X-ray imaging apparatus and image processing apparatus

Abstract

To display an X-ray image in which a region of interest of an operator is converted into an emphasized pixel value which is easily visually recognized by the operator.SOLUTION: In fluoroscopy, a region of interest included in an X-ray image is extracted, and a pixel value conversion parameter for converting pixel values so as to emphasize the distribution of the pixel values of pixels in the region of interest is calculated. The pixel value of the X-ray image generated after the time point when the pixel value conversion parameter is calculated is converted by using the calculated pixel value conversion parameter, and the converted image is displayed.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an X-ray imaging apparatus that irradiates an object with X-rays to obtain an image. [Background technology]

[0002] In X-ray images, the image brightness and contrast of the region of interest (the part, place, or object to be examined in the fluoroscopic image) of the operator within the subject (image of the subject) are not constant due to the subject's body shape and other external factors. Therefore, if the contrast of the region of interest is not appropriate, the operator can adjust the contrast by manually changing the image processing gradation conversion processing parameters (e.g., WW / WL (= (window width) / (window level)), gradation gamma correction table), but this requires a lot of manual effort on the part of the operator and is inefficient.

[0003] Techniques for performing gradation conversion of X-ray images using image processing are known from Patent Documents 1 and 2, etc. The technique in Patent Document 1 extracts the contour of the subject from the X-ray image, analyzes the curvature of the contour, etc., to detect a region of interest, and performs gradation conversion of the X-ray image so that the density feature (pixel value) of the region of interest becomes a predetermined value. On the other hand, the technique in Patent Document 2 extracts a region of interest from each of an X-ray fluoroscopic image and a still image of one frame of fluoroscopy, and performs gradation conversion so that the density feature of the region of interest becomes a predetermined value. To prevent differences in image density between the fluoroscopic image and the still image, the density feature of the region of interest of the subject is extracted from either the fluoroscopic image or the still image and stored. The stored density feature is used to perform gradation conversion on both the fluoroscopic image and the still image, thereby aligning the visibility of the fluoroscopic image and the still image. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-94829 [Patent Document 2] Japanese Patent Application Laid-Open No. 2006-263224 Summary of the Invention [Problem to be solved by the invention]

[0005] In X-ray fluoroscopy (moving images), it is desirable to display images that have undergone gradation conversion processing in real time. The technologies in Patent Documents 1 and 2 disclose a technique for extracting density features from either a radiographic image of an arbitrary frame of a moving image or a still image, and then converting the gradation of the moving image and still image radiographic images based on the density features, but it is difficult to realize gradation conversion processing of moving images in real time. Furthermore, the image brightness value of a region of interest within a subject varies depending on the subject's body thickness and other external factors. The techniques described in Patent Documents 1 and 2 set a region of interest within the subject based on the contour of the subject, and perform gradation conversion so that the pixel values ​​of the region of interest become predetermined values. However, the pixel values ​​of the region behind the region of interest are not taken into consideration. Therefore, the contrast between the region of interest and the background region may not be easy for the surgeon to see.

[0006] Furthermore, the region of interest of the surgeon varies depending on the target site (organ) on which the procedure is performed, and even if the target site on which the procedure is performed is the same, the region of interest may differ depending on the surgical procedure. Therefore, in the technology of extracting the contour of the subject and setting a region of interest within the subject, as in Patent Documents 1 and 2, the set region of interest does not necessarily match the region of interest that the surgeon actually wants to see. It is also possible that the pixel values ​​of a region of interest different from the region of interest of the surgeon are subjected to gradation conversion.

[0007] Furthermore, in the techniques of Patent Documents 1 and 2, the outline of the subject is required to set the region of interest, but in the case of an X-ray image of the abdomen, etc., the outline of the subject may not appear in the X-ray image. Also, as the surgeon proceeds with the procedure, the shape of the target area may change.

[0008] An object of the present invention is to display an X-ray image in which the region of interest to the operator has been converted into enhanced pixel values ​​that make it easier to view. [Means for solving the problem]

[0009] In order to achieve the above object, the X-ray imaging apparatus of the present invention comprises: an X-ray source that irradiates the subject with X-rays; an X-ray detector that detects X-rays that have passed through the subject; an image generating unit that generates a fluoroscopic image by repeatedly generating X-ray images based on the output of the X-ray detector; a region of interest extraction unit that extracts a region of interest included in an X-ray image; a pixel value conversion parameter calculation unit that calculates a pixel value conversion parameter that converts pixel values ​​so as to enhance the distribution of pixel values ​​of pixels within the region of interest; The image processing unit includes a region of interest enhancement processing unit that uses the pixel value conversion parameters calculated by the pixel value conversion parameter calculation unit to convert pixel values ​​of the X-ray image generated by the image generation unit after the pixel value conversion parameters are calculated. [Effects of the Invention]

