JP2017164205A - Image processing system, image processing method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make determination made by an operator about an image after gray-scale conversion easy.SOLUTION: A statistical amount of a pixel value within an ROI of a predetermined shape and size around a target pixel of image data before gray-scale conversion is a reference value in the target pixel. Derivation of such a reference value is performed with each target image. The reference value in each target pixel is compared with a reference value calculated by an automatic analysis, and a position of the target pixel having a pixel value corresponding to the reference value calculated in the automatic analysis is identified. The identified position is displayed over an image after the gray-scale conversion as an ROI position corresponding to a pixel having the reference value obtained in the automatic analysis.SELECTED DRAWING: Figure 4

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program, and is particularly suitable for use in performing gradation conversion of an image.

  Conventionally, a technique for visualizing the inside of a subject by irradiating the subject with radiation represented by X-rays and imaging a radiation component transmitted through the subject has been used.

  By the way, when outputting image data obtained by radiography to a monitor, an X-ray diagnostic film or the like, gradation conversion is performed on the image data so that the output image can be easily observed. Is called.

  In the gradation conversion, for example, a pixel value representing a region of interest (referred to as ROI (Region Of Interest)) of the subject included in the image data is derived as a reference value. A desired output image is obtained by associating the pixel value of the image data with the gradation conversion curve so that the reference value becomes a predetermined output value in the gradation conversion curve indicating the input / output relationship of the pixel value. .

  In such a method of automatically analyzing the reference value (hereinafter referred to as automatic analysis), gradation conversion may not be optimal if the analysis fails. Therefore, Patent Document 1 discloses a method in which after an image obtained by automatic analysis is displayed, an operator designates an ROI, and automatic analysis is performed again based on the designated ROI.

  Patent Document 1 discloses that when an operator designates an ROI, an image of the same part of the same subject and an ROI corresponding to the image are extracted and displayed from image data captured in the past. In addition, it is disclosed that in the absence of such information, a typical schematic diagram (actual image or design image) of the same part of the subject and an ROI for the schematic diagram are displayed.

JP 2004-88322 A Japanese Patent No. 4208333 JP 2004-173077 A

  However, in the method described in Patent Document 1, if the operator does not look at the result of gradation conversion again by specifying the displayed ROI, when the ROI is specified, appropriate gradation conversion is performed. It cannot be determined whether or not it is performed. For this reason, the operator cannot easily determine whether the result of automatic analysis, that is, the result of gradation conversion is appropriate.

  The present invention has been made in view of such problems, and it is an object of the present invention to facilitate an operator's judgment on the result of gradation conversion of an image.

  An image processing apparatus according to the present invention is an image processing apparatus that performs image processing on an image obtained by photographing a subject, and performs gradation conversion of the image based on a pixel value serving as a reference of the image. And a display control unit for displaying information indicating a position corresponding to the reference pixel value on the image on which the tone conversion has been performed by the tone conversion unit. To do.

  According to the present invention, it is possible to facilitate determination performed by an operator on an image after gradation conversion.

It is a figure which shows the 1st example of a structure of a radiography apparatus. It is a flowchart explaining the 1st example of operation | movement of an image process part. It is a figure which shows a gradation conversion curve. It is a figure explaining the display method of the position of ROI. It is a figure which shows the 2nd example of a structure of a radiography apparatus. It is a flowchart explaining the 2nd example of operation | movement of an image process part.

Hereinafter, embodiments will be described with reference to the drawings.
(First embodiment)
First, the first embodiment will be described.
FIG. 1 is a diagram illustrating an example of the configuration of the radiation imaging apparatus 100. The radiation imaging apparatus 100 has an image processing function for performing gradation conversion on the image data based on a reference value obtained from image data obtained by radiation imaging. The reference value is a pixel value used as a reference when gradation conversion is performed. The radiation imaging apparatus 100 includes a radiation generation unit 101, a radiation detector 104, a data collection unit 105, a preprocessing unit 106, a CPU 108, a main memory 109, an operation unit 110, a display unit 111, and an image processing unit 112. These units are connected to each other through the CPU bus 107 so that data can be exchanged between them.

