X-ray CT apparatus, image processing apparatus, and motion corrected image reconstruction method

Abstract

The invention provides an X-ray CT apparatus, an image processing apparatus, and a motion correction image reconstruction method, and aims to suppress misrecognition of motion components and excessive smoothing and improve motion correction precision when motion information of a shot object in scanning is acquired from an image pair and motion correction image reconstruction is performed to generate a CT image. The motion information acquisition unit is provided with: a filter unit that reduces noise of the image pair; a registration unit that performs registration processing using the filtered image pair; and a processing condition adjustment unit that adjusts the conditions of the filter processing and the registration processing on the basis of the tube current ratio of the image pair. The processing condition adjustment unit refers to a table in which a relationship between a tube current ratio and a parameter for each process is preset, selects a parameter corresponding to the tube current ratio of the target image pair, and specifies a parameter for filtering and registration. The motion information acquisition unit performs filtering and registration using the determined parameters and acquires motion information, and the image reconstruction unit performs motion-corrected image reconstruction using the motion information.

Description

Technical Field

[0001] The present invention relates to an X-ray CT apparatus for obtaining medical images by irradiating an object with X-rays, and to a motion correction image reconstruction processing technology for improving the image reading accuracy and reading efficiency of a moving object. Background Art

[0002] In a CT scan of a moving object such as a heart, motion artifacts may be generated in the image due to the movement of the object during the CT scan. The image quality is degraded due to the motion artifacts, which may result in a reduction in the accuracy and efficiency of the diagnosis of the disease by doctors or inspection technicians (hereinafter collectively referred to as inspectors). Therefore, a motion correction image reconstruction process is performed to reduce the motion artifacts generated in the CT image of the moving object.

[0003] In the motion correction image reconstruction process, the motion of the subject is estimated from a pair of images (first image and second image) reconstructed at temporally opposite positions with the target image reconstruction position as the center, and the image at the target reconstruction position is corrected using the estimated motion information while back-projection is performed to perform reconstruction. By applying this motion correction image reconstruction, it is possible to correct the motion of the heart that cannot be solved by matching the conditions of the cardiac phase alone, and the motion of organs in ECG asynchronous photography, etc.

[0004] However, if noise exists in the two images used for estimating motion, there are problems such as misidentifying the noise as motion or failing to detect motion mixed in the noise. Patent document 1 discloses a method for extracting only the motion 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 by applying a low-pass filter according to the amount of X-rays when the first image and the second image are acquired.

[0005] Patent Document 1: U.S. Patent No. 10165989

[0006] In the method disclosed in Patent Document 1, although the noise is reduced based on the noise of each of the first image and the second image, the correlation of the noise between the images is not considered. The noise increases when the tube current is low when acquiring the transmission X-ray data, but if the difference in the tube current when acquiring the image pair is large, excessive smoothing is performed due to the influence of the noise component in the low-dose area, resulting in distortion in the fringe component and other problems. On the other hand, if excessive correction such as excessive smoothing is excessively suppressed, the phase that can be static will be affected. Summary of the invention

[0007] Therefore, an object of the present invention is to suppress the erroneous recognition and excessive smoothing of the above-mentioned motion components and to improve the accuracy of motion correction.

[0008] In order to solve the above-mentioned problems, the present invention adjusts the processing conditions performed when acquiring motion information for a pair of images, a first image and a second image (hereinafter also referred to as an image pair), according to the tube current ratio when each image of the image pair is acquired. The main processing performed when acquiring motion information is image noise reduction processing and registration processing after noise reduction, and the adjustment of the processing conditions can be performed by adjusting the type of algorithm used in these processing or the parameters included therein.

[0009] 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 acquires transmission X-ray data of the subject within a specified angle range; and an image reconstruction unit that generates a reconstructed image using the transmission X-ray data acquired by the imaging unit. The image reconstruction unit comprises: an image pair generation unit that generates a first image and a second image at a facing position using a portion of the transmission X-ray data; and a motion information acquisition unit that performs noise reduction processing on the first image and the second image, and registration processing on the first image and the second image after noise reduction, and calculates motion information of the subject being scanned. The image reconstruction unit uses the motion information calculated by the motion information acquisition unit to correct the motion of the subject being scanned to generate a reconstructed image.

[0010] According to a first embodiment of the present invention, the motion information acquisition unit includes a processing condition adjustment unit, which classifies image pairs consisting of a first image and a second image according to a tube current ratio when transmission X-ray data for generating the first image and the second image are acquired, and adjusts the conditions of noise reduction processing and registration processing for each category.