[0010] According to the present invention, an X-ray image can be displayed in which the region of interest to the surgeon has been converted into enhanced pixel values ​​that are easy for the surgeon to view. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing the configuration of an X-ray imaging apparatus according to a first embodiment of the present invention. [Figure 2] 5(a) to 5(f) are diagrams for explaining the processing operation of the X-ray imaging apparatus according to the first embodiment. [Figure 3] 6 is a flowchart showing the operation of an image processing unit of the X-ray imaging apparatus of the first embodiment. [Figure 4] 10(a) to 10(f) are diagrams for explaining the processing operation of the X-ray imaging apparatus according to the second embodiment. [Figure 5] 10(a) to 10(f) are diagrams for explaining the processing operation of the X-ray imaging apparatus according to the third embodiment. [Figure 6] FIG. 10 is a diagram showing target regions (classes) of image segmentation for each of a plurality of examination protocols of the X-ray imaging device of the fourth embodiment, regions that can be set as predetermined regions of interest, background regions, and regions that are not the target of histogram calculation. [Figure 7] 13 is a diagram showing a ROI automatic contrast enhancement button 53a and a pixel value conversion parameter automatic update button 53b of the X-ray imaging apparatus of the fifth embodiment. FIG. [Figure 8] 13 is a flowchart showing the operation of an image processing unit of the X-ray imaging apparatus of the fifth embodiment. [Figure 9] FIG. 20 is a diagram showing a ROI automatic contrast enhancement button 53a of the X-ray imaging apparatus according to the sixth embodiment. [Figure 10] 13 is a flowchart showing the operation of an image processing unit of the X-ray imaging apparatus of the sixth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0013] <<Embodiment 1>> An X-ray imaging device 1 of the first embodiment will be described with reference to FIGS.

[0014] 1, the X-ray imaging device 1 is configured to include an X-ray irradiation device 10 including an X-ray source that irradiates an object 2 with X-rays, a flat X-ray detector 20 that detects X-rays that have passed through the object 2, an operation unit 50, and an image processing unit 40. An image display monitor 60 is connected to the image processing unit 40.

[0015] The operation unit 50 includes a button 51 for receiving instructions from the user to start and stop fluoroscopy, a button 52 for receiving instructions to take a snapshot, and a button 53 for receiving instructions to execute a process to highlight a region of interest.

[0016] When the button 51 of the operation unit 50 is operated by the user and an instruction to start fluoroscopy (fluoroscopy ON) is received, the X-ray irradiator 10 supplies power to the X-ray source, causes it to emit X-rays, and irradiates the subject 2 with them.

[0017] The flat X-ray detector 20 detects the X-rays that have passed through the subject 2 .

[0018] The image processing unit 40 includes an image generating unit 41, a region of interest extracting unit 42, a pixel value conversion parameter calculating unit 43, and a region of interest enhancing unit 44.

[0019] When the button 51 of the operation unit 50 is operated by the user and an instruction to start fluoroscopy (fluoroscopy ON) is received, the image generation unit 41 generates an X-ray image based on the output of the X-ray detector 20. As shown in FIG. 2(a), the image generation unit 41 generates a fluoroscopic image (moving image) by repeating the generation of an X-ray image at a predetermined frame rate. The image generation unit 41 outputs the generated fluoroscopic image to the image display monitor 60 for display. Furthermore, the image generation unit 41 is capable of generating not only fluoroscopic images but also captured images.

[0020] The image generating unit 41 includes an LIH memory 41a and a snapshot memory 41b. Each time an X-ray image (fluoroscopic image) is generated at a predetermined frame rate, the generated X-ray image is overwritten and saved in the LIH memory 41a. As a result, the last generated X-ray image (Last Image Hold: LIH) is always saved in the LIH memory 41a. The X-ray image stored in the LIH memory 41a is not limited to the last generated X-ray image. Instead of or in addition to the last generated X-ray image, an X-ray image a predetermined number of frames before the last generated X-ray image, such as one frame or two frames before, may be saved and stored. Furthermore, by using the LIH memory 41a with a capacity capable of saving a predetermined number of X-ray images, it is also possible to save the last generated multiple frames of X-ray images in the LIH memory 41a.

[0021] Furthermore, when the user operates the snapshot acquisition instruction button 52 of the operation unit 50, the image generation unit 41 stores the X-ray image (fluoroscopic image) generated by the image generation unit 41 at that timing in the snapshot memory 41b.

[0022] When the user operates button 53 of operation unit 50 and receives an instruction to execute region of interest enhancement processing, region of interest extraction unit 42 receives the X-ray image (see FIG. 2(b)) stored in LIH memory 41a or snapshot memory 41b, and extracts a region of interest 2a that the user is interested in that is included in the X-ray image, as shown in FIG. 2(c). For example, in the examples of FIGS. 2(b) and 2(c), the image of the bile duct into which a contrast agent has been injected in the X-ray image is region of interest 2a.

[0023] Various methods can be used to extract the region of interest 2a, such as setting a region of a predetermined shape, such as a rectangle, on the X-ray image or extracting pixels belonging to the same region. However, it is preferable that the region of interest 2a does not include any region other than the region of the object (organ or tissue) in which the user is interested. Furthermore, it is preferable that the region of interest 2a does not include only a portion of the region of the object in which the user is interested, but rather the entire object in which the user is interested. Therefore, it is preferable to use a method for extracting pixels belonging to the same region, such as an image segmentation method or a region growing method.

[0024] As shown in FIG. 2(c), the region of interest extraction unit 42 includes a region extraction unit 42a that extracts regions of images of multiple predetermined objects included in the X-ray image, and a selection unit 42b that selects a region of a predetermined object from the multiple regions extracted by the region extraction unit 42a as a region of interest 2a. The region extraction unit 42 is configured to extract regions of images of multiple predetermined objects included in the X-ray image by performing image segmentation using a learning model. The region extraction unit 42 may also be configured to specify a single seed point in the image, and if the pixel values ​​of pixels near the pixel of the seed point are within a predetermined range, consider the pixels to be part of the same region and extract the regions of the images of multiple objects using a region growing method that successively expands the region.

[0025] Examples of images of multiple objects from which regions are extracted include, in the case of a contrast examination of the biliary tract and pancreatic duct, images of a bile duct 21 into which contrast agent has been injected, an endoscopic device 22, a field region 23, and a background region (ribs, vertebrae, and soft tissue) 24, as shown in Figure 2(c).

