROI including the artifact line, so the average value differs greatly from other ROIs, and the average value for each ROI The variation of the is increased. The image analysis unit 103 can determine whether the grid stripe suppression process has failed by taking a ratio before and after the grid stripe suppression process and comparing the ratio with a threshold value. If the grid stripe suppression process has failed, the display unit 104 stops the constancy test in the automatic detection mode and displays a message on the display device to prompt the constancy test in the synchronous detection mode (in step 1107). Yes, step 1111). The processing from steps 1108 to 1110 is the same as the processing from steps 306 to 308 of the embodiment 1-1, and thus the description thereof is omitted.

  In this embodiment, grid stripe suppression processing is taken up as image processing before calculating the evaluation index. However, the present embodiment is applied to processing that may cause image processing to fail due to artifact lines, which may be other image processing. Is possible.

  As described above, according to the first embodiment, X-ray irradiation detection is performed in the automatic detection mode, and even if an artifact has occurred in the preprocessed image (invariance test image), the image after the influence of the artifact is suppressed. To evaluate the quality. This makes it possible to perform a quality control test on the X-ray image apparatus with high reliability.

[Embodiment 2]
FIG. 12 shows a configuration diagram of the X-ray image processing apparatus according to the present embodiment. The X-ray image processing apparatus according to this embodiment includes an image acquisition unit 1201, an imaging condition determination unit 1202, a quality management unit 1203, and a display unit 1204. The image acquisition unit 1201 is the same as the image acquisition unit 101 of FIG. 1 described in the first embodiment, and a description thereof is omitted. The shooting condition determination unit 1202 receives the preprocessed image output from the image acquisition unit 101, analyzes the preprocessed image, determines whether the shooting condition is appropriate, and outputs a determination result. The quality management unit 1203 receives the preprocessed image output from the image acquisition unit 101, analyzes the preprocessed image, calculates an evaluation index of a quality control test, and outputs an evaluation result. The display unit 1204 receives the evaluation result and the determination result from the imaging condition determination unit 1202 and the quality management unit 1203, and outputs the evaluation result and the determination result to a display device such as a monitor.

  A configuration example when the configuration of FIG. 12 is realized using a PC is the same as that of FIG. 2 described in the first embodiment. The image processing unit 1201, the shooting condition determination unit 1202, and the quality management unit 1203 illustrated in FIG. 12 are stored in the storage unit 208 as software modules. Hereinafter, the processing details of the X-ray image processing apparatus according to the present embodiment will be described along the four embodiments 2-1 to 2-4. The premise of the processing according to the present embodiment is the same as that of the first embodiment.

[Embodiment 2-1]
The embodiment 2-1 will be described along the flow of processing using the configuration diagram of FIG. 12 and the overall flow of FIG. FIG. 13 is a flowchart of processing of the X-ray image processing apparatus according to the present embodiment. The image acquisition unit 1201 acquires an image with an X-ray sensor and performs preprocessing (steps 1301 and 1302). The preprocessing is processing for correcting sensor characteristics such as offset correction and gain correction. The image acquisition unit 1201 brings the acquired image into a state in which the correlation with the peripheral pixels is maintained by such processing.

  Next, the imaging condition determination unit 1202 calculates the amount of artifact in the automatic detection mode from the preprocessed image (step 1303). In the present embodiment, as a method for automatically detecting X-rays, as in the first embodiment, since the detection is performed for each scanning line as in Patent Documents 1 and 2, the artifact is a region as shown in FIG. Occurs. In order to calculate the amount of artifact, the imaging condition determination unit 1202 first obtains the value of the artifact line PV using Equation 1 by the same method as described in the first embodiment.

  The constancy test in this embodiment is performed in the absence of a subject. Therefore, the sensor output is substantially uniform. Since the dark current component is also substantially uniform, it can be considered by reducing to a linear expression. Therefore, the projection data (PV (y)) is fitted with the primary expression F (y). Then, the linear position is estimated as the start line by taking the average of the lines where no artifact has occurred. Thereby, an artifact area can be extracted.