[0011] Furthermore, according to the second aspect of the present invention, the motion information acquisition unit includes a processing condition adjustment unit that adjusts parameters of the non-rigid registration algorithm according to a tube current ratio when transmission X-ray data for generating the first image and the second image are acquired.

[0012] The image processing device of the present invention is a device having the function of the image reconstruction unit of the above-mentioned X-ray CT apparatus.

[0013] Furthermore, in the motion correction image reconstruction method of the present invention, the motion of the subject being scanned is corrected using the transmission X-ray data to reconstruct the CT image, and the method includes the following steps: using a part of the transmission X-ray data to generate a first image and a second image at a facing position; performing noise reduction processing and registration processing after noise reduction on each of the first image and the second image, and acquiring the motion information of the subject; and reconstructing the image using the motion information and the transmission X-ray data. In the step of acquiring the motion information, at least one of the type and parameters of the noise reduction filter used in the noise reduction processing and the parameters of the non-rigid registration algorithm used in the registration processing are adjusted according to the tube current ratio when the first image and the second image are acquired.

[0014] Effects of the Invention

[0015] According to the present invention, when reconstructing a motion-corrected image using motion information acquired from an image pair, it is possible to prevent excessive correction caused by differences in tube current when each image pair is acquired, thereby obtaining an image in which the motion of the subject is appropriately corrected and the influence on the stationary portion is minimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a diagram showing the overall structure of the X-ray CT apparatus of the present invention.

[0017] Figure 2 It is a diagram showing the constituent elements of the X-ray CT apparatus of the present invention.

[0018] Figure 3 This is a functional block diagram of an image reconstruction unit according to an embodiment.

[0019] Figure 4 This is a diagram showing the processing flow of the X-ray CT apparatus according to the first embodiment.

[0020] Figure 5 It is a diagram showing the relationship between the first image and the second image with respect to the target reconstruction position.

[0021] Fig. 6A This is a diagram showing the relationship between electrocardiogram waveform data and the magnitude of tube current value in electrocardiogram-synchronized imaging, and is a diagram showing the electrocardiogram waveform and the target imaging cardiac phase.

[0022] Figure 6B This is a diagram showing the relationship between electrocardiogram waveform data and the magnitude of tube current value in electrocardiogram-synchronized imaging, and is a diagram showing the relationship between the imaging cardiac phase and the tube current.

[0023] Figure 6C This is a diagram showing the relationship between electrocardiogram waveform data and the magnitude of tube current value in electrocardiogram-synchronized imaging, and is a diagram showing the relationship between the imaging cardiac phase and the tube current.

[0024] Figure 7 It is a diagram illustrating the tube current ratio.

[0025] Figure 8 This is a diagram showing an example of a table in which the relationship between the tube current ratio and the parameter set according to the first embodiment is set.

[0026] Fig. 9 : is a diagram showing an example of the characteristics of the adjusted filter and the parameters of the non-rigid registration.

[0027] Fig.10 This is a diagram showing the details of parameter adjustment in the first embodiment.

[0028] Fig.11 This is a diagram showing the processing of implementation mode 2.

[0029] Fig.12 This is a diagram for explaining the determination of the presence or absence of motion.

[0030] Explanation of symbols

[0031] 1-X-ray CT device, 3-object, 10-camera unit, 20-operating unit, 30-calculating unit, 100-gantry, 101-bed device, 102-X-ray generating device, 103-X-ray detecting device, 104-collimator device, 105-high voltage generating device, 107-driving device, 106-data collecting device, 200-central control device, 210-input and output device, 211-display device, 212-input device, 213-storage device, 310-image reconstruction unit, 320-image pair generating unit, 330-motion information acquiring unit, 335-processing condition adjusting unit, 340-motion correction image reconstruction unit. DETAILED DESCRIPTION

[0032] Hereinafter, embodiments of the present invention will be described using the drawings.

[0033] First, the overall structure of an X-ray CT apparatus to which the present invention is applied will be described.

[0034] like Figure 1 As shown, the X-ray CT apparatus 1 includes an imaging unit 10 for capturing a tomographic image and a fluoroscopic image of a subject 3 , and including a gantry 100 and a bed device 101 ; and an operating unit 20 for operating and controlling the imaging unit 10 .

[0035] like Figure 2As shown, the gantry 100 is provided with: an X-ray generator 102 equipped with an X-ray tube for generating X-rays to be irradiated to the subject 3; a collimator device 104 for reducing the beam of X-rays generated by the X-ray generator 102; an X-ray detector 103 for detecting X-rays transmitted through the subject; a scanner 108 equipped with the above devices; a high voltage generator 105 for applying a high voltage to the X-ray generator 102; a data collection device 106 for collecting the transmitted X-ray data obtained by the X-ray detection device 103; and a drive device 107 for rotating the scanner around the subject 3. Although not shown, the X-ray generator 102 is provided with an X-ray tube, and irradiates the subject 3 with a predetermined amount of X-rays by passing a predetermined tube current through the X-ray tube. In the case of ECG synchronous photography, photography is performed based on information from an electrocardiograph (not shown) worn on the subject 3, and in the case of ECG asynchronous photography, photography is performed under automatic exposure control.