  The image processing unit 112 performs image processing including gradation conversion on the image captured by the radiation detector 104. The image processing unit 112 includes an automatic analysis unit 113, a gradation conversion unit 114, an ROI shape acquisition unit 115, an ROI position specification unit 116, an ROI display control unit 117, and a position designation unit 118.

  The main memory 109 stores various data necessary for processing by the CPU 108 and functions as a working memory for the CPU 108. The CPU 108 uses the main memory 109 to perform operation control of the entire apparatus according to the operation from the operation unit 110. As a result, the radiation imaging apparatus 100 operates as follows.

  First, when an imaging instruction is input from the operator via the operation unit 110, the imaging instruction is transmitted to the data collection unit 105 by the CPU 108. Further, when receiving an imaging instruction, the CPU 108 controls the radiation generation unit 101 and the radiation detector 104 to execute radiation imaging.

  In radiography, first, the radiation generation unit 101 irradiates the subject 103 with the radiation beam 102. The radiation beam 102 emitted from the radiation generation unit 101 passes through the subject 103 while being attenuated, and reaches the radiation detector 104. The radiation detector 104 outputs a signal corresponding to the intensity of the arrived radiation. In this embodiment, the subject 103 is a human body. Therefore, the signal output from the radiation detector 104 is data obtained by photographing the human body.

  The data collection unit 105 converts the signal output from the radiation detector 104 into a predetermined digital signal and supplies it to the preprocessing unit 106 as image data. The preprocessing unit 106 performs preprocessing including offset correction and gain correction on the image data supplied from the data collection unit 105.

  Even if the radiation is not irradiated, the signal level output from the radiation detector 104 is not completely zero due to the dark current inherent in each photodiode arranged corresponding to each pixel of the radiation detector 104. Has an offset component. By performing the offset correction by the preprocessing unit 106, the signal level output when the dose is zero can be made uniform for all pixels. Specifically, the preprocessing unit 106 performs correction by subtracting offset image data acquired in a state where no radiation is irradiated from captured image data. Each pixel of the radiation detector 104 has a slightly different gain (sensitivity). The pre-processing unit 106 corrects the gain variation of each pixel of the radiation detector 104, and makes the signal level output with respect to the dose uniform for all pixels. Specifically, the preprocessing unit 106 irradiates the radiation detector 104 with substantially uniform radiation, regards this radiation distribution as uniform, and creates a gain image representing the gain difference of each pixel. Then, the gain image is corrected by subtracting it from the offset corrected image data.

  The image data preprocessed by the preprocessing unit 106 is sequentially transferred to the main memory 109 and the image processing unit 112 via the CPU bus 107 under the control of the CPU 108. In the present embodiment, radiation imaging is performed as described above.

  The image processing unit 112 performs irradiation field recognition on the image data. An irradiation field edge is extracted from a plurality of edge components existing in the preprocessed image data using a restriction condition such as its position and intensity. The image processing unit 112 recognizes an area surrounded by the irradiation field edge as an irradiation field area. The image processing unit 112 analyzes the image data on which preprocessing and irradiation field recognition have been performed. The image processing unit 112 calculates a reference value that is a representative pixel value of the ROI from the analysis result by the automatic analysis unit 113. For example, in the case of chest imaging, the image processing unit 112 sets the ROI so as to straddle the ribs of the chest, and calculates a reference value that is a representative pixel value of the ROI. Then, the image processing unit 112 converts the pixel value of the preprocessed image data into the gradation value so that the reference value becomes a predetermined output value in the gradation conversion curve indicating the input / output relationship of the pixel value. By associating with the conversion curve, the gradation value (pixel value) of the output image data is obtained.

  The image after the gradation conversion is performed in this way is displayed on the display unit 111. After confirming this image, the operator re-specifies the ROI using the operation unit 110 as necessary, and instructs re-execution of the gradation conversion described above. When the operator determines that an appropriate image has been displayed, the operator operates the operation unit 110 to instruct the storage of the image, the output to a printer (not shown), or the like. Based on this instruction, an image is stored in a storage medium inside or outside the radiation imaging apparatus 100, and an image is output from a printer or the like. As described above, a series of photographing operations is completed.

