X-ray CT equipment, image processing equipment, and motion correction image reconstruction methods for CT images.

Abstract

This invention provides an X-ray CT apparatus, an image processing apparatus, and a method for motion-corrected image reconstruction of CT images. It obtains motion information of a subject during scanning from image pairs, and when generating CT images through motion-corrected image reconstruction, it suppresses misidentification of motion components and excessive smoothing, thereby improving the accuracy of motion correction. The method includes a filtering unit for reducing noise in the image pairs. The filtering unit calculates the relationship between an index of noise levels obtained based on information affecting the noise levels of the first and second images constituting the image pair, and uses this relationship to adjust the smoothing degree of each image. For image pairs with reduced noise through filtering, it determines whether motion is present, and performs normalization based on the difference in normalization degree between the two images caused by the presence or absence of motion, detecting the magnitude and direction of the motion.

Description

Technical Field

[0001] This invention relates to an X-ray CT apparatus for obtaining medical images by irradiating a subject with X-rays, and to an activity correction image reconstruction processing technology for improving the accuracy and efficiency of reading images of moving subjects. Background Technology

[0002] In CT scans of moving subjects such as the heart, motion artifacts sometimes occur in the images due to the subject's movement. These artifacts degrade image quality, potentially reducing the accuracy and efficiency of diagnosis for physicians and technicians (hereinafter collectively referred to as examiners). Therefore, motion-corrected image reconstruction processing is performed to reduce motion artifacts in CT images of moving subjects.

[0003] In motion-corrected image reconstruction processing, the motion of the subject is inferred from a pair of images (a first image and a second image) reconstructed at temporally opposite positions centered on the target image reconstruction location. Using the inferred motion information, backprojection is performed while correcting the image at the target reconstruction location to perform reconstruction. However, if the two images used in motion inference contain noise, problems arise such as misidentifying noise as motion or failing to detect motion mixed with noise. Therefore, it is crucial to accurately extract the subject's motion from the noisy first and second images.

[0004] Patent Document 1 discloses a method for extracting only the activity of a subject by performing noise reduction processing on a first image and a second image. In the method disclosed in Patent Document 1, noise reduction is performed using a low-pass filter based on the X-ray dose at the time the first image and the second image were acquired.

[0005] Existing technical documents

[0006] Patent documents

[0007] Patent Document 1: US Patent No. 10,165,989

[0008] In the method disclosed in Patent Document 1, noise is reduced according to the noise of the first and second images respectively, but the relationship between the noise between the images is not considered. Therefore, there is a possibility that the difference in noise between the two images is misidentified as a motion component, or that the motion component is lost due to excessive smoothing, which would impair the motion correction effect. Furthermore, in existing motion correction methods, since the current motion of the subject is not considered, there is a possibility that unnatural distortion may occur in the image when motion correction is performed on a subject with very little motion. Summary of the Invention

[0009] Therefore, the objective of this invention is to suppress the misidentification of the aforementioned active components and excessive smoothing, thereby improving the accuracy of activity correction.

[0010] To address the aforementioned issues, this invention prevents a reduction in the effectiveness of activity correction by determining the smoothing parameters of the filter based on the relationship between the noise levels of a pair of images used in activity correction, namely the first image and the second image, and applying these parameters to each image.

[0011] That is, the X-ray CT apparatus of the present invention comprises: an imaging unit having an X-ray source and an X-ray detector that rotate around a subject, and acquiring transmitted X-ray data of the subject within a given angular range; an image reconstruction unit that generates a reconstructed image using the transmitted X-ray data acquired by the imaging unit; and an image processing unit. The image processing unit further comprises: an image pair generation unit that generates image pairs at opposing positions using a portion of the transmitted X-ray data; a noise reduction unit that reduces noise in each image pair generated by the image pair generation unit; and a motion information acquisition unit that acquires motion information of the subject being scanned using the noise-reduced image pairs. The noise reduction unit reduces noise in each image pair based on an index relating the noise levels of the images contained in the image pairs. The image reconstruction unit uses the motion information calculated by the motion information acquisition unit to correct for motion of the subject being scanned and generates a reconstructed image.

[0012] The noise level index is a measure of the amount of noise obtained from various conditions that affect the noise level of an image during photography.

[0013] Furthermore, the image processing apparatus of the present invention is an image processing apparatus that processes X-ray data collected by an X-ray CT apparatus, and has the same functions as the image processing unit of the aforementioned X-ray CT apparatus.