[0036] The operation unit 20 includes a central control unit (CPU) 200 for controlling each device built into the frame, and an input / output device 210 that functions as a user interface for user interaction with the CPU 200, and a computing unit 30 is mounted in the CPU 200, and the computing unit 30 performs various operations such as image reconstruction on the transmission X-ray data collected by the data collection device 106. However, a computing device different from the CPU 200 may be provided, and the computing device may function as the computing unit 30. The functions of the CPU 200 are realized by the CPU 200 reading and executing a program that records a computing algorithm or a control processing sequence, but part of the computing or processing performed by the computing unit 30 may also be performed using a PLD (programmable logic device) such as an ASIC or FPGA.

[0037] The input / output device 210 includes an input device 212 for the operator to input imaging conditions, etc., a display device 211 for displaying data such as imaging images or a GUI, and a storage device 213 for storing data required for imaging such as programs and device parameters.

[0038] The CPU 200 controls the imaging unit 10 (X-ray generating device 102, X-ray detecting device 103, high voltage generating device 105, collimator device 104, bed device 101, driving device 107, data collecting device 106), the input / output device 210, and the computing unit 30 according to the operation instruction from the operator via the input device 212. Under the control of the CPU 200, these units operate to reconstruct the CT image and correct the reconstructed CT image.

[0039] In the case of ECG synchronous photography, information from an electrocardiograph (not shown) worn on the subject 3 is sent to the CPU 200, and the CPU 200 controls the high voltage generator 105 in synchronization with the phase of the heart movement, and controls the tube current supplied to the X-ray generator (X-ray tube). In addition, in the case of ECG asynchronous photography, an automatic exposure control function is provided to automatically control the tube current. The dose is adjusted by controlling the tube current to prevent continuous high doses, thereby reducing unnecessary radiation.

[0040] The operation unit 30 includes an image reconstruction unit 310 that performs back-projection processing on the transmission X-ray data obtained by the data collection device 106 to generate a tomographic image. In addition to normal image reconstruction, the image reconstruction unit 310 also includes motion correction image reconstruction in the imaging of a moving part. In the imaging of a moving part, in addition to imaging synchronized with the phase of the movement of the heart (cardiac phase) such as ECG-synchronized imaging, imaging of organs that are moving even though they are ECG-asynchronous is also included.

[0041] In order to perform motion-corrected reconstruction, e.g. Figure 3 As shown, the image reconstruction unit 310 includes: an image pair generation unit 320, which uses a part of the transmitted X-ray data to generate a first image and a second image at a facing position (hereinafter, the pair consisting of the first image and the second image is also referred to as an image pair); a motion information acquisition unit 330, which uses the image pair to extract the motion of the subject; and a motion correction image reconstruction unit 340.

[0042] The motion information acquisition unit 330 performs a process for reducing the noise contained in the image pair (noise reduction process) and a registration process of the image pair after noise reduction, and acquires motion information. At this time, the motion information acquisition unit 330 acquires the tube current information when the transmission X-ray data of the image pair is acquired by the imaging unit 10, calculates the tube current ratio of the two transmission X-ray data, and uses the calculated ratio to adjust the conditions of the noise reduction process and the registration process. The adjustment method will be described in detail in the embodiment described later, but there are the following methods: classifying the image pair according to the tube current ratio and selecting a pre-set condition according to the classification; and adjusting the type or parameter of the algorithm used in each process according to the tube current ratio.

[0043] Hereinafter, an embodiment of the processing of the X-ray CT apparatus when performing motion correction image reconstruction will be described.

[0044] [Implementation Method 1]

[0045] like Figure 3As shown in FIG. 1 , the motion information acquisition unit 330 of the present embodiment includes a filtering unit 331 that performs a noise reduction process, a registration unit 333 that performs a registration process, and a processing condition adjustment unit 335 that adjusts the conditions of each process according to the tube current ratio of the image pair. The processing condition adjustment unit 335 includes a tube current ratio calculation unit that calculates the tube current ratio of the image pair, and a table that predetermines the relationship between the tube current ratio and the conditions of the filtering process and the registration process.

[0046] In the present embodiment, as an example, a case where processing conditions, in particular, parameters are adjusted for both the noise reduction process and the registration process will be described.

[0047] Below, reference Figure 4 The processing flow of this embodiment is described.