[0026] The pixel value conversion parameter calculation unit 43 calculates pixel value conversion parameters for converting pixel values ​​so as to enhance the distribution of pixel values ​​of pixels within the region of interest 2a.

[0027] Specifically, pixel value conversion parameter calculation unit 43 generates a histogram of pixel values ​​within region of interest 2a as shown in Fig. 2(d), and calculates the average value ave and standard deviation sd of the pixel values ​​as statistical values ​​from the histogram. Then, as shown in Fig. 2(e), pixel value conversion parameter calculation unit 43 converts the average value ave of the pixel values ​​into a predetermined target pixel value T, and calculates pixel value conversion parameters that convert the pixel value (ave+sd) obtained by adding the standard deviation sd to the average value ave to a pixel value (sd×α+T) obtained by multiplying the standard deviation sd by a coefficient α (α>1) and adding the result to the target pixel value T.

[0028] The pixel value conversion parameters are preferably pixel value conversion curves represented on a graph with the pixel values ​​of the X-ray image before conversion on the horizontal axis and the pixel values ​​after conversion on the vertical axis, as shown in Figure 2(e). The pixel value conversion curve is set to pass through three points: point A, where the pixel value before conversion is the average value ave and the pixel value after conversion is the target pixel value T; point B, where the pixel value before conversion is the pixel value (ave+sd) obtained by adding the standard deviation sd to the average value ave and the pixel value after conversion is the pixel value (sd×α+T) obtained by multiplying the standard deviation sd by a coefficient α (α>1) and adding the result to the target pixel value T; and point C, where the pixel value before conversion is the pixel value (ave-sd) obtained by subtracting the standard deviation sd from the average value ave and the pixel value after conversion is the pixel value (T-sd×α) obtained by subtracting the result of multiplying the standard deviation sd by a coefficient α (α>1) from the target pixel value T. For example, a straight line is set to pass through points C, A, and B.

[0029] As a result, the distribution of pixel values ​​in the region of interest 2a, which has a standard deviation sd and an average pixel value ave as its center ((ave-sd) to (ave+sd)), is expanded to a pixel value range ((T-sd×α) to (sd×α+T)) that is α times the standard deviation sd and has a target pixel value T as its center. This allows the distribution of pixel values ​​in the region of interest 2a to be converted into a distribution of pixel values ​​that emphasizes it.

[0030] Furthermore, it is preferable that the curve 25 on the pixel value conversion curve between point B and the intersection point of the maximum pixel value max be an upwardly convex curve, and it is preferable that the curve 26 on the pixel value conversion curve between point C and zero be a downwardly convex curve.

[0031] The pixel value conversion parameters calculated by the pixel value conversion parameter calculation unit 43 are stored in an LUT memory 43a within the pixel value conversion parameter calculation unit 43. The pixel value conversion parameters are stored in the LUT memory 43a in the form of a lookup table (LUT) that indicates the relationship between pre-conversion pixel values ​​and post-conversion pixel values ​​of a pixel value conversion curve. Alternatively, the pixel value conversion parameters may be stored in the memory 43a in the form of a function that represents the pixel value conversion curve.

[0032] The region of interest enhancement processing unit 44 uses the pixel value conversion parameters stored in the LUT memory 43a to convert the pixel values ​​of the X-ray image generated by the image generating unit 41 after the pixel value conversion parameters are calculated. As a result, the distribution of pixel values ​​in the region of interest 2a is converted into an enhanced X-ray image (see FIG. 2(f)), and the image is displayed on the image display monitor 60.

[0033] It is preferable that the timing at which region of interest extraction unit 42 extracts region of interest 2a and pixel value conversion parameter calculation unit 43 calculates the pixel value conversion parameters is the timing at which the nth fluoroscopy is started (time t1), button 53 instructing execution of region of interest enhancement processing is turned ON, and then button 51 is used to instruct stop of fluoroscopy (time t2).This makes it possible to calculate pixel value conversion parameters and store the LUT in memory 43a during the fluoroscopy-off period (between times t2 and t3).

[0034] The region of interest enhancement processing unit 44 converts the pixel values ​​of the X-ray image generated by the image generation unit 41 after the timing (time t3) when the resumption of fluoroscopy is instructed by button 51 using pixel value conversion parameters in the LUT memory 43a, and can display an X-ray image that emphasizes the distribution of pixel values ​​within the region of interest 2a.

[0035] This prevents pixel values ​​from being converted while fluoroscopic images are being generated continuously at a predetermined frame rate, thereby maintaining visibility for the user. After fluoroscopy is resumed after being turned off, an X-ray image that makes it easy to recognize the region of interest 2a can be displayed.

[0036] If the user so desires, it is of course possible to generate pixel value conversion parameters while continuously generating fluoroscopic images at a predetermined frame rate, and to use the generated pixel value conversion parameters to perform pixel value conversion while continuously generating fluoroscopic images.

[0037] The operation of the image processing unit 40 of the X-ray imaging apparatus of the first embodiment will be described below with reference to the flowchart of FIG.

[0038] (Step S100) In step S100, when the user operates an operation button or the like (not shown) of the operation unit 50 to select an examination protocol (region to be examined and examination contents) and instructs to start the examination, the X-ray imaging apparatus executes the following step S101 and subsequent steps. Here, an example will be described in which the region extraction unit 42a extracts regions by image segmentation.

[0039] (Step S101) The region of interest extraction unit 42 reads a learning model for image segmentation.

