JP5405293B2 - Image processing apparatus and magnetic resonance imaging apparatus - Google Patents

Description

  The present invention relates to an image processing apparatus and a magnetic resonance imaging apparatus.

  Conventionally, one of inspections using an X-ray CT (Computed Tomography) apparatus or a magnetic resonance imaging apparatus is an inspection called a dynamic study. The dynamic study is a test for examining a temporal change in blood flow perfused into a tissue. In this dynamic study, images of the subject are taken over a plurality of time phases. For example, imaging for several seconds per time phase is performed over a plurality of time phases. Although the time phase to be imaged varies depending on the examination, it is approximately several tens to several tens, and is several tens of seconds to several tens of minutes.

  The purpose of the dynamic study is to obtain information such as the blood flow rate of the tissue by examining the temporal change of the pixel of interest. Therefore, an accurate result cannot be obtained unless the same tissue is imaged on the same pixel in all temporal images. However, the subject often moves during the examination due to the injection of a contrast medium used in the perfusion study and the influence of tasks performed in fMRI (functional MRI). Therefore, the object motion correction is performed on the image in advance before the image analysis.

  There are many types of methods for subject motion correction as known techniques (see, for example, Non-Patent Document 1). For example, an arbitrary time phase image is selected as a reference image from a plurality of time phase images, and a time phase image other than the reference image (hereinafter referred to as a target image) is matched with the selected reference image. There is a method of alignment. In this method, first, the degree of coincidence between the target image and the reference image is calculated. Next, the target image is moved while searching for a moving direction in which the degree of coincidence increases. This process is performed sequentially, and the alignment process ends when the degree of coincidence between both images reaches a predetermined value. Such alignment processing is performed on all target images.

T.R. Oakes, T. Johnstone, K.s. Ores Walsh, L.L. Greischar, A.L. Alexander, A.S. Fox, and R.J. Davidson, "Comparison of fMRI motion correction software tools", ELSEVIER, NeuroImage 28 (2005) p.529-543

  However, in the conventional technique described above, there are cases where the load and time required for the motion correction processing of the subject increase.

  For example, in the motion correction using the reference image, there may be a case where an image in a time phase in which the subject has moved is selected as the reference image. In this case, the distance that the target image must be moved until the degree of coincidence between the reference image and the target image reaches a predetermined value increases. As a result, the number of processes performed sequentially increases, and the load and time required for the motion correction process increase.

  The present invention has been made in view of the above, and an object of the present invention is to provide an image processing apparatus and a magnetic resonance imaging apparatus that can reduce the load and time required for the process of correcting the motion of a subject.

  In order to solve the above-described problems and achieve the object, according to the present invention, the image processing apparatus acquires a plurality of time-phase images obtained by imaging the subject in time series, and acquires each acquired image. A calculation unit that calculates an imaging position of the subject, and a specification that identifies a set of imaging positions that minimizes the variance of the imaging position by analyzing the distribution of the imaging position calculated for each image by the calculation unit Selecting means for selecting at least one image as a reference image from among the images of the plurality of time phases based on imaging positions included in the set of imaging positions specified by the specifying means, and the selecting means And a positioning means for changing the position of an image other than the reference image so as to match the imaging position of the subject in the reference image selected by (1).

  According to a sixth aspect of the present invention, the magnetic resonance imaging apparatus uses a magnetic resonance phenomenon to image the subject in time series, and images of a plurality of time phases captured by the imaging unit. And calculating the imaging position of the subject for each acquired image, and analyzing the distribution of the imaging position calculated for each image by the calculating means, thereby minimizing the variance of the imaging position A specifying unit that specifies a set of imaging positions, and at least one image selected from the plurality of time-phase images as a reference image based on an imaging position included in the set of imaging positions specified by the specifying unit And a position of an image other than the reference image so as to match the imaging position of the subject in the reference image selected by the selection unit. Characterized in that a 置合 allowed means.

  According to the first or sixth aspect of the present invention, there is an effect that it is possible to reduce the load and time required for the process of correcting the motion of the subject.

FIG. 1 is a diagram illustrating an overall configuration of the MRI apparatus according to the present embodiment. FIG. 2 is a functional block diagram illustrating a detailed configuration of the computer system according to the present embodiment. FIG. 3 is a diagram (1) for explaining region extraction by the imaging position calculation unit. FIG. 4 is a diagram (2) for explaining region extraction by the imaging position calculation unit. FIG. 5 is a diagram (1) for explaining identification of the most frequent position group by the most frequent position identifying unit. FIG. 6 is a diagram (2) for explaining identification of the most frequent position group by the most frequent position identifying unit. FIG. 7 is a diagram illustrating an example of a reference selection screen displayed by the reference selection screen control unit. FIG. 8 is a flowchart illustrating a processing procedure of motion correction processing by the computer system according to the present embodiment. FIG. 9 is a diagram illustrating region extraction from a sagittal image. FIG. 10 is a diagram illustrating a change in the center of gravity position in the sagittal image. FIG. 11 is a diagram illustrating an example of a reference selection screen when the imaging position after motion correction is displayed.