なお、（式６）において、ストップラインにおける真値（右辺第２項の分母）はストップラインの隣接行から補間したものとしている。 Next, the imaging condition determination unit 1202 uses, as an index representing the amount of artifact, a true value of a pixel value (a pixel value when operating normally without loss) on a stop line (y _stop ) where the artifact occurs at the maximum. An artifact generation rate expressed as a ratio to the actual pixel value is obtained. For example, the artifact occurrence rate S in the case of driving by interlace is shown in Equation 6. w is the number of pixels in the line direction.
(Formula 6)

In (Expression 6), the true value (the denominator of the second term on the right side) in the stop line is assumed to be interpolated from the adjacent line of the stop line.

  In the present embodiment, it is assumed that the artifact occurrence rate Smax that is the correction limit of the artifact correction processing in step 1305 is calculated in advance. The imaging condition determination unit 1202 compares whether or not the occurrence rate S obtained by Equation 6 exceeds the correction limit occurrence rate Smax (step 1304). When the occurrence rate S exceeds the correction limit occurrence rate Smax, the artifact cannot be corrected by the artifact correction process, so the quality management unit 1203 does not perform quality control evaluation, and the display unit 1204 displays a message prompting re-photographing. . The display unit 1204 displays on the monitor so as to set the imaging condition so that the dose is higher than that in the current imaging in order to suppress the occurrence rate of the artifact (step 1308).

  On the other hand, when the occurrence rate S is smaller than the correction limit occurrence rate Smax in step 1304, the quality control evaluation unit 103 performs the artifact correction process (step 1304) because the influence of the artifact can be suppressed by the artifact correction process. Yes, step 1305). Then, the quality management evaluation unit 103 performs quality management evaluation on the artifact corrected image (step 1306). For example, in a constancy test as a quality control method, a plurality of ROIs are set in an image, and an average value and a variance value are calculated for each ROI. The quality control evaluation unit 103 can confirm the invariance of the output value by checking that the average value and the variance value are within a predetermined threshold range in each ROI. The display unit 1204 displays the quality management evaluation result for the corrected image (step 1307).

[Embodiment 2-2]
As the embodiment 2-2, an example different from the embodiment 2-1 in the method for determining the photographing condition will be described. The process flow will be described with reference to the configuration diagram of FIG. 12 and the entire flow of FIG. FIG. 14 is a flowchart of processing of the X-ray image processing apparatus according to the present embodiment. The processing in steps 1401 to 1402 corresponds to the processing in steps 1301 to 1302 in the embodiment 2-1, and thus description thereof is omitted.

ここで、V’は画素値の真値、V_Dは残存している暗電流成分、Rは電荷解放時間に比例した値を表す。なお、V’は、インタレースで駆動している場合、隣接行の画素値から補間することによって求められる。また、Rは実測値から求めることができる。X線センサ２０２はV_Aを走査線と垂直方向に積分した値が検知判定閾値TH_Xdを超えるとX線センサ２０２はX線が照射されていると判断し、照射を検知をする。すなわち、X線センサ２０２は、式８が真の場合に、照射を検知する。
（式８）

The shooting condition determination unit 102 calculates an average value of a specific area from the preprocessed image (step 1403). The area for calculating the average value may be the entire image range or a predetermined ROI. The average value is an index representing the amount of artifacts due to automatic detection. Since the amount of artifact is the value V _A lost due to the charge release, it can be represented by a value R proportional to the charge release time.
(Formula 7)

Here, V ′ is the true value of the pixel value, V _D is the remaining dark current component, and R is a value proportional to the charge release time. Note that V ′ is obtained by interpolating from pixel values in adjacent rows when driving in an interlaced manner. R can be obtained from an actual measurement value. When the value obtained by integrating _VA in the direction perpendicular to the scanning line exceeds the detection determination threshold TH _Xd , the X-ray sensor 202 determines that X-rays are being irradiated and detects irradiation. That is, the X-ray sensor 202 detects irradiation when Expression 8 is true.
(Formula 8)