[0014] Furthermore, the activity correction image reconstruction method of the present invention uses transmitted X-ray data to correct the activity of the subject in a scan, and is a method for reconstructing CT images. The method is characterized in that a pair of images is generated using a portion of the transmitted X-ray data, and noise reduction is performed on the pair of images respectively. At this time, the filtering smoothness during noise reduction is adjusted based on the relationship between the noise levels of the images contained in the pair of images. The activity information of the subject is obtained from the pair of noise-reduced images, and reconstruction is performed using the activity information and the transmitted X-ray data.

[0015] Invention Effects

[0016] This system can perform noise reduction processing on both the first and second images to minimize the loss of motion information and maximize the effect of motion correction. Furthermore, it can adjust the intensity of motion correction based on the presence or absence of subject motion, suppressing artifacts caused by image noise in the corrected image. As a result, it improves the accuracy and efficiency of disease diagnosis for users. Attached Figure Description

[0017] Figure 1 This is a diagram showing the overall structure of the X-ray CT device of the present invention.

[0018] Figure 2 This is a diagram showing the constituent elements of the X-ray CT apparatus of the present invention.

[0019] Figure 3A It is a diagram showing the shooting process that includes activity correction and reconstruction.

[0020] Figure 3B This is a diagram illustrating the process of image reconstruction with active correction.

[0021] Figure 4 This is a functional block diagram of the image processing unit in Embodiment 1.

[0022] Figure 5 This is a diagram illustrating the flow of the active correction image reconstruction process in Implementation 1.

[0023] Figure 6 It is a diagram showing the relationship between the first image and the second image relative to the target reconstruction position.

[0024] Figure 7A It is a graph showing the relationship between the ECG waveform data and the tube current value in synchronous ECG imaging, and a graph showing the ECG waveform and the phase of the heart being captured.

[0025] Figure 7B This is a graph showing the relationship between ECG waveform data and tube current value during synchronous ECG recording, and a graph showing the relationship between the phase of the captured heart and tube current.

[0026] Figure 7C This is a graph showing the relationship between ECG waveform data and tube current value during synchronous ECG recording, and a graph showing the relationship between the phase of the captured heart and tube current.

[0027] Figure 8A The diagram shows a smoothing parameter adjustment method based on the relationship of tube current values ​​in Implementation 1, and the diagram shows an adjustment using the average tube current value.

[0028] Figure 8BThe diagram shows a smoothing parameter adjustment method based on the relationship between tube current values ​​in Implementation 1, and the diagram shows an adjustment using the average tube current ratio.

[0029] Figure 9 This is a diagram showing an example of the filtering result of Implementation Method 1.

[0030] Figure 10 This is a functional block diagram of the image processing unit in Embodiment 2.

[0031] Figure 11 This is a diagram illustrating the process of setting image reconstruction conditions in Implementation Method 2.

[0032] Figure 12 This is a diagram showing an example of image reconstruction condition settings for Implementation Method 2.

[0033] Figure 13 This is a diagram illustrating the setting of the reconstructed cardiac phase position.

[0034] Explanation of reference numerals in the attached figures

[0035] 1: X-ray CT apparatus; 3: Subject; 10: Imaging unit; 20: Operation unit; 100: gantry; 101: Bed unit; 102: X-ray generating device; 103: X-ray detection device; 104: Collimation device; 105: High voltage generating device; 107: Drive device; 106: Data collection device; 200: Central control unit; 210: Input / output device; 211: Display device; 212: Input device; 213: Storage device; 301: Image reconstruction unit; 302: Image processing unit; 310: Image pair generation unit; 330: Filtering unit; 340: Tube current calculation unit; 350: Motion information acquisition unit; 351: Motion determination unit; 352: Motion detection unit; 360: Reconstructed cardiac phase setting unit. Detailed Implementation

[0036] The embodiments of the present invention are illustrated below using the accompanying drawings.

[0037] The overall structure of the X-ray CT device using the present invention is initially described.

[0038] like Figure 1 As shown, the X-ray CT apparatus 1 includes: an imaging unit 10 with a gantry 100 and a bed 101 for capturing tomographic and fluoroscopic images of the subject 3; and an operation unit 20 for operating and controlling the imaging unit 10.

[0039] like Figure 2As shown, the frame 100 includes: an X-ray generating apparatus 102 that generates X-rays to irradiate the subject 3; a collimating device 104 that converges the X-ray beam generated from the X-ray generating apparatus 102; an X-ray detection apparatus 103 that detects X-rays transmitted through the subject; a scanner 108 equipped with these apparatuses; a high-voltage generating apparatus 105 that applies a high voltage to the X-ray generating apparatus 102; a data collection apparatus 106 that collects transmitted X-ray data obtained from the X-ray detection apparatus 103; and a drive device 107 that rotates the scanner around the subject 3. Although not shown, the X-ray generating apparatus 102 includes an X-ray tube, through which a given tube current is flowed to irradiate the subject 3 with a given amount of X-rays.