An example of the operation of the image processing unit 112 of the radiation imaging apparatus 100 having the above configuration will be specifically described with reference to the flowchart of FIG.
As described above, the image data preprocessed by the preprocessing unit 106 is transferred to the image processing unit 112 via the CPU bus 107. Then, first, the automatic analysis unit 113 calculates a reference value (s201). As described above, in this embodiment, based on the pixel value representing the region of interest (ROI) of the image or the histogram of the pixel value of the image, the pixel value serving as the reference of the image is derived as the reference value. Here, the calculation method of the reference value is not particularly limited. For example, for chest imaging, for example, the method described in Patent Document 2 can be used. Moreover, if it is abdominal radiography, the method of patent document 3 can be used, for example. Further, the reference value may be calculated from the pixel value representing the ROI, or the reference value may be calculated from the histogram of the image (pixel value thereof). In this way, a known automatic analysis corresponding to the imaging region may be used. Note that the automatic analysis unit 113 obtains information input in advance by the operator from the main memory 109 using the operation unit 110 when performing an imaging instruction, and performs automatic analysis according to the imaging region (calculation of a reference value). (Method) may be switched.

  Next, the gradation conversion unit 114 performs gradation conversion on the preprocessed image data based on the reference value obtained by the automatic analysis unit 113, and the image after gradation conversion is displayed on the display unit 111. Display (s202). As described above, in the present embodiment, tone conversion of the image is performed based on the pixel value (reference value) serving as the reference of the image. Here, an example of a specific method of gradation conversion will be described. FIG. 3 is a diagram illustrating an example of the gradation conversion curve 301. In the example shown in FIG. 3, the shape of the function used for gradation conversion is exemplified by a case where an S-shaped curve similar to the characteristic curve of the silver salt film is used as the gradation conversion curve 301. In FIG. 3, the gradation conversion unit 114 has each pixel value (input pixel value (gradation value)) of the preprocessed image so that the reference value calculated by the automatic analysis unit 113 becomes a predetermined value. To obtain an output pixel value. Note that the gradation conversion curve used for gradation conversion is not limited to the S-shaped curve shown in FIG. 3, and may be a straight line, for example. In addition, a value set in advance by the operator using the operation unit 110 as the predetermined value may be acquired from the main memory 109 and used.

  Next, the ROI shape acquisition unit 115 acquires the size and shape of the ROI (s203). Thus, in this embodiment, information on the shape and size of the region including the target pixel of the image is acquired. Here, the ROI shape and size may be set in advance by the operator using the operation unit 110. As a method for setting the shape and size of the ROI, for example, a plurality of types of templates are prepared in advance, and the operator can select from among the plurality of types of templates using the operation unit 110. In addition, the operator may freely set the shape and size of the ROI using the operation unit 110. Alternatively, a predetermined fixed value may be used as the ROI shape and size without the operator setting the ROI shape and size. In the present embodiment, the following description will be given by taking as an example a case where a square whose side is X [cm] is set as the shape and size of the ROI.

Next, the ROI position specifying unit 116 specifies the position of the ROI in which the statistic of the pixel value substantially matches the reference value calculated by the automatic analysis unit 113 from the preprocessed image data (s204). . A specific example of the method for specifying the position of the ROI will be described below.
First, the ROI position specifying unit 116 has a region (with a shape and a size acquired by the ROI shape acquisition unit 115 centered on a target pixel of preprocessed image data (that is, image data before gradation conversion). An average value of pixel values in a square having a side length of X [cm] is calculated. The region becomes the ROI, the position of the target pixel becomes the position of the ROI, and the average value becomes the value of the ROI. The ROI position specifying unit 116 performs the calculation of the average value of such pixel values using all the pixels of the preprocessed image data as target pixels. However, all the pixels of the image data need not be the target pixels. For example, a pixel in a region unrelated to the subject need not be a target pixel.
The ROI position specifying unit 116 calculates the reference value for each target pixel of the image data that has been preprocessed as described above.