[0048] <Photography: S41>

[0049] The subject 3 is placed on the bed device 101 and positioning photography is performed. Positioning photography is photography for setting the imaging range of the subject 3, and a transmission X-ray image is acquired along the body axis direction while changing the relative position of the scanner 108 and the bed device 101 (subject 3). The inspector sets the imaging range using the transmission X-ray image. Next, the imaging unit 10 performs tomography with the rotation of the scanner 108 for the imaging range set based on the positioning image, and collects transmission X-ray data of the subject.

[0050] <Image reconstruction condition setting: S42>

[0051] The image reconstruction conditions for the transmitted X-ray data of the object acquired in the photographing step S41 are set. The image reconstruction conditions include, for example, the thickness of the image (thickness of the cross section), FOV, filter conditions, etc., and in the case of ECG synchronous photography, the settings of the reconstructed cardiac phase (target reconstructed cardiac phase: which cardiac phase's image is to be reconstructed) and the like. In ECG synchronous photography, the target reconstructed image position is determined by setting the target reconstructed cardiac phase. In the case of ECG asynchronous photography, the target reconstructed image position is determined by the user. The image reconstruction unit 310 receives these image reconstruction conditions set by the user via the input device 212.

[0052] <Motion Correction Image Reconstruction: S43 to S47>

[0053] The image reconstruction unit 310 performs image reconstruction based on the image reconstruction conditions set in the image reconstruction condition setting step S42 using the transmission X-ray data of the object acquired in the imaging step S41 .

[0054] At this time, the motion information of the subject being photographed is acquired, and the motion is corrected to perform reconstruction (motion-corrected image reconstruction).

[0055] <Image pair generation: S43>

[0056] In the motion correction image reconstruction, first, the image pair generator 320 generates two images, namely, a first image and a second image, by a filter-corrected back projection method based on the transmission X-ray data collected by the data collection device 106 of the imaging unit 10. The two images constitute an image pair facing each other with the target reconstruction image position as the center.

[0057] In addition, the first image and the second image are not limited to being one two-dimensional image each, and may be a three-dimensional image composed of a plurality of two-dimensional images.

[0058] The relationship between the target reconstructed image position and the first image and the second image is shown in Figure 5 In. Figure 5 In the figure, position 500 represents the target image reconstruction center position set in the image reconstruction condition setting step S42. In this example, the position where the X-ray tube and the object 3 face each other (the position where the scanner rotates at an angle of 0°) is the target image reconstruction center position. In the reconstruction of the tomographic image, the transmission X-ray data of a specified angle range centered on the position 500 (0°), for example, a range of more than 180°, is used. However, the image pair is generated based on the transmission X-ray data of the image reconstruction ranges 501 and 502 centered at the positions of -90° and +90°, respectively. That is, the image reconstruction ranges 501 and 502 are equal in size, and are angle ranges less than 180°, respectively, and their image reconstruction center positions are 180° apart from each other. The angle range can be set to a default value, or it can be set or changed by the user in the image reconstruction condition setting step S42.

[0059] <Tube Current Ratio Calculation: S44>

[0060] Next, the processing condition adjustment unit 335 (tube current ratio calculation unit) obtains the tube current information when each transmission X-ray data of the first image and the second image is acquired, and calculates their ratio (tube current ratio). Specifically, the tube current information is acquired based on the transmission X-ray data collected in the imaging step S41. There are cases where imaging is performed while keeping the tube current constant, but for example, in cardiac imaging, in order to suppress the radiation dose, the tube current in the stationary phase for generating the reconstructed image and other phases is sometimes changed based on the electrocardiogram waveform data during imaging. This embodiment is applied to imaging accompanied by such a change in the tube current.

[0061] In Figure 6 ( Figure 6A to Figure 6C ) shows an example of the relationship between the electrocardiogram waveform data (600) and the magnitude of the tube current value (610 and 620) when the tube current is changed. Fig. 6AIn the electrocardiogram waveform data 600 shown in FIG. 1 , when a photographing target cardiac phase 601 is set between adjacent R waves, as shown in FIG. Figure 6B As shown, the tube current is set so that the center of its waveform becomes the imaging target cardiac phase 601.

[0062] Here, when the cardiac phase selected at the time of reconstruction (i.e., the target image reconstruction position) 611 set in the image reconstruction condition setting step S42 coincides with the imaging target cardiac phase 601, the tube current values ​​in the image reconstruction range 612 of the first image and the image reconstruction range 613 of the second image, which are located at equal distances (equal time intervals) from the position 611, become equal. However, if Figure 6C As shown, when the selected cardiac phase (target image reconstruction position) 621 deviates to a certain extent from the imaging target cardiac phase 601 during reconstruction, the tube current value in the image reconstruction range 622 of the first image and the tube current value in the image reconstruction range 623 of the second image do not match.