[0040] (Step S102) When the user operates the button 51 of the operation unit 50 to instruct the start of fluoroscopy, the process proceeds to step S103.

[0041] (Step S103) The image generation unit 41 generates an X-ray image using the output of the X-ray detector 20 and displays it on the image display monitor 60. The image generation unit 41 also stores the generated X-ray image in the LIH memory 41a. When the user operates the snapshot acquisition instruction button 52 to indicate that a snapshot should be taken, the image generation unit 41 stores the generated image in the snapshot memory 41b.

[0042] (Step S104) The image generating unit 41 generates and displays X-ray images at a predetermined frame rate by repeating step S103 until the user instructs fluoroscopy to be turned off by operating the button 51 of the operation unit 50. If fluoroscopy to be turned off is instructed, the process proceeds to step S105.

[0043] (Step S105) The region of interest extraction unit 42 determines whether the region of interest enhancement processing execution instruction button 53 is on, and if it is off, returns to step S102 and waits until an instruction to start fluoroscopy is given again. On the other hand, if the region of interest enhancement processing execution instruction button 53 is on, the process proceeds to step S106.

[0044] (Step S106) The region extraction unit 42a of the region of interest extraction unit 42 acquires the X-ray image stored in the LIH memory 41a or the snapshot memory 41b (FIG. 2(b)), and performs image segmentation using the learning model read in step S101. As a result, each pixel of the X-ray image is classified into one of multiple predetermined objects (e.g., bile duct 21 into which contrast agent is injected, endoscopic device 22, field region 23, background region (ribs, vertebrae, soft tissue) 24), and image regions of the multiple objects are extracted (see FIG. 2(c)).

[0045] (Step S107) The selector 42b of the region of interest extractor 42 selects the bile duct 21, into which the contrast agent, which is a predetermined target, is injected, as the region of interest 2a (see FIG. 2(c)).

[0046] (Step S108) The pixel value conversion parameter calculation unit 43 generates a histogram of pixel values ​​within the region of interest 2a, as shown in FIG. 2(d), and calculates the average value ave and standard deviation sd of the pixel values.

[0047] (Step S109) 2(e), pixel value conversion parameter calculation unit 43 calculates pixel value conversion parameters (pixel value conversion curves) that enhance the distribution of pixel values ​​within region of interest 2a, and stores the calculated pixel value conversion parameters in LUT memory 43a.

[0048] (Step S110) When the button 51 of the operation unit 50 is operated to instruct the resumption of fluoroscopy, the image generation unit 41 proceeds to step S111.

[0049] (Step S111) The image generating unit 41 resumes generating the X-ray image and stores the generated X-ray image in the LIH memory 41a.

[0050] (Step S112) After resumption, the region of interest enhancement processing unit 44 converts the pixels of the X-ray image generated by the image generation unit 41 using pixel value conversion parameters stored in the LUT memory 43a, generates an X-ray image in which the distribution of pixel values ​​within the region of interest 2a is enhanced, and displays it on the image display monitor 60.

[0051] (Step S113) The image generator 41 and region of interest enhancement processor 44 repeat steps S111 and S112 until the user issues a fluoroscopy-off instruction by operating the button 51. If a fluoroscopy-off instruction is issued, the process returns to step S105.

[0052] In the X-ray imaging device of the first embodiment, the contrast of the distribution of pixel values ​​within the region of interest 2a is enhanced in a protocol for examining a subject using X-ray fluoroscopy, thereby improving the visibility of the region of interest (organs or tissues) of the user (operator).

[0053] In the first embodiment, a configuration has been described in which an instruction to execute a region of interest enhancement process is received from the user by operating the button 53, but it is also possible to start the region of interest enhancement process in response to something other than a user instruction, as in the fourth and fifth embodiments described below. In such a case, it is also possible to configure the operation unit 50 without providing the button 53 for receiving an instruction to execute a region of interest enhancement process.

[0054] <<Embodiment 2>> The X-ray imaging apparatus of the second embodiment will be described with reference to FIG.

[0055] The X-ray imaging apparatus of the second embodiment has the same configuration as the X-ray imaging apparatus 1 of the first embodiment, but differs from the first embodiment in that the region of interest extraction unit 42 performs a predetermined expansion process on the extracted region of interest 2a to further set an adjacent background region 2b around the region of interest 2a (see FIG. 4(c)). In the second embodiment, pixel value conversion parameters are calculated using statistics obtained from histograms of pixel values ​​of the region of interest 2a and the adjacent background region 2b (FIGS. 4(d) and (e)).

[0056] The pixel value conversion parameter calculation unit 43 converts the average value ave1 of the pixel values ​​in the region of interest 2a into a predetermined target pixel value T, and calculates pixel value conversion parameters that convert the average value ave2 of the pixel values ​​in the adjacent background region 2b into the predetermined target pixel value T, which is the value (T+sub×β) obtained by multiplying the difference sub between the average values ​​ave1 and ave2 by a coefficient β.

[0057] Specifically, the pixel value conversion curve representing the pixel value conversion parameters is set to pass through points A and D, as shown in FIG. 4(e). At point A, the pixel value before conversion is the average pixel value ave1 of the pixel values ​​in the region of interest 2a, and the pixel value after conversion is the target pixel value T. At point D, the pixel value before conversion is the average pixel value ave2 of the pixel values ​​in the adjacent background region 2b, and the pixel value after conversion is the pixel value (T+sub×β) obtained by adding the difference sub between the average pixel values ​​ave1 and ave2 multiplied by a coefficient β (β>1) to the target pixel value T. For example, a straight line passing through points A and D is set.