  For example, assuming that the pixel value of i row j column of the image before gradation conversion is I (i, j), the ROI position specifying unit 116 uses the reference value R (i, j) at the position (i, j) of the target pixel. ) Is calculated as in the following equation (1).

Here, p is the size [cm] of the target pixel of the image.
Here, in Formula (1), the case where an average value is used as a reference value is shown as an example. However, the reference value is not limited to an average value as long as it is a statistic. For example, an order statistic value such as a mode value robust to an abnormal value or a median value may be used. Moreover, you may combine at least 2 of these statistics. As described above, in this embodiment, the statistic of the pixel value in the region including the target pixel of the image is derived using the plurality of pixels of the image as the target pixel. Note that the reference value R (i, j) obtained by Expression (1) is equivalent to the reference value calculated when the operator specifies the position of the target pixel in the position specifying unit 118. Therefore, the reference value R (i, j) in each target pixel obtained by Expression (1) represents how the reference value changes when each target pixel is designated.

Next, the ROI position specifying unit 116 calculates an error between the reference value obtained by the automatic analysis unit 113 and the reference value R (i, j) obtained by Expression (1). To describe a specific example, ROI position specifying unit 116 a reference value determined by the automatic analyzer 113 and R _a, the position of the target pixel (i, j) error definitive E (i, j) following the Calculation is performed as shown in Equation (2).

Here, in equation (2), and the reference value R _a determined by automatic analysis unit 113, the reference value R (i, j) obtained by the equation (1) error E (i, j) and is subject It is also calculated including the distribution around the pixels (error variation in the range of ± D). Indicating that the error E (i, j) as the value is small, the area around the target pixel, an area suitable for deriving a reference value R _a equivalent value determined by the automatic analyzer 113 . Note that w (i, j) in equation (2) is an arbitrary weighting function. For example, a Gaussian function that gives a large weight to the center (target pixel position) may be used as the weighting function w (i, j).

Then, ROI position specifying unit 116, the error E (i, j), to identify the position of the reference value R _a equivalent reference value determined by automatic analysis unit 113 is obtained (i, j). As described above, in the present embodiment, the position corresponding to the reference pixel value (reference value) is determined based on the statistic in each of the regions including the target pixel of the image and the reference pixel value (reference value). Identify. As described above, as the value of the error E (i, j) is small, the reference value R (i, j) obtained by the equation (1), the reference value R _a equivalent standards determined by automatic analysis unit 113 It is likely that the value. Therefore, ROI position specifying section 116, error E (i, j) is minimum position (i, j) are as the position of the ROI reference value R _a value obtained substantially coincide in the automatic analyzer 113 It only has to be specified. Note that the position of the specified ROI does not have to be one. For example, the following may be used. First, the ROI position specifying unit 116 selects a plurality of positions (i, j) whose error E (i, j) is smaller than a predetermined value. Then, the ROI position specifying unit 116, among the selected positions (i, j), the distance between each position (i, j) is greater than a predetermined value (plural positions) (i, j). the reference value R _a and value may be specified as the position of the ROI substantially coincident. Thus, it is only necessary to specify the position (i, j) where the error E (i, j) satisfies the predetermined condition as the position of the ROI. As described above, in the present embodiment, when the difference between the statistic in the region including the target pixel of the image and the reference pixel value (reference value) satisfies a predetermined condition, the position of the target pixel is set as the reference. Is specified as a position corresponding to a pixel value (reference value).

  Next, the ROI display control unit 117 overlay-displays the position of the ROI specified by the ROI position specifying unit 116 on the image after gradation conversion (s205). As described above, in this embodiment, information indicating a position corresponding to a reference pixel value (reference value) is displayed on an image on which gradation conversion has been performed. Here, the display method of the position of the ROI is not particularly limited. FIG. 4 is a diagram illustrating an example of a method for displaying the position of the ROI. For example, as shown in FIG. 4 (a), a circle having the radius D as the center and the ROI position (i, j) specified by the ROI position specifying unit 116 as the center is defined as the ROI. It is displayed overlaid on the image after the key conversion. In addition, when a plurality of positions are specified as the ROI position (i, j) by the ROI position specifying unit 116, a circle having the radius D in Expression (2) is used for each of the plurality of positions. What is necessary is just to superimpose and display on the image after gradation conversion as ROI. In this way, the size and shape of the ROI displayed on the image after gradation conversion can be made different from the size and shape of the ROI acquired by the ROI shape acquisition unit 115. However, the size and shape of these ROIs may be the same. In addition, the figure to be displayed on the image after gradation conversion is not limited to a circle or a square. As described above, in the present embodiment, a predetermined area centered on a position corresponding to a reference pixel value (reference value) is displayed on the image on which gradation conversion has been performed.