[0040] The operation unit 20 includes: a central control unit 200 controlling the various devices built into the control rack; and an input / output device 210 serving as a user interface for communication between the user and the central control unit 200. The central control unit 200 houses a computational unit 30 that performs various calculations, such as image reconstruction, on the X-ray data collected by the data collection device 106. However, it may also include a different computational unit than the central control unit 200, which functions as the computational unit 30. The functions of the central control unit 200 are implemented by reading and executing a program that describes the calculation algorithms and control processing steps. Some of the calculations and processing performed by the computational unit 30 can also be performed using PLDs (Programmable Logic Devices) such as ASICs or FPGAs.

[0041] The input / output device 210 includes: an input device 212 for the operator to input shooting conditions, etc.; a display device 211 for displaying shooting data such as captured images and a GUI; and a storage device 213 for storing shooting data such as programs and device parameters.

[0042] The arithmetic unit 30 includes: an image reconstruction unit 301 that performs back-projection processing on the X-ray data obtained by the data collection device 106 to create a tomographic image; and an image processing unit 302 that performs image data analysis, image correction, etc. Image correction includes active correction image reconstruction. Details will be described later.

[0043] The central control unit 200 controls the imaging unit 10 (X-ray generating device 102, X-ray detection device 103, high-voltage generating device 105, collimating device 104, bed device 101, drive device 107, data collection device 106), input / output device 210, and processing unit 30 via operation instructions from the operator through the input device 212. Under the control of the central control unit 200, these units operate to perform CT image reconstruction, correction of the reconstructed CT image, and other tasks.

[0044] refer to Figure 3A , Figure 3B The process is used to illustrate the general operation of the X-ray CT device under the control of the central control unit 200.

[0045] <Step S1>

[0046] Subject 3 is placed on the bedside table 101 for positioning imaging. Positioning imaging is used to set the imaging range of subject 3. While changing the relative position of scanner 108 and bedside table 101 (subject 3), a transmitted X-ray image is acquired along the body axis. The examiner uses the transmitted X-ray image to set the imaging range. Then, the imaging unit 10 performs tomographic imaging of the imaging range set based on the positioning image, accompanied by the rotation of scanner 108, and collects transmitted X-ray data of the subject.

[0047] <Step S2>

[0048] Image reconstruction conditions are set for the X-ray data of the subject acquired in S1. These conditions include, for example, image thickness (thickness of the cross-section), field of view (FOV), filter conditions, and, in the case of simultaneous ECG imaging, settings such as reconstructing the cardiac phase (target cardiac phase reconstruction: which cardiac phase to reconstruct). In simultaneous ECG imaging, the target reconstructed cardiac phase is set to determine the position of the target reconstructed image. The image processing unit 302 receives these image reconstruction conditions set by the user via the input device 212.

[0049] <Step S3>

[0050] The image processing unit 302 performs image reconstruction based on the image reconstruction conditions set in the condition setting step S2, using the transmitted X-ray data of the subject obtained in the shooting step S1. At this time, the motion information of the subject during shooting is obtained, the motion is corrected, and reconstruction is performed (motion-corrected image reconstruction).

[0051] like Figure 3B As shown, the motion-corrected image reconstruction step S3 includes: an image pair generation step S31, which generates a pair of images (image pair) used for motion detection based on the X-ray data collected in the shooting step S1; an motion information acquisition step S32, which uses the image pair to detect motion; and an image reconstruction step S33, which uses the X-ray data collected in the shooting step S1 and the motion information detected in step S32. In the image pair generation step S31, noise reduction processing (filtering) is performed on the first image and the second image constituting the image pair. Details of these processing will be described later.

[0052] <Step S4>

[0053] Finally, the active correction image data created in reconstruction step S3 is displayed on display device 211.

[0054] The corrected CT images are obtained through the above steps S1 to S4 and presented to the examiner. The following describes the specific implementation method of the corrected image reconstruction process.

[0055] [Implementation Method 1]

[0056] This embodiment is characterized in that, when generating image pairs, filtering is performed based on the relationship between the noise levels of the two images. Furthermore, information regarding the presence or degree of activity generated by the subject is obtained, and this information is reflected in the activity-corrected image reconstruction. In this embodiment, the case where tube current is used as an indicator of noise level when obtaining the transmitted X-ray data used in generating the image pairs will be described.