[0058] As a result, the difference sub between the average pixel value ave1 in the region of interest 2a and the average pixel value ave2 in the adjacent background region 2b is expanded to a range of pixel values ​​that is β times the difference sub, with the target pixel value T as the center.

[0059] Therefore, compared to the first embodiment, the contrast difference between the region of interest 2a and the background adjacent to the region of interest 2a (adjacent background region 2b) can be more appropriately emphasized, improving the visibility of the target of interest to the surgeon.

[0060] The configuration and operation of the X-ray imaging apparatus of the second embodiment other than those described above are the same as those of the first embodiment, and therefore a description thereof will be omitted.

[0061] <<Embodiment 3>> The X-ray imaging apparatus of the third embodiment will be described with reference to FIG.

[0062] The X-ray imaging device of the third embodiment, like the X-ray imaging device 1 of the second embodiment, is configured to further set an adjacent background region 2b, but differs from the second embodiment in that a second region of interest 28 is also selected in the third embodiment (see FIG. 5(c)). In the third embodiment, pixel value conversion parameters are calculated using statistics obtained from histograms of pixel values ​​of the region of interest 2a, the second region of interest 28, and the adjacent background region 2b (FIGS. 5(d) and 5(e)).

[0063] For example, the region extraction unit 42a of the region of interest extraction unit extracts, as a plurality of predetermined targets, the bile duct 21 into which a contrast agent has been injected, the endoscopic device 22, the field region 23, and the background region (ribs, vertebrae, and soft tissue) 24, as well as the bile duct transparent region 28 (see FIG. 5(c)). The selection unit 42b selects the bile duct 21 into which a contrast agent has been injected as the region of interest 2a, and the bile duct transparent region 28 as the second region of interest 2c.

[0064] 5(d), pixel value conversion parameter calculation unit 43 generates histograms for region of interest 2a, second region of interest 2c, and adjacent background region 2b. Then, pixel value conversion parameters are calculated to convert the average value ave1 of pixel values ​​in first region of interest 2a to a predetermined target pixel value T, to convert the average value ave2 of pixel values ​​in the adjacent background region to a value obtained by multiplying the difference sub1 between the average values ​​ave1 and ave2 by a coefficient β, and to convert the average value ave3 of pixel values ​​in the second region of interest to a value obtained by multiplying the difference sub2 between the average values ​​ave1 and ave3 by a coefficient γ.

[0065] Specifically, in the third embodiment, the pixel value conversion curve representing the pixel value conversion parameters is set to pass through points A, D, and E, as shown in FIG. 5( e). At point A, the pixel value before conversion is the average value ave1 of the pixel values ​​in the region of interest 2a, and the pixel value after conversion is the target pixel value T. At point D, the pixel value before conversion is the average value ave2 of the pixel values ​​in the adjacent background region 2b, and the pixel value after conversion is the pixel value (T+sub1×β) obtained by adding the difference sub1 between the average values ​​ave1 and ave2 multiplied by a coefficient β (β>1) to the target pixel value T. At point E, the pixel value before conversion is the average value ave3 of the pixel values ​​in the second region of interest 2c, and the pixel value after conversion is the pixel value (T-sub2×γ) obtained by subtracting the difference sub2 between the average values ​​ave1 and ave3 multiplied by a coefficient γ (γ>1) from the target pixel value T.

[0066] As a result, the difference sub1 between the average pixel value ave1 in the region of interest 2a and the average pixel value ave2 in the adjacent background region 2b is expanded to a range of pixel values ​​obtained by multiplying the difference sub1 by β, with the target pixel value T as the center. Also, the difference sub2 between the average pixel value ave1 in the region of interest 2a and the average pixel value ave3 in the second region of interest 2c is expanded to a range of pixel values ​​obtained by multiplying the difference sub2 by γ, with the target pixel value T as the center.

[0067] Therefore, even when a plurality of regions of interest 2a, 2c are present in an X-ray image, an appropriate contrast difference with the adjacent background region 2b can be obtained, improving the visibility of the regions of interest 2a, 2c for the operator.

[0068] The configuration and operation of the X-ray imaging apparatus of the third embodiment other than those described above are the same as those of the first and second embodiments, and therefore a description thereof will be omitted.

[0069] <<Embodiment 4>> The X-ray imaging apparatus of the fourth embodiment will be described with reference to FIGS.

[0070] The X-ray imaging apparatus of the fourth embodiment is capable of applying any one of the region of interest enhancement processes of the first to third embodiments to a plurality of types of examination protocols (areas of the examination target and examination contents). Therefore, in the fourth embodiment, a learning model is prepared for each of the plurality of types of examination protocols, and the image of the target from which the region is to be extracted and the number of targets are determined in advance. The learning model for each of the plurality of types of examination protocols is stored in the region extraction unit 42a of the region of interest extraction unit 42.

[0071] 3 in the first embodiment, if the user selects an examination protocol on the operation unit 50 in step S100, the region extraction unit 42a of the region of interest extraction unit 42 reads a learning model corresponding to the selected examination protocol in step S101. Then, in steps S106 and S107, regions of a plurality of predetermined targets are extracted for each examination protocol, and the selection unit 42b selects the predetermined targets as regions of interest.