  As described above, the position of the ROI displayed superimposed on the image after gradation conversion is a guideline that intuitively indicates where the gradation conversion was performed with reference to the position in the automatic analysis. . As a result, the operator can empirically grasp where the ROI is appropriate. Further, when a position different from the normal position is displayed as the ROI position, the operator can also know that the automatic analysis may have failed. For example, the position of the ROI that is normally displayed in chest radiographing is within the lung field as shown in FIG. 4A, whereas the position that is not within the lung field is displayed as the ROI position as shown in FIG. 4B. In this case, it can be easily determined that the automatic analysis has failed.

  Then, the operator confirms the position of the image after the gradation conversion and the ROI displayed on the display unit 111, determines whether or not the gradation of the image after the gradation conversion is appropriate, and the determination An operation corresponding to the result is performed on the operation unit 110. Based on the result of this operation, the image processing unit 112 determines whether or not the gradation of the image after gradation conversion is appropriate (s206). If the gradation of the image after gradation conversion is appropriate as a result of this determination, the process is confirmed and the process according to the flowchart of FIG. 2 is terminated. At this time, the image processing unit 112 stores the image for which the gradation is determined to be appropriate and the ROI position designated at that time in a storage device inside or outside the radiation imaging apparatus 100. On the other hand, when the gradation of the image after gradation conversion is not appropriate, the operation for specifying the ROI is started. In this case, by pressing the ROI adjustment button in the operation unit 110, the operation for specifying the ROI is started, and the position specifying unit 118 inputs the position of the ROI based on the operation of the operation unit 110 by the operator ( s207). As described above, in this embodiment, information indicating the position corresponding to the reference pixel value (reference value) is displayed on the image subjected to the gradation conversion, and then the position in the image is designated.

  For example, the operator designates the position to be the center point of the ROI on the image after gradation conversion displayed on the display unit 111 with a mouse or the like. In the ROI position designation operation, the same position as that displayed when the automatic analysis is successful may be designated. As described above, since the operator can empirically grasp the position of the ROI when the automatic analysis is successful, in such a case, the operator can easily specify the position of the ROI. It becomes. In addition, since the ROI shape corresponding to the pixel having the reference value obtained by the automatic analysis is converted into the ROI having a predetermined shape, the operator can change the complicated shape depending on the imaging region and the automatic analysis method. There is no need to specify, and the work of specifying the position of the ROI can be easily performed.

  Next, the automatic analysis unit 113 recalculates the reference value at the designated ROI position (i, j) based on the equation (1) (s208). Next, based on the reference value recalculated by the automatic analysis unit 113, the gradation conversion unit 114 performs processing on the preprocessed image data as described with reference to FIG. Tone conversion is performed again (s209). As described above, in the present embodiment, when the designation of the position in the image is received, the reference pixel value is derived again based on the designated position. Then, gradation conversion of the image is performed based on the re-derived reference pixel value. Then, the gradation conversion unit 114 displays the image displayed on the display unit 111 by replacing the image with the gradation conversion again. Then, the operator confirms the image displayed on the display unit 111, determines whether or not the gradation of the image subjected to gradation conversion again is appropriate, and performs an operation according to the result of the determination. 110. Based on the result of this operation, the image processing unit 112 determines whether or not the gradation of the image subjected to gradation conversion again is appropriate (s206). The above processing from s206 to s209 is repeated until it is determined in s206 that the gradation of the image is appropriate. If it is determined that the gradation of the image is appropriate, the process is determined as described above, and the process according to the flowchart of FIG. 2 is terminated.