When the time required for detection of irradiation increases, the number of scanning lines to be integrated increases, and the left side of Equation 8 becomes larger and easier to detect, but the value of R, which is proportional to time, also increases. As the value of R increases, the amount of artifacts increases. The line position when the amount of the correction limit artifact that can be corrected by the artifact correction processing in step 1405 is Rmax is defined as yRmax. At this time, the dose detected at the line position yRmax is the minimum dose to be irradiated. That is, the minimum dose to be irradiated is expressed as in Equation 9.
(Formula 9)

  When a constancy test is performed as a quality control test, the pixel values of the entire image are substantially uniform. Therefore, if Equation 9 is divided by the number of lines from the estimated start line ystart to the yRmax line and the average value in the line is calculated, the pixel value PxVmax for the lowest dose to be irradiated is obtained. Therefore, the imaging condition determination unit 102 calculates an average value of pixels in a specific region, and checks whether or not the value is equal to or greater than PxVmax, so that the dose is sufficient for correcting artifacts due to automatic detection. Can be determined. In the right side of Equation 9, since the second term is a sufficiently small value compared to the first term, the photographing condition determination unit 102 may calculate only the first term. The imaging condition determination unit 102 compares the average value of the preprocessed images with the threshold value PxVmax (step 1404).

  When the average value exceeds the threshold value PxVmax, it is possible to correct artifacts due to automatic detection occurring in the preprocessed image. At that time, the quality management evaluation unit 1203 performs artifact correction (step 1405). The quality management evaluation unit 1203 may perform the artifact correction by the method described in the first embodiment. Then, the quality management evaluation unit 1203 performs quality management evaluation on the artifact corrected image (step 1406). The quality control evaluation unit 1203 sets a plurality of ROIs in an image and calculates an average value and a variance value for each ROI when performing an invariance test as a quality control evaluation. The display unit 1204 displays the result of the quality control evaluation test on a display device such as a monitor (step 1407).

  On the other hand, when the threshold value PxVmax is not exceeded, artifacts caused by automatic detection occurring in the preprocessed image cannot be corrected. At that time, the display unit 1204 displays a message on a display device such as a monitor so as to increase the dose and perform imaging again (step 1408). At that time, the display unit 1204 may indicate to the user how much is insufficient by displaying the threshold value PxVmax and the average value together.

  In addition, if an artifact line exists in the set ROI, the average value is greatly different from the true value. Therefore, it is necessary to calculate the average value after removing the artifact line. The region where the artifact occurs can be calculated by the method described above.

  In the present embodiment, the imaging condition is a condition for setting an artifact amount within a range in which artifacts generated by automatic detection can be corrected. However, even if it cannot be completely corrected, there is no problem as long as the evaluation index is not affected. It is good also as an imaging condition which made the artifact amount the target.

[Embodiment 2-3]
As Embodiment 2-3, another method for calculating the threshold value PxVmax in Embodiment 2-2 will be described. The amount of the correction limit artifact that can be corrected by the artifact correction processing used in step 1405 in FIG. 14 is not the attenuation rate with respect to the true value of the pixel value, but the width (number of lines of the artifact line). For example, in the artifact correction processing in interlaced driving, when correction is performed using values of adjacent rows, it is assumed that the adjacent rows are normal. Therefore, if the width of the artifact exceeds half of the number of scanning lines, the relationship is not satisfied and correction cannot be performed. Therefore, if the maximum allowable width regarding the width of the artifact is Wmax, the dose detected at this time is the lowest dose to be irradiated. In Equation 9, the threshold value PxVmax can be obtained in the same manner as in the embodiment 2-2 by calculating by replacing the line position yRmax with Wmax.

[Embodiment 2-4]
In Embodiment 2-1, the target of the quality control test is an image acquired in the absence of a subject such as an invariance test, but this embodiment shows an embodiment in the state of having a subject. Description will be made along the flow of processing using the configuration diagram of FIG. 15 and the entire flow of FIG. FIG. 16 is a flowchart of processing of the X-ray image processing apparatus according to the present embodiment. The processing in steps 1601 to 1602 corresponds to the processing in steps 1301 to 1302 in the embodiment 2-1, and thus description thereof is omitted.