  Embodiments of an image processing apparatus and a magnetic resonance imaging apparatus according to the present invention will be described below in detail with reference to the drawings. In the following embodiment, a case where the image processing apparatus according to the present invention is applied to a computer system included in a magnetic resonance imaging apparatus will be described. In the following embodiments, the magnetic resonance imaging apparatus is called an MRI (Magnetic Resonance Imaging) apparatus. In the following embodiment, a case where a brain dynamic study is performed by an MRI apparatus will be described.

  First, the overall configuration of the MRI apparatus according to the present embodiment will be described.

  FIG. 1 is a diagram illustrating an overall configuration of the MRI apparatus according to the present embodiment. As shown in FIG. 1, the MRI apparatus 100 according to the present embodiment includes a static magnetic field magnet 1, a gradient magnetic field coil 2, a gradient magnetic field power source 3, a bed 4, a bed control unit 5, a transmission RF coil 6, a transmission unit 7, and a reception. An RF coil 8, a receiver 9, and a computer system 10 are included.

  The static magnetic field magnet 1 is formed in a hollow cylindrical shape and generates a uniform static magnetic field in a space in the cylinder. The static magnetic field magnet 1 is formed using, for example, a permanent magnet or a superconducting magnet.

  The gradient magnetic field coil 2 is formed in a hollow cylindrical shape and is disposed inside the static magnetic field magnet 1. The gradient coil 2 is formed by combining three coils corresponding to X, Y, and Z axes orthogonal to each other. These three coils are individually supplied with a current from a gradient magnetic field power source 3 to be described later, and generate a gradient magnetic field whose magnetic field strength changes along the X, Y, and Z axes. The Z-axis direction is the same direction as the static magnetic field.

  Here, the gradient magnetic fields of the X, Y, and Z axes generated by the gradient magnetic field coil 2 correspond to, for example, the slice selection gradient magnetic field Gs, the phase encoding gradient magnetic field Ge, and the readout gradient magnetic field Gr, respectively. The slice selection gradient magnetic field Gs is used to arbitrarily determine an imaging section. The phase encoding gradient magnetic field Ge is used to change the phase of an NMR (Nuclear Magnetic Resonance) signal emitted from the subject in accordance with the spatial position. The readout gradient magnetic field Gr is used for changing the frequency of the NMR signal in accordance with the spatial position.

  The gradient magnetic field power supply 3 supplies a current to the gradient magnetic field coil 2 under the control of a computer system 10 described later.

  The couch 4 has a couchtop 4a on which the subject P is placed, and the gradient magnetic field coil 2 is moved by moving the couchtop 4a up and down or in the longitudinal direction under the control of a couch controller 5 described later. The subject is moved in and out of the imaging area inside. Usually, the bed 4 is installed such that the longitudinal direction is parallel to the central axis of the static magnetic field magnet 1.

  The couch controller 5 drives the couch 4 so as to move the top plate 4a in the longitudinal direction or the vertical direction under the control of the computer system 10 to be described later.

  The transmission RF coil 6 is arranged inside the gradient magnetic field coil 2 and receives a high frequency pulse from the transmission unit 7 to generate a high frequency magnetic field.

  The transmission unit 7 transmits a high-frequency pulse corresponding to the Larmor frequency to the transmission RF coil 6 under the control of the computer system 10 described later.

  The reception RF coil 8 is disposed inside the gradient magnetic field coil 2 and receives an NMR signal emitted from the subject due to the influence of the high-frequency magnetic field generated by the transmission RF coil 6. Then, the reception RF coil 8 outputs the received NMR signal to the receiving unit 9 described later.

  The receiving unit 9 generates NMR signal data based on the NMR signal output from the receiving RF coil 8 under the control of the computer system 10 described later. The receiving unit 9 transmits the generated NMR signal data to the computer system 10.

  The computer system 10 performs overall control of the MRI apparatus 100, data collection, image reconstruction, and the like. The computer system 10 includes an interface unit 11, a data collection unit 12, an image reconstruction unit 13, a storage unit 14, a display unit 15, an input unit 16, and a control unit 17.