  In s205, the following processing may be performed. That is, the ROI display control unit 117 determines that the appropriate gradation conversion has been performed in s206, and is the position of the ROI stored in the storage device inside or outside the radiation imaging apparatus 100, and the same subject, The ROI position corresponding to the image of the same imaging region is read out. Then, the ROI display control unit 117 displays the read ROI position in addition to the ROI position specified by the ROI position specifying unit 116 on the image after gradation conversion. In this way, the operator can more easily determine that the position of the ROI specified by the ROI position specifying unit 116 is not appropriate and the automatic analysis has failed. At this time, it is preferable to change the display form of the position of the ROI specified by the ROI position specifying unit 116 and the position of the ROI stored in the storage device (for example, different colors and line types). When a plurality of positions are stored in the storage device as ROI positions corresponding to images of the same subject and the same imaging region, the average value, mode value, or median value of the ROI positions, etc. The order statistic value may be calculated and displayed.

  As described above, in this embodiment, the statistic of the pixel value in the ROI having a predetermined shape and size centered on the target pixel of the image data before gradation conversion is used as the reference value for the target pixel. Such derivation of the reference value is performed for each target image. The reference value of each target pixel is compared with the reference value calculated by automatic analysis, and the position of the target pixel having the pixel value corresponding to the reference value calculated by automatic analysis is specified. Then, the specified position is displayed as an ROI position corresponding to the reference value obtained by the automatic analysis, superimposed on the image after gradation conversion. Therefore, according to the predetermined ROI shape and size, it is possible to match the ROI corresponding to the pixel having the reference value calculated by the automatic analysis and the ROI when the operator manually designates the ROI. . That is, when an image subjected to gradation conversion by automatic analysis is displayed, when the operator manually designates the position of the image as the position of the ROI, can the image subjected to the gradation conversion be obtained? Can be specified by the position of the ROI that is displayed superimposed on the image. Therefore, it is possible to empirically grasp where the image is appropriate as the ROI from the image after the gradation conversion and the position of the ROI. As a result, the operator can determine the result of the tone conversion of the image, such as determining whether the automatic analysis is successful or determining which ROI should be manually specified when the automatic analysis fails. The determination to be made can be easily made by simply looking at the image after gradation conversion. Furthermore, it is possible to reduce the work load when manually specifying the ROI.

  Further, in this embodiment, it is not necessary to register an ROI that assumes every situation in advance, and therefore it is not necessary to prepare an ROI for every automatic analysis. Therefore, for example, the position of the ROI can be easily specified even when the imaging method changes depending on the facility, such as the radiation irradiation range of the target part.

  Moreover, the size and shape of the ROI extracted by automatic analysis are various, and in some cases, the ROI extracted by automatic analysis has a complicated shape. For example, the ROI shape may be a polygonal region surrounding the entire lung field. Therefore, when displaying the result of automatic analysis of image data captured in the past, it is necessary for the operator to specify a complicated shape similar to the ROI obtained by automatic analysis as the ROI, which requires very complicated work. become. On the other hand, in this embodiment, the shape of the ROI corresponding to the pixel having the reference value obtained by the automatic analysis is converted into the ROI having a predetermined shape. The operator only has to specify the position of the displayed ROI when the automatic analysis is successful, and it is not necessary to specify a complicated shape depending on the imaging region or the automatic analysis method, and the operation of specifying the position of the ROI Can be easily performed.

  In the present embodiment, whether the reference value is obtained from the ROI of the image or the reference value is obtained from the histogram of the image (pixel value thereof), the processes other than the calculation of the reference value are the same. That is, even when performing histogram analysis without using ROI in automatic analysis, the position of ROI corresponding to the reference value obtained by automatic analysis can be specified.

(Second Embodiment)
Next, a second embodiment will be described. In the present embodiment, in contrast to the first embodiment, it is further determined whether or not the automatic analysis is successful based on the specified position of the ROI, and further notifies the operator of information according to the determined result. Do. As described above, the present embodiment is obtained by adding a configuration and processing for doing so to the first embodiment. Therefore, in the description of the present embodiment, the same parts as those in the first embodiment are denoted by the same reference numerals as those in FIGS.