[0072] For example, as shown in Figure 6, the X-ray imaging device of embodiment 3 allows the selection of "endoscopic retrograde cholangiopancreatography" and "spinal canalography / nerve root block examination" as examination protocols, and a learning model is prepared for each. In the case of "endoscopic retrograde cholangiopancreatography," the regions to be extracted by image segmentation are the following (1) to (8), and the regions that can be selected as regions of interest are the following (1) to (3). The background region is (7), and the regions that are not subject to histogram calculation by the pixel value conversion parameter calculation unit 43 are (4), (5), (6), and (8). (1)Cholangiography area (2) Transparent area within the cholangiography area (3) Artificial Devices (4) Lung field area (5) Direct radiation (X-ray penetration) region (6) Intestinal gas (7) Other background areas (ribs, vertebrae, soft tissue) (8) X-ray aperture area On the other hand, in the case of a "spinal cavity imaging / nerve root block examination," the regions to be extracted by image segmentation are the following (1) to (8), and the regions that can be set in advance as regions of interest are the following (1) to (3). The background region is (7), and the regions that are not included in the histogram calculation by the pixel value conversion parameter calculation unit 43 are (4), (5), (6), and (8). (1) Contrast area (2) Spine area (3) Needle (4) Lead gloves (5) Direct radiation (X-ray penetration) region (6) Direct radiation (X-ray penetration) region (7) Other background areas (ribs and soft tissues) (8) X-ray aperture area In the X-ray imaging device of embodiment 4, by preparing a learning model for image segmentation for each examination protocol (area to be examined and examination content), it is possible to perform contrast enhancement in accordance with the examination protocol for X-ray fluoroscopic diagnosis.

[0073] The configuration and operation of the X-ray imaging apparatus of the fourth embodiment other than those described above are the same as those of the first to third embodiments, and therefore a description thereof will be omitted.

[0074] <<Embodiment 5>> The X-ray imaging apparatus of the fifth embodiment will be described with reference to FIGS.

[0075] 7, the ROI enhancement processing execution instruction button 53 of the operation unit 50 includes an ROI automatic contrast enhancement button 53a and a pixel value conversion parameter automatic update button 53b displayed on the image display monitor 60. This allows the user to select whether to execute the ROI enhancement processing and the timing of execution, such as whether to execute it automatically when the X-ray image in the LIH memory 41a or snapshot memory 41b is updated.

[0076] The operation of the image processing unit 40 of the fifth embodiment will be described below with reference to the flow of FIG.

[0077] (Step S700) In step S700, if the user selects an examination protocol (region to be examined and examination contents) using the operation unit 50 and issues an instruction to start the examination, the process proceeds to step S701.

[0078] (Steps S701 and S702) The region of interest extraction unit 42 determines whether the region of interest automatic contrast enhancement button 53a is on, and if it is off, proceeds to step S702. In step S702, steps S102 to S104 of the flow in Fig. 3 are executed, but calculation of pixel value conversion parameters and region of interest enhancement processing are not performed.

[0079] On the other hand, if the ROI automatic contrast enhancement button 53a is on, the ROI extraction unit 42 proceeds to step S703.

[0080] (Step S703) In step S700, the region extraction unit 42a of the region of interest extraction unit 42 reads a learning model corresponding to the examination protocol (region of the examination target and examination contents) selected by the user.

[0081] (Steps S704 and S705) Then, the region of interest extraction unit 42 determines whether the pixel value conversion parameter automatic update button 53b is on, and if it is off, the process proceeds to step S705.

[0082] 3 is executed only when the user temporarily instructs with the pixel value conversion parameter automatic update button 53b or another instruction button (not shown). That is, the pixel value conversion parameters are calculated using the X-ray image in the LIH memory 41a or snapshot memory 41b, the LUT memory 43a is updated, and the region of interest enhancement process of the generated X-ray image is performed.

[0083] (Step S706) On the other hand, if pixel value conversion parameter automatic update button 53b is on in step S704, region of interest extraction unit 42 proceeds to step S706. In step 706, region of interest extraction unit 42 executes steps S106 to S113 in Fig. 3 every time a new X-ray image is stored in LIH memory 41a or snapshot memory 41b, calculates pixel value conversion parameters, updates LUT memory 43a, and performs region of interest enhancement processing on the generated X-ray image.

[0084] In this way, in the fifth embodiment, by providing the ROI automatic contrast enhancement button 53a and the pixel value conversion parameter automatic update button 53b, it is possible to select whether to execute the ROI enhancement process and the timing of execution, whether to execute it automatically when the X-ray image in the LIH memory 41a or the snapshot memory 41b is updated. Therefore, it is possible to operate the ROI enhancement process only when the user needs it.

[0085] The configuration and operation of the X-ray imaging apparatus of the fifth embodiment other than those described above are the same as those of the first to fourth embodiments, and therefore a description thereof will be omitted.

[0086] <<Embodiment 6>> The X-ray imaging apparatus of the sixth embodiment will be described with reference to FIGS.

[0087] In the sixth embodiment, whether or not to perform region of interest enhancement processing is automatically determined based on whether or not a predetermined region of interest 2a, etc. is extracted during image segmentation of the X-ray image stored in the LIH memory 41a or snapshot memory 41b.

[0088] As shown in FIG. 9, the ROI enhancement process execution instruction button 53 of the operation unit 50 includes an ROI automatic contrast enhancement button 53a displayed on the image display monitor 60 as shown in FIG.

[0089] When the region of interest automatic contrast enhancement button 53a is turned on by the operator, the image processing unit 40 of the sixth embodiment operates as shown in the flow of FIG.

[0090] (Step S800) In step S800, if the user selects an examination protocol (site to be examined and examination contents) using the operation unit 50 and issues an instruction to start the examination, the process proceeds to step S801.

[0091] (Step S801) In step S800, the region extraction unit 42a of the region of interest extraction unit 42 reads a learning model corresponding to the examination protocol (region of the examination target and examination contents) selected by the user.