[0057] exist Figure 4 An example of the structure of the image processing unit 302 of this embodiment is shown. As shown, the image processing unit 302 includes: an image pair generation unit 310; a noise reduction filtering unit 330; an activity information acquisition unit 350; and an activity-corrected image reconstruction unit 370. The filtering unit 330 includes a smoothing filter 331 and a parameter adjustment unit 332 for adjusting the smoothing parameters of the filter. In smoothing, filters such as low-pass filters, high-pass filters, Gaussian filters, and bilateral filters can be used, and the filtering unit 330 uses one or more of these as smoothing filters to reduce noise. The activity information acquisition unit 350 includes: an activity determination unit 351 for determining the presence or absence of activity; and an activity detection unit 352 for detecting the magnitude and direction of activity.

[0058] The following is for reference Figure 3B as well as Figure 5 This section will explain in detail the active correction image reconstruction processing of each part of the image processing unit 302.

[0059] First, the image pair generation process is performed (S31).

[0060] <Step S51>

[0061] Image generation unit 310 according to shooting step S1 ( Figure 3A The transmitted X-ray data collected by the data collection device 106 is used to generate two images, namely a first image and a second image, through a filter-corrected backprojection method. The two images constitute an image pair facing each other with the reconstructed image position of the target as the center.

[0062] exist Figure 6 The image shows the relationship between the reconstructed image location of the target and the first and second images. Figure 6In this example, position 600 represents the target image reconstruction center position set in reconstruction condition setting step S2. The position where the X-ray tube and the subject 3 face each other (the position where the scanner's rotation angle is 0°) is the target image reconstruction center position. In the reconstruction of the tomographic image, using transmitted X-ray data centered at this position 600 (0°), within a given angular range, for example, a range greater than 180°, the image pair is generated using transmitted X-ray data centered at positions of -90° and +90°, respectively, as the image reconstruction ranges. That is, image reconstruction ranges 601 and 602 are of equal size, each with an angle less than 180°, and these image reconstruction center positions are 180° apart from each other. This angular range can be set to a default value or can be set / changed by the user in reconstruction condition setting step S2.

[0063] Furthermore, the first image and the second image are not limited to being a single two-dimensional image, but rather become a three-dimensional image composed of multiple two-dimensional images.

[0064] <Step S52>

[0065] This step is a process used to set a smoothness that matches the level of noise when reducing noise in the first and second images by filtering. First, the parameter adjustment unit 332 obtains the tube current, which is an indicator of the level of noise (noise amount).

[0066] Specifically, tube current information is obtained from the X-ray data collected in the imaging step S1. There are cases where the tube current is kept constant during imaging, and cases where the tube current varies according to imaging conditions. For example, in imaging the heart, to suppress radiation dose, the tube current in the stationary phase and other phases used to create the reconstructed image is sometimes changed based on the electrocardiogram waveform data at the time of imaging.

[0067] In Figure 7 ( Figures 7A to 7C The diagram illustrates an example of the relationship between ECG waveform data (700) and the magnitude of the tube current values ​​(710 and 720) under varying tube current conditions. Figure 7A In the ECG waveform 700 shown, when a target heart phase 701 is set between adjacent R waves, the tube current is as follows: Figure 7B As shown, the center of its waveform is set to be the target heart phase 701.

[0068] Here, when the heart phase (i.e., the target image reconstruction position) 711 selected in the condition setting step S2 is consistent with the target heart phase 701, the tube current values ​​in the image reconstruction range 712 of the first image and the image reconstruction range 713 of the second image, which are equidistant (equidistant time intervals) from position 711, are equal. Figure 7C As shown, when the heart phase (target image reconstruction position) 721 is selected during reconstruction and deviates from the target heart phase 701 to a certain extent, the tube current value in the image reconstruction range 722 of the first image and the tube current value in the image reconstruction range 723 of the second image are inconsistent.

[0069] Since the tube current information acquired at the time of acquisition is included as supplementary information in the X-ray data, the parameter adjustment unit 332 uses the tube current information acquired at the time of acquisition of the first and second images to calculate the average tube current value and average tube current ratio of the first and second images acquired in the image pair generation step S51. The average tube current value is... Figure 6 The average value of the varying tube currents in the image reconstruction ranges 601 and 602 shown is the average tube current ratio, which is the value obtained by dividing the average tube current value of the first image by the average tube current value of the second image (or its reciprocal).

[0070] <Step S53>

[0071] The parameter adjustment unit 332 uses the average tube current value and average tube current ratio of the first and second images obtained in the tube current acquisition step S52 to determine the smoothing parameters of the filter used in the subsequent filtering process S54.

[0072] Specifically, when the average tube current ratio of the first image and the second image is within a certain range centered at 1, for example when the ECG phase 700 (target image heart phase 701) and tube current 710 are within a certain range... Figure 7B In that case, the same smoothing parameter is set for both the first and second images based on the information of the average tube current value. A certain range can be determined, for example, within the range of an average tube current ratio of 0.9 to 1.1.