FIG. 5 is a diagram illustrating an example of the configuration of the radiation imaging apparatus 500. FIG. 6 is a flowchart for explaining an example of the operation of the image processing unit 501 of the radiation imaging apparatus 500.
The image processing unit 501 of the radiation imaging apparatus 500 includes a determination unit in addition to the automatic analysis unit 113, the gradation conversion unit 114, the ROI shape acquisition unit 115, the ROI position specification unit 116, the ROI display control unit 117, and the position designation unit 118. 502 and a notification unit 503 are provided. As described above, the radiation imaging apparatus 500 of this embodiment has a configuration in which the determination unit 502 and the notification unit 503 are added to the radiation imaging apparatus 100 of the first embodiment. Hereinafter, an example of processing of the image processing unit 501 of the present embodiment will be described with reference to the flowchart of FIG.

  First, in the image processing unit 501, as in the first embodiment, s201 to s205 are executed, and the image after gradation processing is displayed on the display unit 111, and the position of the ROI is displayed on the image as an overlay.

  Next, the determination unit 502 determines whether or not the automatic analysis in the automatic analysis unit 113 has failed based on the position of the ROI specified by the ROI position specifying unit 116 (s601). As described above, in the present embodiment, the reference pixel value (reference value) is evaluated based on the position corresponding to the reference pixel value (reference value). Here, a case will be described as an example in which it is determined whether or not the automatic analysis has failed depending on where in the radiation irradiation range the position of the ROI specified by the ROI position specifying unit 116 is located. Generally, an anatomical region to be diagnosed is arranged and imaged at the center of the radiation irradiation range. In the automatic analysis, there is a tendency to calculate the reference value from the anatomical region to be diagnosed in order to stably display the anatomical region to be diagnosed. Therefore, the position of the ROI specified by the ROI position specifying unit 116 is likely to be near the center of the radiation field range.

  Therefore, the determination unit 502 fails in the automatic analysis when the position of the ROI specified by the ROI position specifying unit 116 is greatly deviated from the center of the radiation irradiation range (when it is away from a predetermined value). It is determined that A specific example will be described. When the ROI position specified by the ROI position specifying unit 116 does not exist within a range of 25 [cm] in diameter from the center position of the radiation irradiation range, the determination unit 502 performs automatic analysis. Judge that it failed. Note that the extent to which these positions are shifted is not limited to such a method as to determine that the automatic analysis has failed. For example, you may make it set suitably according to the photography tendency for every institution, and from the past tendency of the distribution of the position of ROI, the identification boundary for judging failure of automatic analysis and a correct answer is determined by machine learning etc. It may be determined. In addition, the operator may specify in advance a position to be compared with the position of the ROI specified by the ROI position specifying unit 116 by operating the operation unit 110. As described above, in this embodiment, the reference pixel value is determined by determining whether or not the reference pixel value is correctly derived based on the position corresponding to the reference pixel value (reference value). Evaluate the value (reference value).

  Next, when it is determined that the automatic analysis has failed, the notification unit 503 notifies the operator that the automatic analysis may have failed (s602). The notification method is not particularly limited. For example, the notification unit 503 displays a warning message on the display unit 111. As described above, in this embodiment, information indicating the result of evaluation of a reference pixel value (reference value) is notified. Information indicating the result of evaluation of a reference pixel value (reference value) is notified as information indicating that, for example, when it is determined that the reference pixel value (reference value) is not correctly derived. Is done.

  As described above, in the present embodiment, the position of the ROI is displayed on the image after gradation conversion, and if it is determined that the automatic analysis has failed, it indicates that the automatic analysis may have failed. Display a warning message. Therefore, in addition to the effect of the first embodiment, an effect that the operator can reduce the oversight that the automatic analysis has failed is obtained.