[0063] In this way, when the tube currents of the two images constituting the image pair are greatly different, the acquisition of the motion information of the subject using these images becomes inaccurate, and even if the motion correction image reconstruction is performed, it will cause overcorrection. Therefore, in this embodiment, the calculated tube current ratio is used to adjust the conditions of the filtering process and the registration process for acquiring the motion information to be described later, thereby preventing the motion correction from becoming inaccurate.

[0064] The tube current ratio calculated by the tube current ratio calculation unit is a 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. Figure 7 In the example shown, if the average tube current of the image (first image) 701 of the phase that will serve as the basis for motion correction is set to 70 mA and the average tube current of the image (second image) 702 of the corresponding phase is set to 350 mA, the tube current ratio (mA ratio) is 0.2.

[0065] The processing condition adjustment unit 335 classifies the image pairs according to the calculated tube current ratio, and adjusts the parameters of the noise reduction process S45 and the registration process S46 according to the classification. The details of the adjustment will be described later.

[0066] <Noise reduction processing: S45>

[0067] The filter unit 331 filters the first image and the second image generated in the image pair generation step S43. When filtering, a smoothing filter such as a low-pass filter, a high-pass filter, a Gaussian filter, a bilateral filter, etc. can be used, and the filter unit 331 uses one or more of them as a smoothing filter to reduce noise. At this time, the characteristics (parameters) of the filter used are adjusted according to the tube current ratio calculated in the tube current ratio calculation step S44.

[0068] The smoothing parameters to be adjusted differ depending on the filter used, and as an example, adjustable parameters when a bilateral filter is used will be described.

[0069] The bilateral filter is represented by the following formula (1), in which the three parameters w, σ1, and σ2 are parameters for determining the strength of smoothing, and the strength of smoothing is determined by adjusting one or more of the parameters. In the case of strengthening smoothing, the adjustment is made in the direction of increasing the values ​​of w, σ1, and σ2, and in the case of weakening smoothing, the adjustment is made in the direction of decreasing the values ​​of w, σ1, and σ2.

[0070] [Formula 1]

[0071]

[0072] In the formula, f(i, j) represents the arrangement of input image data, g(i, j) represents the arrangement of 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 in pixel value from the object voxel.

[0073] Here, the tube current ratio can take various values ​​depending on the time when the first image and the second image are acquired, so it is not easy to adjust these parameters for all tube current ratios. In this embodiment, the tube current ratio is divided into a plurality of ranges according to a threshold value in advance, and a parameter value or a function that determines the value is preset for each range. The preset parameter can be set, for example, by using statistical data such as past clinical data to obtain a value that can achieve smoothing without overcorrection and using it as the parameter value. The parameter value for each range is stored in the storage device 213 of the operation unit 30 in the form of a table, for example.

[0074] exist Figure 8In FIG. 1 , as an example, a table is shown in which the range of the tube current ratio is divided into three ranges of “mA ratio < 0.7”, “0.7 ≤ mA ratio < 0.9”, and “mA ratio ≥ 0.9” using two values ​​of 0.7 and 0.9 as thresholds, and parameters corresponding to each range are set (Table 1). In this example, a parameter set including parameters for the registration process described later is set in each range of the tube current ratio.

[0075] Furthermore, regarding filtering, in the example where σ(σ1, σ2) among the three parameters (w, σ1, σ2) shown in equation (1) is set, the parameter σ is set as follows: in "mA ratio <0.7", it becomes a nonlinear function of the tube current ratio; in "0.7≤mA ratio<0.9", it becomes a linear function of the tube current ratio; and in "mA ratio≥0.9", it becomes a fixed value. Fig. 9 is a graph showing the characteristics of the filter adjusted according to the setting. The parameter set is not limited to this example, and other parameters w can also be set for each range. Moreover, in Table 1, in the filtering conditions, the type of filter is fixed, and only the parameters are made different for each classification, but the type of filter can also be made different for each classification.

[0076] like Fig. 9 As shown, by setting the filter characteristics to be asymmetric around mA ratio=1.0, different filters are applied to the first image and the second image according to the tube current.

[0077] Regarding the adjustment of parameters when the parameters are preset for each range of the tube current ratio, Fig.10 As shown, after the tube current ratio is calculated (S44), the following processing is added: the image pairs are classified according to the range in which the tube current ratio calculated for a predetermined image pair falls (S44-1), and a preset parameter is selected for each classification (each range of the tube current ratio) (S44-2).