[0073] On the other hand, when the average tube current ratio between the first image and the second image is outside a certain range centered at 1, for example, when the ECG phase 700 (target image heart phase 701) and tube current 720 are... Figure 7C In such a relationship, the same or different smoothing parameters are set for the first image and the second image based on one or more pieces of information, such as the average tube current value and the average tube current ratio.

[0074] In Figure 8 ( Figure 8A , Figure 8B The diagram illustrates parameter adjustments based on both the average tube current value and the average tube current ratio. Figure 8AAs shown, when using the average tube current value of 800 for both the first and second images, the larger this value, the more the smoothing parameters are adjusted to weaken the smoothing. That is, since a larger average tube current value results in less noise, and a smaller average tube current value results in more noise, when the average tube current values ​​of the first and second images are different, the smoothing degree of the side with the larger average tube current value is reduced, and the smoothing degree of the side with the smaller average tube current value is enhanced.

[0075] like Figure 8B As shown, when using the average tube current ratio 801 between the first image and the second image, the further the value moves from 1 in the increasing direction, the more the smoothing parameter is adjusted in the direction of enhancing the smoothing of the image that serves as the reference; the further the value moves from 1 in the decreasing direction, the more the smoothing parameter is adjusted in the direction of enhancing the smoothing of images other than the image that serves as the reference.

[0076] By using the average tube current value or average tube current ratio in this way, noise reduction that reflects the relationship between noise in the image pair can be achieved. Furthermore, regardless of whether the average tube current value or average tube current ratio of the first and second images is used, an upper limit is set for the smoothing parameters. This prevents excessive smoothing.

[0077] In addition, in Figure 8, the average tube current value and the smoothing intensity, and the average tube current ratio and the smoothing intensity are linearly correlated, but they can also be set as nonlinear correlations.

[0078] Furthermore, in Figure 8, the smoothing parameters that should be adjusted are generally represented as the "smoothing intensity" (vertical axis), but the smoothing parameters vary depending on the filter used. As an example, the parameters that can be adjusted when using a bilateral filter are illustrated.

[0079] The bilateral filter is characterized by the following equation (1), where the three parameters w, σ1, and σ2 determine the strength of the smoothing. The strength of the smoothing is determined by adjusting one or more of these parameters. To enhance smoothing, the values ​​of w, σ1, and σ2 are increased; to weaken smoothing, the values ​​of w, σ1, and σ2 are decreased.

[0080]

Mathematical Formula 1

[0081]

[0082] In the formula, f(i,j) represents the arrangement of the input image data, g(i,j) represents the arrangement of the output image data, w represents the size of the kernel, σ1 represents the weight considering the distance to the object voxel, and σ2 represents the weight considering the difference between the pixel values ​​of the object voxel.

[0083] <Step S54>

[0084] In this step, the filter of the parameters set by the parameter adjustment unit 332 is applied to the first image and the second image obtained in the image pair generation step S51 to obtain the filtered first image and the second image.

[0085] exist Figure 9 The results of filtering a first image and a second image with significantly different average tube current values ​​are shown. As shown, the first image 900 and the second image 901 before filtering have significantly different noise tendencies due to the large difference in average tube current values. By applying optimized parameters to the first image and the second image respectively, the noise tendencies of the first image 910 and the second image 911 after filtering are similar.

[0086] Through the above steps S51 to S54 Figure 3B The image pair generation step S31 is completed.

[0087] Next, the filtered image pairs are used to detect activity (activity information acquisition step S32). In activity information acquisition step S32, before activity detection, the difference between the two images is used to determine whether activity exists. This will be explained in detail below.

[0088] <Step S55>

[0089] In this step, firstly, image differences are obtained from the first and second images acquired in S51, or from the filtered first and second images in S54. As explained in S51, in the case of a three-dimensional image, the first and second images are generated for each of multiple cross-sections. That is, multiple image pairs and multiple difference images are obtained. The standard deviation of pixel values ​​is calculated for each of all the obtained difference images, and the median value of the standard deviation of all difference images is set as the representative value of the standard deviation of the image pair. In other words, the representative value of the standard deviation becomes an indicator of the overall activity of the obtained three-dimensional image.

[0090] <Step S56>

[0091] The presence or absence of activity in the subject is determined by calculating the standard deviation obtained in step S55 based on the aforementioned standard deviation. Specifically, if the standard deviation is above a threshold, it is determined that there is activity in the subject; if the standard deviation is below the threshold, it is determined that there is no activity in the subject.