[0092] (Step S802,) The region extraction unit 42a of the region of interest extraction unit 42 acquires the X-ray image from the LIH memory 41a or the snapshot memory 41b.

[0093] (Step S803) The region extraction section 42a of the region of interest extraction section 42 performs image segmentation to extract a region of the image of a predetermined target.

[0094] (Step S804) The region extraction unit 42a of the region of interest extraction unit 42 analyzes the extracted region and determines whether a predetermined region of interest 2a has been extracted. For example, in the case of an "endoscopic retrograde cholangiopancreatography examination," if a cholangiography region is preset as the region of interest 2a, it determines whether the cholangiography region has been extracted by image segmentation.

[0095] If the region of interest 2a has not been extracted and only the background region has been extracted, the process proceeds to step S805, and if the region of interest 2a and the background region have been extracted, the process proceeds to step S806.

[0096] (Step S805) If the region of interest 2a is not extracted in step S804, the region of interest extraction unit 42 does not perform the region of interest enhancement process in step S805, ie, executes steps S102 to S104 of the flow in FIG.

[0097] (Step S806) If the region of interest 2a is extracted in step S804, the region of interest extraction unit 42 calculates pixel value conversion parameters and performs region of interest enhancement processing every time the X-ray image in the LIH memory 41a or snapshot memory 41b is updated in step S806. Specifically, steps S106 to S113 in the flow of FIG. 3 are executed.

[0098] In this way, in embodiment 6, whether or not to perform region of interest enhancement processing is switched based on whether or not the region of interest 2a is extracted from the X-ray image in the LIH memory 41a or the snapshot memory 41b, without any user operation, thereby reducing the burden on the operator.

[0099] The configuration and operation of the X-ray imaging apparatus of the sixth embodiment other than those described above are the same as those of the first to fourth embodiments, and therefore a description thereof will be omitted. [Explanation of symbols]

[0100] 1 X-ray imaging device 2. Subject 2a Area of ​​interest 2b Adjacent background area 2c Second Area of ​​Interest 10 X-ray irradiation device 20 X-ray detector 21 Bile duct 22 Endoscopic Devices 23 Field area 24 Background area 25 curve 26 curve 28 Translucency area in the bile duct 40 Image processing section 41 Image generation unit 41a LIH memory 41b Snapshot memory 42 Region of interest extraction unit 42a Region extraction part 42b Selection section 43 Pixel value conversion parameter calculation unit 43a LUT memory 44 Region of interest enhancement processing unit 50 Control section 51 Fluoroscopy start / stop button 52 Snapshot acquisition instruction button 53 Region of interest enhancement processing execution instruction button 53a Region of Interest Auto Contrast Enhancement Button 53b Pixel value conversion parameter automatic update button 60 Image display monitor

Claims

1. an X-ray source that irradiates an object with X-rays; an X-ray detector that detects X-rays that have passed through the subject; an image generating unit that generates a fluoroscopic image by repeatedly generating X-ray images based on an output of the X-ray detector; a region of interest extraction unit that extracts a region of interest included in the X-ray image; a pixel value conversion parameter calculation unit that calculates a pixel value conversion parameter that converts the pixel values ​​so as to enhance the distribution of pixel values ​​of the pixels within the region of interest; and a region of interest enhancement processing unit that converts pixel values ​​of the X-ray image generated by the image generation unit after the pixel value conversion parameter is calculated, using the pixel value conversion parameter calculated by the pixel value conversion parameter calculation unit.

2. 2. The X-ray imaging device according to claim 1, an X-ray imaging device, characterized in that the region of interest extraction unit includes: a region extraction unit that extracts regions of images of predetermined objects included in the X-ray image; and a selection unit that selects a region of a predetermined object as the region of interest from the plurality of regions extracted by the region extraction unit.

3. 3. The X-ray imaging apparatus according to claim 2, further comprising: a memory for storing the X-ray image generated by the image generating unit; and an operation unit for receiving instructions from a user to start and stop image generation; the memory stores the last generated X-ray image among the X-ray images repeatedly generated by the image generation unit, or the X-ray image generated at a predetermined timing; the region of interest extraction unit extracts the region of interest from the X-ray image stored in the memory when the operation unit receives an instruction to stop image generation from a user, and the pixel value conversion parameter calculation unit calculates the pixel value conversion parameter for the extracted region of interest; The X-ray imaging device is characterized in that, when the operation unit receives an instruction from a user to resume image generation, the region of interest enhancement processing unit converts the pixel values ​​of the X-ray image generated by the image generation unit after the resumption using the pixel value conversion parameters.

4. 3. The X-ray imaging device according to claim 2, wherein the region extraction unit extracts regions of images of a plurality of predetermined objects included in the X-ray image by performing image segmentation using a learning model.

5. 2. The X-ray imaging device according to claim 1, wherein the pixel value conversion parameter calculation unit calculates the average value ave and standard deviation sd of pixel values ​​within the region of interest, converts the average value ave into a predetermined target pixel value T, and calculates the pixel value conversion parameter to convert a pixel value obtained by adding the standard deviation sd to the average value ave to a pixel value obtained by adding a value obtained by multiplying the standard deviation sd by a coefficient α (α > 1) to the target pixel value T.