[0078] In this way, the image pairs are classified according to the tube current ratio, and the preset parameters are selected based on the classification, so that the parameter adjustment can be easily achieved.

[0079] <Registration Processing: S46>

[0080] After the noise reduction process performed by the filter unit, a registration process using the image pair is performed to obtain motion information. In the registration process, for example, non-rigid registration of the second image is performed with the first image as a reference, and a deformation function that minimizes the error between the first image and the second image is obtained. In this embodiment, a parameter (regularization term weight λ) for adjusting the degree of consideration of the regularization term of the cost function in the calculation of the non-rigid registration is adjusted according to the tube current ratio.

[0081] In non-rigid registration, a free-form deformation (FFD) model based on a B-spline function is used as a motion model. When a 3D FFD model is used, the deformation function is expressed by the following equation (2).

[0082] [Formula 2]

[0083]

[0084] In formula (2), T is the FFD model based on 3D B-spline, x is the voxel, j is the control point of interest, B is the third-order tensor product composed of cubic B-spline, d is the interval of the control points in the spatial region, Θ j is the displacement vector (3D) of the control point j of interest, and Θ is a group of displacement vectors (3D) of the control points representing the relationship between the reference time and the motion at each time.

[0085] The parameters at the control points of the deformation function are obtained by minimizing the dissimilarity (the following equation (3)) based on the squared difference (SSD) between the first image in the reference phase and the second image in the phase opposite to the reference phase.

[0086] [Formula 3]

[0087]

[0088] In the formula, D(θ) represents the dissimilarity based on SSD, P target (x) represents the image in the reference phase (first image), P source (x) represents an image (second image) in a phase 180° away from the reference phase.

[0089] Here, the convergence calculation for minimization is an ill-set problem with multiple conversion parameters. In order to solve it efficiently and robustly, a regularization term represented by equation (4) is introduced.

[0090] [Formula 4]

[0091]

[0092] In the formula, R(θ) represents the regularization term that penalizes the difference in parameters of spatially adjacent control points, and Kj Represents the index group that is spatially adjacent to the j-th control point.

[0093] This equation (4) is a regularization term that penalizes the difference in parameters of spatially adjacent control points. In other words, it is a term used to suppress excessive deformation in the image of interest. The cost function of non-rigid registration uses the non-similarity D(θ) represented by equation (3) and the regularization term R(θ) represented by equation (4), and is represented by equation (5).

[0094] [Formula 5]

[0095]

[0096] In the formula, λ is a weight parameter of the regularization term R(θ). In the present embodiment, the weight λ is adjusted according to the tube current ratio.

[0097] In the weight adjustment, similarly to the adjustment of the filter parameters corresponding to the tube current ratio, the weight value is predetermined for each classification based on the range of the tube current ratio, and the above-mentioned non-rigid registration operation is performed using the weight λ predetermined according to the range (classification) to which the tube current ratio of the image pair belongs. Figure 8 As shown in Table 1, the weight λ preset according to the classification is stored in the form of a table, and when the non-rigid registration is performed by the registration unit, the weight corresponding to the tube current ratio of the image pair is read from the table.

[0098] The motion vector between the images is calculated through this registration process.

[0099] <Motion Correction Image Reconstruction: S47>

[0100] In the motion-corrected image reconstruction, image reconstruction is performed using the transmission X-ray data obtained in the imaging step S41 and the motion information acquired in the above-mentioned step, that is, the motion vector.

[0101] In motion-corrected image reconstruction, the magnitude and direction of the subject's motion when acquiring each transmission X-ray data used in image reconstruction is estimated based on the motion vector calculated from the first image and the second image generated at positions 180° apart with the target reconstruction position as the center. The image at the target reconstruction position is corrected based on this information while back-projection is performed to reconstruct the tomographic image.

[0102] By performing this process on a plurality of cross sections, motion-corrected 3D tomographic image data can be obtained. The obtained tomographic image is stored in the storage device 213 as needed and displayed on the display device 211.

[0103] As described above, according to the present embodiment, by adjusting the conditions of filtering and registration performed when detecting motion for an image pair for detecting motion using the tube current ratio when each image pair is acquired, it is possible to suppress overcorrection caused by the difference in tube current of the image pair, and reduce the generation of streak artifacts or the deformation of the blood vessel shape. In particular, by adjusting the parameters of the cost function not only for filtering but also for registration, even if the tube current ratio deviates greatly from 1, it is possible to prevent the accuracy of registration from decreasing, and to obtain high-precision motion information.

[0104] Furthermore, according to the present embodiment, the ratio of the tube current of the first image to the tube current of the second image is divided into a plurality of ranges and a parameter is set in advance to adjust the ratio for each range, thereby making it possible to easily adjust the parameters.