[0092] exist Figure 10 This illustrates an example of determining the presence or absence of activity in a subject that is not currently in motion. Figure 10In the diagram, the difference image 1010 is a difference image between the filtered first image 1000 and the second image 1001. As shown, the pixel values ​​of the difference image 1010 are inherently low, and the calculated standard deviation is below a threshold. Therefore, it is determined that there is no activity in the subject.

[0093] <Step S57>

[0094] Based on the result of the determination of the presence or absence of activity (S56), the pixel values ​​of the filtered first and second images are normalized. Normalization may be performed on all pixels of the first and second images using the Min-Max method, but different processing is applied according to the determination result in step S56.

[0095] For images determined to be active in the subject, the input image is normalized to a range with a minimum value of 0 and a maximum value of 1 by applying equation (2).

[0096] For images determined to be inactive within the subject, the denominator (f) of equation (2) is adjusted by applying a formula to the input image. max -f min Increase to a predetermined fixed value T (T≥f) max -f min The formula is used to normalize the range to a minimum value of 0 and a maximum value of M (0 < M ≤ 1).

[0097]

Mathematical Formula 2

[0098]

[0099] f(i, j): Arrangement of input image data; g(i, j): Arrangement of output image data.

[0100] f min : The minimum pixel value of the input image, f max The maximum pixel value of the input image.

[0101] By determining the presence or absence of activity in this way, and adjusting the normalization method based on the determination result (i.e., adjusting to make the normalized value smaller in the absence of activity), the intensity of activity compensation is adjusted. Thus, even in the absence of activity, excessive activity compensation can be suppressed.

[0102] <Step S58>

[0103] After completing steps S55 to S57 above, normalized image pairs are used to detect activity. Figure 3B (S32). This process is similar to existing activity detection techniques, performing non-rigid alignment of two images and calculating the activity vector between the images.

[0104] <Step S59>

[0105] Finally, the image reconstruction unit 301 uses the activity vector obtained in step S58 and the shooting step ( Figure 3A Image reconstruction is performed using X-ray data collected in S1.

[0106] The motion-corrected image reconstruction uses the motion vectors calculated from the first and second images taken at a position 180° away from the target reconstruction position to estimate the magnitude and direction of the motion of the subject when the X-ray data used in the image reconstruction were acquired. The tomographic image is then reconstructed by back-projecting the image at the target reconstruction position based on this information.

[0107] By performing this processing on each of multiple cross sections, 3D tomographic image data with active correction is obtained. The obtained tomographic images are stored in storage device 213 as needed and displayed on display device 211.

[0108] As explained above, according to this embodiment, by performing noise reduction on the image pairs used for detecting activity that reflects the relationship between the tube currents when these images were acquired, problems such as misidentifying the noise difference between the two images as an active component are prevented, and the loss of active components caused by excessive smoothing is prevented, thereby improving the accuracy of activity correction.

[0109] Furthermore, according to this embodiment, even when the tube current values ​​of the first and second images are unknown at the image reconstruction condition setting time point, the tube current values ​​are obtained from the accompanying information of the transmitted X-ray data and reflected in the noise reduction. Therefore, even when the tube current value changes, noise reduction corresponding to each change can be performed, thereby suppressing the reduction in activity correction accuracy and reducing unnatural image distortion.

[0110] Furthermore, according to this embodiment, by determining whether the subject is active or not, and changing the conditions for normalizing the two images according to the degree of activity (presence or absence), it is possible to eliminate problems such as excessive correction even when the activity is small, which would result in unnatural distortion caused by image noise.

[0111] In addition, this embodiment is an example of performing (1) noise reduction reflecting the relationship of tube current and (2) activity correction reflecting the determination result of the presence or absence of the subject's activity, but implementing only one of the two means is also included in the present invention.

[0112] [Implementation Method 2]

[0113] In Implementation 1, the image reconstruction position (reconstructing the heart phase) is set as the target in the image reconstruction condition setting S2, thereby determining the position of the image pair used in the active correction. Therefore, noise reduction processing is performed using the tube current values ​​of the actually acquired first and second images.

[0114] In contrast, in this embodiment, during the step of generating image pairs (Figure 3: S31), a range suitable for motion correction (possible motion correction range) is determined, and the reconstructed cardiac phase is set within this range. This reduces the risk of decreased motion correction accuracy and unnatural distortion in advance.

[0115] In this embodiment, the structure of the image processing unit 302 is also similar to... Figure 4 The embodiment shown is the same as 1, but as Figure 11 As shown, an additional tube current calculation unit 340 and a reconstructed cardiac phase setting unit (cardiac phase setting unit) 360 are added. The tube current calculation unit 340 can be common to the parameter adjustment unit 332 of Embodiment 1 that performs tube current calculation.