6. 6. The X-ray imaging apparatus according to claim 5, wherein the pixel value conversion parameter is a pixel value conversion curve expressed on a graph with pixel values ​​of the X-ray image before conversion on the horizontal axis and pixel values ​​of the X-ray image after conversion on the vertical axis, an X-ray imaging device characterized in that the pixel value conversion curve passes through a point where the pixel value before conversion is the average value ave and the pixel value after conversion is the target pixel value T; a point where the pixel value before conversion is the pixel value obtained by adding the standard deviation sd to the average value ave and the pixel value after conversion is the pixel value obtained by multiplying the standard deviation sd by a coefficient α (α > 1) and adding that value to the target pixel value T; and a point where the pixel value before conversion is the pixel value obtained by subtracting the standard deviation sd from the average value ave and the pixel value after conversion is the pixel value obtained by subtracting the value obtained by multiplying the standard deviation sd by a coefficient α (α > 1) from the target pixel value T.

7. 2. The X-ray imaging apparatus according to claim 1, wherein the region of interest extraction unit further sets an adjacent background region around the region of interest by performing a predetermined expansion process on the region of interest; the pixel value conversion parameter calculation unit calculates the pixel value conversion parameters to convert an average value ave1 of pixel values ​​in the region of interest into a predetermined target pixel value T, and to convert an average value ave2 of pixel values ​​in the adjacent background region into a predetermined target pixel value T obtained by multiplying a difference sub between the average value ave1 and the average value ave2 by a coefficient β.

8. 3. The X-ray imaging device according to claim 2, the selection unit of the region of interest extraction unit selects a first region of interest and a second region of interest as the regions of interest from the plurality of regions extracted by the region extraction unit; the region of interest extraction unit further sets an adjacent background region adjacent to a periphery of the first region of interest by performing a predetermined expansion process on the first region of interest; wherein the pixel value conversion parameter calculation unit calculates pixel value conversion parameters that convert an average value ave1 of pixel values ​​in the first region of interest into a predetermined target pixel value T, convert an average value ave2 of pixel values ​​in the adjacent background region into a predetermined target pixel value T, and convert an average value ave3 of pixel values ​​in the second region of interest into a predetermined target pixel value T, and convert an average value ave3 of pixel values ​​in the second region of interest into a predetermined target pixel value T, and convert an average value ave2 ...1 of pixel values ​​in the first region of interest into a predetermined target pixel value T

9. 3. The X-ray imaging apparatus according to claim 2, further comprising an operation unit that accepts a selection of an examination target of a subject from an operator, The region of interest selection unit selects, as the region of interest, a region corresponding to the inspection object accepted by the operation unit from among the plurality of regions extracted by the region extraction unit. An X-ray imaging device characterized by:

10. 5. The X-ray imaging apparatus according to claim 4, further comprising an operation unit that accepts a selection of an examination target of a subject from an operator, The learning model is prepared for each of a plurality of test subjects, The X-ray imaging device is characterized in that the region extraction unit selects and uses the learning model corresponding to the examination object accepted by the operation unit to extract the region.

11. 5. The X-ray imaging apparatus according to claim 4, further comprising an operation unit that accepts a selection of an examination target of a subject from an operator, The learning model is prepared for each of a plurality of test subjects, The X-ray imaging device is characterized in that the region extraction unit selects and uses the learning model corresponding to the examination object accepted by the operation unit to extract the region.

12. 4. The X-ray imaging device according to claim 3, wherein the region of interest extraction unit extracts the region of interest each time the X-ray image is stored in the memory, the pixel value conversion parameter calculation unit calculates the pixel value conversion parameter for the extracted region of interest, and the region of interest enhancement processing unit performs processing to convert pixel values ​​of the X-ray image generated by the image generation unit using the calculated pixel value conversion parameter.

13. 2. The X-ray imaging apparatus according to claim 1, further comprising an operation unit that receives a selection from an operator as to whether or not to perform an enhancement process for a region of interest of a subject, An X-ray imaging device characterized in that, when the operation unit receives a selection from the operator not to perform enhancement processing of the region of interest, the image generation unit displays the generated image on the display unit as is.

14. 2. The X-ray imaging device according to claim 1, wherein the region of interest extraction unit extracts the region of interest from the X-ray image, then evaluates the extraction result, and automatically determines whether to cause the pixel value conversion parameter calculation unit to calculate pixel value conversion parameters depending on whether the region of interest has been extracted.

15. an X-ray source that irradiates an object with X-rays, and an X-ray detector that detects the X-rays that have passed through the object; an operation unit that receives instructions from a user to start and stop image generation; an image generating unit that generates a fluoroscopic image by repeatedly generating X-ray images based on the output of the X-ray detector; a memory for storing the last generated X-ray image or an X-ray image generated at a predetermined timing among the X-ray images repeatedly generated by the image generating unit; a region of interest extraction unit that extracts a region of interest included in the X-ray image stored in the memory when the operation unit receives an instruction to stop image generation from a user; a pixel value conversion parameter calculation unit that calculates a pixel value conversion parameter for converting the pixel value using the pixel value of the pixel in the region of interest; and a region of interest enhancement processing unit that, when the operation unit receives an instruction from a user to resume image generation, converts pixel values ​​of the X-ray image generated by the image generation unit after the restart using the pixel value conversion parameters.

16. a region of interest extraction unit that receives an X-ray image generated as a fluoroscopic image by an X-ray imaging device and extracts a predetermined region of interest included in the received X-ray image; a pixel value conversion parameter calculation unit that calculates a pixel value conversion parameter that converts the pixel values ​​so as to enhance the distribution of pixel values ​​of the pixels within the region of interest; and a region of interest enhancement processing unit that uses the pixel value conversion parameters calculated by the pixel value conversion parameter calculation unit to convert pixel values ​​of the X-ray image generated by the X-ray imaging device after the pixel value conversion parameters are calculated.

Images