[0105] In the above-described embodiment, both the conditions of filtering and registration are adjusted according to the tube current ratio, but the present invention also includes the case where only the parameters of the registration process are adjusted.

[0106] [Implementation Method 2]

[0107] In the first embodiment, the registration process is performed after filtering the image pair, but the feature of this embodiment is that the image pair is used to determine the presence or absence of motion before detecting motion in the registration process, and the pixel values ​​are normalized based on the determination result.

[0108] Below, reference Fig.11 The processing of this embodiment is described. Fig.11 In, with Figure 4 The same processing as shown is represented by the same symbols, and repeated description is omitted.

[0109] After filtering the image pair under conditions corresponding to the tube current ratio (S43 to S45), the difference between the image pair (the first image and the second image) is used to determine whether there is motion. The details are described below.

[0110] <Determination of whether there is movement: S451>

[0111] In this step, first, image differences are obtained for the first image and the second image acquired in the image pair generation step S43, or for the first image and the second image filtered by the noise reduction process S45. In the case where the first image and the second image are three-dimensional images, they are generated for a plurality of cross sections respectively. That is, a plurality of image pairs are obtained, and a plurality of differential images are also obtained. For all the differential images acquired, the standard deviation of the pixel value is calculated for each image, and the central value of the standard deviation of all the differential images is set as the representative value of the standard deviation of the image pair. That is, the representative value of the standard deviation becomes an indicator of the overall motion of the acquired three-dimensional image.

[0112] Next, the presence or absence of motion of the subject is determined based on the calculated standard deviation value. Specifically, if the standard deviation value is greater than a threshold value, it is determined that there is motion in the subject, and if the standard deviation value is less than the threshold value, it is determined that there is no motion in the subject.

[0113] The example of determining whether a non-moving subject has motion is shown in Fig.12 In. Fig.12 , the differential image 1010 is a differential image of the filtered first image 1000 and the second image 1001. As shown in the figure, the pixel value of the differential image 1010 itself is low, and the value of the standard deviation calculated from it is lower than the threshold value. Therefore, it is determined that there is no motion in the subject.

[0114] <Normalization: S452>

[0115] According to the result of the motion determination (S451), the pixel values ​​of the filtered first image and the second image are normalized. Regarding normalization, for example, the Min-Max method is used to perform normalization on all pixels of the first image and the second image, but different processing is performed according to the determination result in the determination step S451.

[0116] For an image determined to have motion in the subject, equation (6) is applied to the input image to normalize it within a range where the minimum value is 0 and the maximum value is 1.

[0117] For an image determined to have no motion in the subject, the denominator (f max -f min ) increases to a preset value T(T≥f max -f min ) and apply it to the input image to normalize it to a range with a minimum value of 0 and a maximum value of M (0<M≤1).

[0118] [Formula 6]

[0119]

[0120] f(i, j): the arrangement of input image data, g(i, j): the arrangement of output image data,

[0121] f min : The minimum pixel value of the input image, f max : The maximum pixel value of the input image

[0122] In this way, the presence or absence of motion is determined, and the normalization method is changed according to the determination result, that is, the strength of motion correction is adjusted so that the normalized value when there is no motion becomes smaller. Thus, even when there is no motion, excessive motion correction can be suppressed.

[0123] Thereafter, the step of performing a registration process ( S46 ) using the normalized image pair and a step of performing a motion-corrected image reconstruction ( S47 ) using parameters corresponding to the tube current ratio are the same as those in the first embodiment.

[0124] According to this embodiment, by determining whether there is motion of the subject and changing the conditions for normalizing the two images according to the degree of motion (presence or absence), it is possible to solve the problem of excessive correction even when the motion is small, causing unnatural distortion due to image noise.

[0125] In the above, the embodiments of the X-ray CT apparatus and the motion correction image reconstruction of the present invention are described. However, the present invention is characterized in that the conditions of the noise reduction process and the registration process are adjusted using the tube current ratio of the image pair for detecting motion information, and the tube current ratio is preferably divided into a plurality of ranges according to a threshold value, and a preset condition is selected and applied for each range. The present invention is not limited to these embodiments, and various modifications are possible.

Claims

1. An X-ray CT device, characterized in that: The invention comprises: an imaging unit having an X-ray source and an X-ray detector rotating around a subject and acquiring transmission X-ray data of the subject within a predetermined angle range; and an image reconstruction unit generating a reconstructed image using the transmission X-ray data acquired by the imaging unit. The image reconstruction unit comprises: an image pair generating unit that generates a first image and a second image at a facing position using a portion of the transmission X-ray data; and a motion information acquisition unit, which performs noise reduction processing on the first image and the second image, and registration processing on the first image and the second image after noise reduction, and calculates motion information of the subject being scanned, The motion information acquisition unit includes a processing condition adjustment unit that classifies an image pair consisting of the first image and the second image according to a tube current ratio when the transmission X-ray data for generating the first image and the second image are acquired, and adjusts the conditions of the noise reduction processing and the registration processing for each class. The image reconstruction unit corrects the motion of the object being scanned using the motion information calculated by the motion information acquisition unit to generate a reconstructed image.