[0116] The following is for reference Figure 12 The process flow of this embodiment will be explained.

[0117] <Step S41>

[0118] The tube current calculation unit 340 obtains tube current information from the transmitted X-ray data collected in the imaging step S1. The obtained tube current information is used as... Figure 7A The diagram shows the tube current as shown, where the tube current is supplied so that it is high in a given phase range that includes the phase of the target heart being photographed.

[0119] Using the tube current information, the tube current calculation unit 340 calculates the average tube current value for two images directly opposite each other within a given cardiac phase range at regular intervals. Furthermore, the ratio of their average tube currents can also be calculated. The average tube current value and the average tube current ratio are the same as those obtained in the tube current calculation step S52 of Embodiment 1.

[0120] <Step S42>

[0121] Next, the cardiac phase setting unit 360 uses at least one of the average tube current value or the average tube current ratio calculated by the tube current calculation unit 340 to predict whether a certain effect can be obtained through active correction. Specifically, for example, if the average tube current value of an image pair within a certain cardiac phase range is above a preset threshold, that cardiac phase range (cardiac phase) is determined to be a cardiac phase that can be actively corrected. Furthermore, if the average tube current value is lower than the threshold, it is determined to be a cardiac phase that cannot be actively corrected. By performing such determinations while staggering cardiac phase ranges at given intervals, it is finally determined whether active correction can be performed over the entire range of the acquired transmitted X-ray data. When using the average tube current ratio, it is also possible to determine whether active correction can be performed or not based on whether the value is close to 1 (e.g., whether the average tube current ratio is within the range of 0.9 to 1.1 or outside the range).

[0122] exist Figure 13 An example of the result obtained from such a determination is shown. Figure 13 In the region 1300 where the reconstructed cardiac phase should be set, region 1301, shown in white, is determined to be a region where activity correction can be performed, and region 1302, shown in gray, is determined to be a region where activity correction cannot be performed, corresponding to the regions of the cardiac phase respectively.

[0123] <Step S43>

[0124] The cardiac phase setting unit 360 determines the reconstructed cardiac phase (image reconstruction position) based on the determination result of determination step S42, and sets it as a reconstruction condition. Figure 3A (S2). The decision to reconstruct the cardiac phase can be made automatically based on the judgment result, or it can be set when the reconstruction conditions are set. Figure 13 The screen 130, as shown, includes the tube current value 1304, the possible area for active correction 1301, and the impossible area for active correction 1302, along with a GUI (Guideline for Reconstructing the Heart Phase) 1305, which allows the user to set the reconstructed heart phase at any position by operating the GUI 1305. Alternatively, the screen can display an electrocardiogram (ECG) and waveform acquired parallel to the camera, aligned on the time axis. This allows the user to specify a suitable reconstructed heart phase even when the target heart phase and tube current deviate during imaging.

[0125] <Step S44>

[0126] The image generation unit 310 selects the image based on the set reconstructed cardiac phase (image reconstruction position). Figure 6 The regions of the first image and the second image shown are used to generate the first image and the second image.

[0127] <Step S45>

[0128] Image processing after generation and Figure 5 The illustrated implementation 1 similarly performs filter parameter adjustment (S53), filtering processing (S54), calculation of the standard deviation of pixel values ​​(S55), determination of the presence or absence of activity (S56), normalization of pixel values ​​(S57), and calculation of the activity vector (S58), etc. However, regarding... Figure 5 The acquisition of tube current information (S52) in the processing can be omitted because the average tube current value or average tube current ratio has already been calculated in step S41 above. The value calculated from the image used in the activity detection can be used.

[0129] According to this embodiment, the image pairs used in the activity correction can be adapted during the reconstruction condition setting stage, thereby improving the effectiveness of the activity correction.

[0130] In the above implementation, the tube current value is used as an indicator of noise level, but information that influences the noise bias of the image pair can also be used. For example, the tube voltage value can be used to calculate the noise level indicator when acquiring X-ray data, just like the tube current. Furthermore, since the noise bias changes depending on the reconstruction method, reconstruction filter function, detector arrangement, etc., one or more of these factors can be incorporated to calculate the noise level indicator. Here, the reconstruction method refers to filter-corrected backprojection or successive approximation. The reconstruction filter function varies depending on whether it is for the heart, the lung region, or the head. The detector arrangement refers to the number of detector columns (16 columns, 64 columns, 256 columns, etc.).