  The imaging condition determination unit 1502 calculates the amount of artifact in the automatic detection mode from the preprocessed image (step 1603). The amount of artifact represents the attenuation from the true value in the stop line. The area where the artifact is generated is calculated by the method described above. However, when there is a subject, even if projection is performed in the scanning line direction (x direction) as shown in FIG. Therefore, when there is a subject, the calculation is performed by projecting in a region having a specific similar pixel value, and the average of the results of each region is calculated to calculate the parameter of the linear expression F (y).

  The imaging condition determination unit 1502 compares whether or not the occurrence rate S of Expression 6 exceeds the correction limit occurrence rate Smax (step 1604). The correction limit occurrence rate Smax is an artifact occurrence rate Smax that is a correction limit of the artifact correction processing, and is calculated in advance. This value is prepared separately from the threshold value when there is no subject such as the invariance test of the embodiment 2-1. Further, from the viewpoint of preventing diagnostic misdiagnosis, the threshold value is made stricter than when there is no subject. If the occurrence rate S exceeds the correction limit occurrence rate Smax, the artifacts will not disappear in the artifact correction process, so the QA process (image quality enhancement process) is not performed and the user is informed that the artifact has occurred. . The display unit 1504 displays a possibility that an artifact has occurred on the display device. Then, a message for urging to take a picture with a higher dose is also displayed (step 1608).

  On the other hand, if the occurrence rate S is smaller than the correction limit occurrence rate Smax in step 1604, the image processing unit 1503 performs the artifact correction process because the influence of the artifact can be suppressed by the artifact correction process (step 1605). The image processing unit 1503 may perform the artifact correction by the method described in the first embodiment. The image processing unit 1503 performs QA processing (image quality enhancement processing) on the artifact-corrected image and outputs an image suitable for the diagnostic quality (step 1606). The QA processing includes, for example, frequency processing such as noise suppression processing and enhancement processing, and gradation processing for creating gradation suitable for diagnosis.

  The display unit 1504 displays the QA processing result for the corrected image (step 1607). When the occurrence rate S is larger than the correction limit occurrence rate Smax in step 1604, neither the artifact correction process nor the QA process is performed, but both the artifact correction process and the QA process are performed, and the image is displayed on the display device as shown in FIG. You may display. By displaying an image, the user can easily determine whether the level is satisfactory for diagnosis or whether re-imaging should be performed.

  As described above, according to the second embodiment, when the X-ray irradiation detection is performed in the automatic detection mode and an artifact is generated in the preprocessed image (invariance test image), the amount of the artifact is controlled according to the imaging condition. To do. Thereby, the influence of the artifact is reduced, and a test for quality control can be performed with high reliability on the X-ray imaging apparatus.

[Other Embodiments]
The present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in a computer of the system or apparatus read and execute the program This process can be realized. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

101: Image acquisition unit, 102: Switching unit, 103: Image analysis unit, 104: Image analysis unit, 105: Display unit, 106: Tone processing unit, 410: Artifact area, 411: Projection data, 412: Start line estimation 420: Artifact region, 421: Projection data, 422: Start line estimation, 801: ROI arrangement and automatic detection artifact, 802: Automatic detection artifact in ROI, 803: Region from which artifact is removed, 1201: Image acquisition unit, 1202 : Shooting condition section 1203: Quality management section 1204: Display section 1501: Image acquisition section 1502: Shooting condition section 1503: Image processing section 1504: Display section

Claims (16)