  In the present embodiment, a case has been described in which notification that there is a possibility that automatic analysis has failed has been described. However, it is not always necessary to do this as long as the result of the automatic analysis is evaluated and information about the result is output. For example, it may be notified that automatic analysis is highly likely to be successful (automatic analysis is unlikely to have failed). In this case, notification can be performed both when it is determined as yes in s601 and when it is determined as no. Also, in the case of notifying only that there is a high possibility that the automatic analysis has succeeded (that the automatic analysis has a low possibility of failure), in contrast to FIG. 6, when it is determined to be no in s601 If the determination is yes, the notification is not performed.

  In each of the above-described embodiments, the case where image processing is performed on a radiographed image has been described as an example. However, the image to be subjected to image processing is not limited to a radiographed image as long as it is an image of a subject.

  The above-described embodiments are merely examples of implementation in carrying out the present invention, and the technical scope of the present invention should not be construed in a limited manner. That is, the present invention can be implemented in various forms without departing from the technical idea or the main features thereof.

(Other examples)
The present invention is also realized by executing the following processing. That is, first, software (computer program) for realizing the functions of the above embodiments is supplied to a system or apparatus via a network or various storage media. Then, the computer (or CPU, MPU, etc.) of the system or apparatus reads and executes the computer program.

  113: Automatic analysis unit, 114: Tone conversion unit, 115: ROI shape acquisition unit, 116: ROI position specifying unit, 117: ROI display control unit, 118: Position designation unit

Claims (13)

An image processing apparatus that performs image processing on an image obtained by photographing a subject,
Gradation conversion means for performing gradation conversion of the image based on a pixel value serving as a reference of the image;
An image processing apparatus comprising: display control means for displaying information indicating a position corresponding to the reference pixel value on an image on which gradation conversion has been performed by the gradation conversion means. Derivation means for deriving a statistic of a pixel value in a region including the target pixel of the image as a plurality of pixels of the image as the target pixel;
And a specifying unit that specifies a position corresponding to the reference pixel value based on a statistic in each of the regions including the target pixel of the image and the reference pixel value. The image processing apparatus according to claim 1.   When the difference between the statistic in the region including the target pixel of the image and the reference pixel value satisfies a predetermined condition, the specifying unit sets the position of the target pixel to the reference pixel value. The image processing apparatus according to claim 2, wherein the position is specified as a corresponding position.   The image processing apparatus according to claim 2, further comprising an acquisition unit configured to acquire information on a shape and a size of a region including the target pixel of the image.   The display control means displays a predetermined area centered on a position corresponding to the reference pixel value on an image subjected to gradation conversion by the gradation conversion means. The image processing apparatus according to any one of 1 to 4.   The gradation conversion means displays information indicating a position corresponding to the reference pixel value by the display control means on the image subjected to gradation conversion by the gradation conversion means, and then Upon receiving the designation of the position in the image, the reference pixel value is derived again based on the designated position, and the gradation conversion of the image is performed based on the re-derived reference pixel value. The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus.   The image processing apparatus according to claim 1, further comprising an evaluation unit that evaluates the reference pixel value based on a position corresponding to the reference pixel value.   The image processing apparatus according to claim 7, further comprising notification means for notifying information indicating a result of evaluation of the reference pixel value by the evaluation means. The evaluation unit determines whether or not the reference pixel value is correctly derived based on a position corresponding to the reference pixel value.
9. The image processing according to claim 8, wherein when the evaluation unit determines that the reference pixel value is not correctly derived, the notification unit notifies information indicating the fact. apparatus.   The image processing apparatus according to claim 1, wherein the image is an image obtained by radiography.   The said gradation conversion means derives | leads-out the pixel value used as the said reference | standard based on the pixel value representing the region of interest of the said image, or the histogram of the pixel value of the said image. The image processing apparatus according to any one of the above. An image processing method for performing image processing on an image obtained by photographing a subject,
Based on a pixel value serving as a reference of the pixel value of the image, gradation conversion of the image is performed, and information indicating a position corresponding to the pixel value serving as the reference is displayed on the image on which the gradation conversion has been performed. An image processing method characterized in that at least display is performed.   A program that causes a computer to function as each unit of the image processing apparatus according to any one of claims 1 to 11.