Claims

1. An X-ray CT device, characterized in that, have: The imaging unit has an X-ray source that rotates around the subject and an X-ray detector, which acquires transmitted X-ray data of the subject within a given angular range; An image reconstruction unit that generates a reconstructed image using the transmitted X-ray data acquired by the imaging unit; and Image processing department, The image processing unit includes: An image pair generation unit uses a portion of the transmitted X-ray data to generate image pairs of opposite positions; The noise reduction unit performs noise reduction on each image pair generated by the image pair generation unit. and The activity information acquisition unit uses noise-reduced image pairs to acquire activity information of the subject during scanning. The noise reduction unit performs noise reduction on each image pair based on the relationship between the noise levels of the images contained in the image pair. The image reconstruction unit uses the motion information calculated by the motion information acquisition unit to correct for the motion of the subject during scanning and generate a reconstructed image. The noise reduction unit includes a filter and a parameter adjustment unit for adjusting the smoothing parameters of the filter. The parameter adjustment unit adjusts the parameters based on at least one of the average tube current value obtained from transmitted X-ray data for each image of the image pair and the ratio of the average tube current values. The parameter adjustment unit adjusts the smoothing degree of the image with the smaller average tube current value when it was acquired using transmitted X-ray data to be higher than that of the image with the larger average tube current value.

2. The X-ray CT apparatus according to claim 1, characterized in that, The noise reduction unit uses the tube current at the time of acquisition of each transmitted X-ray data used in the generation of the image pair as an indicator of the noise level.

3. The X-ray CT apparatus according to claim 1, characterized in that, The activity information acquisition unit includes an activity determination unit that uses image pairs with reduced noise to determine the presence or absence of an activity, and performs normalization processing on the image pairs based on the result of the activity determination unit.

4. The X-ray CT apparatus according to claim 3, characterized in that, The activity determination section calculates difference images for multiple image pairs and uses the standard deviation of pixel values ​​in each difference image to determine the presence or absence of activity.

5. The X-ray CT apparatus according to claim 1, characterized in that, The X-ray CT device also features: The cardiac phase setting unit sets the reconstructed cardiac phase during synchronized electrocardiogram (ECG) imaging. The cardiac phase setting unit calculates at least one of the average tube current value and average tube current ratio of each image in the image pair generated from transmitted X-ray data, covering a range of multiple cardiac phases, and determines possible cardiac phase regions for activity correction based on the range of cardiac phases calculated using the calculated results.

6. The X-ray CT apparatus according to claim 5, characterized in that, The cardiac phase setting unit sets the reconstructed cardiac phase in the determined cardiac phase region where activity correction is possible.

7. The X-ray CT apparatus according to claim 5, characterized in that, The cardiac phase setting unit displays the identified possible cardiac phase regions for activity correction on the display device and accepts the user's settings for reconstructed cardiac phase.

8. An image processing apparatus that receives transmitted X-ray data collected by an X-ray CT device and performs motion-corrected image reconstruction, characterized in that, The image processing device includes: An image pair generation unit uses a portion of the transmitted X-ray data to generate image pairs of opposite positions; The noise reduction unit reduces noise by adjusting the smoothness of the image pairs generated by the image pair generation unit based on the relationship between the index of the noise amount of each image in the image pair. The activity information acquisition unit uses noise-reduced image pairs to acquire activity information of the subject being scanned; and The image reconstruction unit generates a CT image that corrects for the movement of the subject during scanning using the movement information calculated by the movement information acquisition unit. The noise reduction unit includes a filter and a parameter adjustment unit for adjusting the smoothing parameters of the filter. The parameter adjustment unit adjusts the parameters based on at least one of the average tube current value obtained from transmitted X-ray data for each image of the image pair and the ratio of the average tube current values. The parameter adjustment unit adjusts the smoothing degree of the image with the smaller average tube current value when it was acquired using transmitted X-ray data to be higher than that of the image with the larger average tube current value.

9. A motion-corrected image reconstruction method, using transmitted X-ray data to correct for the motion of the subject during scanning, to reconstruct CT images, characterized in that... A pair of images is generated using a portion of the transmitted X-ray data. Noise reduction is performed on the pair of images separately. The smoothness of the filter during noise reduction is adjusted based on the relationship between the noise levels of the individual images within the pair. The activity information of the subject is obtained from a pair of images after noise reduction. Reconstruction is performed using the activity information and the transmitted X-ray data. The smoothing parameters of the filter are adjusted based on at least one of the average tube current value obtained from transmitted X-ray data for each of the pair of images and the ratio of the average tube current value. The smoothing degree of the image with the smaller average tube current value when it was acquired using X-ray transmitted data is adjusted to be higher than that of the image with the larger average tube current value.

10. The active correction image reconstruction method according to claim 9, characterized in that, A difference image is generated between a pair of images after noise reduction. The standard deviation of the pixel values ​​in this difference image is used to determine whether the subject is active. The pair of images is normalized based on the determination result.

Images