An acquisition means for acquiring an image generated by a sensor that detects X-rays emitted from a radiation source;
Generating means for generating an evaluation value for evaluating the quality of the image based on the image acquired by the acquiring means;
A determination unit that determines a result of evaluation on the quality of the image based on the evaluation value generated by the generation unit;
Display means for displaying the result of the evaluation determined by the determination means;
Have
The X-ray image is characterized in that each of the generation unit or the determination unit performs different processing in response to a plurality of different detection modes for detecting the start of the X-ray irradiation set in the sensor. Processing equipment.   2. The X-ray image according to claim 1, wherein the plurality of different detection modes include a first detection mode in which imaging is performed by detecting an X-ray irradiation start based on a signal from the sensor. Processing equipment.   3. When the sensor is set to the first detection mode, the generation unit corrects an artifact that appears in the image, and generates the evaluation value based on the corrected image. The X-ray image processing apparatus described in 1.   When the sensor is set to the first detection mode, the sensor further includes image processing means for performing predetermined image processing on the image acquired by the acquisition means, and the generation means fails in the image processing. The X-ray image processing apparatus according to claim 2, wherein the evaluation value is generated based on an image on which the image processing has been performed when the image processing is not performed.   The X-ray image processing apparatus according to claim 4, wherein the predetermined image processing is grid stripe suppression processing.   6. The X-ray according to claim 3, wherein the determination unit determines a result of the evaluation by comparing a threshold value for the first detection mode with the evaluation value. 7. Image processing device. The plurality of different detection modes include a second detection mode in which imaging is started at a timing acquired by the sensor from the radiation source,
The X-ray image processing apparatus according to claim 6, wherein the determination unit calculates a threshold value for the first detection mode from a threshold value for the second detection mode.   When the sensor is set to the first detection mode, the generation unit generates the evaluation value from a region where the artifact is removed when the content rate of the artifact appearing in the image is smaller than a predetermined threshold. The X-ray image processing apparatus according to claim 2.   9. The method according to claim 8, wherein when the content rate of the artifact appearing in the image is larger than a predetermined threshold, the generation unit does not generate the evaluation value, and the display unit displays a message prompting re-photographing. The X-ray image processing apparatus described.   When the sensor is set to the first detection mode, the generation unit replaces the area of the artifact appearing in the image with an image obtained by the previous shooting, and based on the image after the replacement The X-ray image processing apparatus according to claim 2, wherein an evaluation value is generated.   When artifacts appear in the replaced image, the display means displays a message prompting re-photographing, and the generating means replaces the area of the artifact appearing in the image using the image obtained by the re-photographing. The X-ray image processing apparatus according to claim 10, wherein the evaluation value is generated based on the image after the replacement. The plurality of different detection modes include a second detection mode in which imaging is started at a timing acquired by the sensor from the radiation source,
12. The X-ray according to claim 8, wherein the determination unit determines a result of the evaluation by comparing the evaluation value with a threshold value for the second detection mode. 13. Image processing device. When the sensor is set to the first detection mode, and the result of the evaluation is determined to be unacceptable by the determination unit,
The generation unit generates a second evaluation value from a region from which artifacts appearing in the image are removed, and the determination unit determines the result of the evaluation based on the second evaluation value. The X-ray image processing apparatus according to claim 2. When the sensor is set to the first detection mode, and the result of the evaluation is determined to be unacceptable by the determination unit,
The generation unit corrects artifacts appearing in the image, generates a second evaluation value based on the corrected image, and the determination unit determines the evaluation result based on the second evaluation value. The X-ray image processing apparatus according to claim 2. The plurality of different detection modes include a second detection mode in which imaging is started at a timing acquired by the sensor from the radiation source,
The X-ray image processing apparatus according to claim 13, wherein the determination unit determines the evaluation result by comparing the evaluation value with a threshold for the second detection mode. An acquisition step of acquiring an image generated by a sensor that detects X-rays emitted from a radiation source;
A generating step for generating an evaluation value for evaluating the quality of the image based on the image acquired in the acquiring step;
A determination step of determining a result of evaluation for the quality of the image based on the evaluation value generated in the generation step;
A display step for displaying the result of the evaluation determined in the determination step;
Have
The X-ray image is characterized in that each of the generation step or the determination step performs different processing in response to a plurality of different detection modes for detecting the start of the X-ray irradiation set in the sensor. Processing method.

Images