2. The X-ray CT apparatus according to claim 1, characterized in that: The imaging unit receives information from an electrocardiograph worn on the subject and performs electrocardiogram synchronous imaging using a desired cardiac phase as a reconstructed cardiac phase. The transmission X-ray data of the first image and the transmission X-ray data of the second image are located at positions facing each other with the reconstructed cardiac phase as the center.

3. The X-ray CT apparatus according to claim 1, characterized in that: The camera unit performs asynchronous photography under automatic exposure control. The transmission X-ray data of the first image and the transmission X-ray data of the second image are located at positions facing each other relative to a reconstruction center position designated by a user.

4. The X-ray CT apparatus according to claim 1, characterized in that: The processing condition adjustment unit sets two or more threshold values ​​for the tube current ratio and classifies the image pairs into three or more categories.

5. The X-ray CT apparatus according to claim 4, characterized in that: The transmission X-ray data acquired by the imaging unit is three-dimensional data including a slice direction intersecting the rotation direction, and the image pair generating unit generates a plurality of image pairs along the slice direction.

6. The X-ray CT apparatus according to claim 1, characterized in that: The motion information acquisition unit includes a smoothing filter. The processing condition adjustment unit adjusts at least one of the type and parameter of the smoothing filter for each of the categories as a condition of the noise reduction process.

7. The X-ray CT apparatus according to claim 6, characterized in that: The processing condition adjustment unit makes a condition of the noise reduction process for the first image different from a condition of the noise reduction process for the second image.

8. The X-ray CT apparatus according to claim 1, characterized in that: The motion information acquisition unit includes a non-rigid registration algorithm, The processing condition adjustment unit adjusts the parameters of the non-rigid registration algorithm for each of the categories as a condition of the registration process.

9. The X-ray CT apparatus according to claim 1, characterized in that: The motion information acquisition unit includes a noise reduction filter and a non-rigid registration algorithm. The processing condition adjustment unit adjusts at least one of the type and parameters of the denoising filter and the parameters of the non-rigid registration algorithm as conditions of the denoising process and the registration process.

10. An X-ray CT device, characterized in that: The invention comprises: an imaging unit having an X-ray source and an X-ray detector rotating around a subject and acquiring transmission X-ray data of the subject within a predetermined angle range; and an image reconstruction unit generating a reconstructed image using the transmission X-ray data acquired by the imaging unit. The image reconstruction unit comprises: an image pair generating unit that generates a first image and a second image at a facing position using a portion of the transmission X-ray data; and a motion information acquisition unit, for performing registration processing on the first image and the second image according to a non-rigid registration algorithm, and calculating motion information of the subject being scanned, The motion information acquisition unit includes a processing condition adjustment unit that adjusts the parameters of the non-rigid registration algorithm according to a tube current ratio when the transmission X-ray data for generating the first image and the second image are acquired. The image reconstruction unit corrects the motion of the object being scanned using the motion information calculated by the motion information acquisition unit to generate a reconstructed image.

11. The X-ray CT apparatus according to claim 10, characterized in that: The motion information acquisition unit includes a noise reduction filter, and the noise reduction filter performs noise reduction processing on the first image and the second image before the registration processing. The processing condition adjustment unit adjusts at least one of the type and the parameter of the noise reduction filter according to the tube current ratio when the transmission X-ray data for generating the first image and the second image are acquired.

12. An image processing device that receives transmission X-ray data collected by an X-ray CT device and performs motion-corrected image reconstruction, wherein: As an image reconstruction unit that reconstructs an image using transmission X-ray data, the image reconstruction unit according to claim 1 or claim 10 is provided.

13. A motion correction image reconstruction method, which uses transmission X-ray data to correct the motion of a subject being scanned to reconstruct a CT image, the method comprising the following steps: Using a portion of the transmission X-ray data to generate a first image and a second image of the facing position; Performing noise reduction processing and registration processing on the first image and the second image respectively, and acquiring motion information of the object; At this time, at least one of the type and parameters of the noise reduction filter used in the noise reduction process and the parameters of the non-rigid registration algorithm used in the registration process are adjusted according to the tube current ratio when the first image and the second image are acquired; and An image is reconstructed using the motion information and the transmission X-ray data.

Images