  The interface unit 11 is connected to the gradient magnetic field power source 3, the bed control unit 5, the transmission unit 7, and the reception unit 9, and controls input / output of signals exchanged between these connected units and the computer system 10. .

  The data collection unit 12 collects NMR signal data transmitted from the reception unit 9 via the interface unit 11. Then, the data collection unit 12 stores the collected NMR signal data in the storage unit 14.

  The image reconstruction unit 13 generates an image relating to the inside of the subject P by performing reconstruction processing such as Fourier transform on the NMR signal data stored in the storage unit 14. Then, the image reconstruction unit 13 stores the generated image in the storage unit 14.

  The storage unit 14 stores the NMR signal data collected by the data collection unit 12 and the image generated by the image reconstruction unit 13 for each subject P. For example, the storage unit 14 is a hard disk drive, a DVD (Digital Versatile Disk) drive, or the like.

  The display unit 15 displays various information such as an image stored in the storage unit 14 and a GUI (Graphical User Interface) for receiving an imaging condition from the operator under the control of the control unit 17. For example, the display unit 15 is a CRT (Cathode Ray Tube) monitor or a liquid crystal monitor.

  The input unit 16 receives input of various information such as imaging conditions and various operations from the operator. For example, the input unit 16 is a mouse, a keyboard, a trackball, a pointing device, or the like.

  The control unit 17 includes a CPU (Central Processing Unit), a memory, and the like (not shown), and comprehensively controls the MRI apparatus 100. For example, the control unit 17 generates various pulse sequences based on the imaging conditions received from the operator, and controls the gradient magnetic field power source 3, the transmission unit 7, and the reception unit 9 according to the generated pulse sequences, thereby Perform imaging.

  Note that the pulse sequence here refers to the strength of the power supplied from the gradient magnetic field power supply 3 to the gradient magnetic field coil 2 and the timing of supplying the power, the strength of the RF signal transmitted from the transmitter 7 to the transmission RF coil 6, Information defining the procedure for performing imaging, such as the timing at which the receiver 9 transmits and the timing at which the receiving unit 9 detects the NMR signal.

  The overall configuration of the MRI apparatus 100 according to the present embodiment has been described above. Under such a configuration, in the MRI apparatus 100, the computer system 10 calculates the imaging position of the subject P for each of the captured images of a plurality of time phases, and analyzes the distribution of the calculated imaging position, A set of imaging positions that minimizes the variance of the imaging positions is specified. Further, the computer system 10 selects at least one image as a reference image from a plurality of time phase images based on the imaging positions included in the specified set. Then, the computer system 10 changes the position of the image other than the reference image, that is, the position of the target image so as to match the imaging position of the subject P in the selected reference image.

  That is, the computer system 10 detects an imaging position with a high frequency in the images of a plurality of time phases, and selects an image in which the subject is imaged at the detected imaging position as a reference image. This reduces the distance that the target image must be moved before the degree of coincidence between the reference image and the target image reaches a predetermined value, thereby reducing the number of sequential processing performed in the alignment process. Can do. Therefore, according to the present embodiment, it is possible to reduce the load and time required for the motion correction process of the subject P.

  Below, the computer system 10 which has the function mentioned above is demonstrated in detail. In the present embodiment, a case where an axial image is captured in time series by the MRI apparatus 100 will be described.

  FIG. 2 is a functional block diagram illustrating a detailed configuration of the computer system 10 according to the present embodiment. As shown in FIG. 2, the computer system 10 includes a storage unit 14, a display unit 15, an input unit 16, and a control unit 17. In FIG. 2, the interface unit 11, the data collection unit 12, and the image reconstruction unit 13 illustrated in FIG.

  The storage unit 14 includes an image storage unit 14a and a reference selection condition storage unit 14b.

  The image storage unit 14 a stores the image generated by the image reconstruction unit 13. In the present embodiment, the image storage unit 14a stores a plurality of time phase images obtained by imaging the head of the subject P in time series.

  The reference selection condition storage unit 14b stores a selection condition determined according to the type of image or the type of examination. The selection conditions here are set according to, for example, the nature of the image to be captured, the algorithm of the alignment processing performed by the image alignment unit 17e described later, the analysis content in the inspection, and the like.

  For example, the characteristics of the alignment processing algorithm depend on the evaluation criteria and the evaluation function used in the algorithm. In some alignment processing algorithms, the alignment accuracy may be lowered when images having different pixel values are aligned. When such an algorithm is used, if a time-phase image at the timing when the contrast agent is flowing in is selected as a reference image, alignment with other time-phase images not stained with the contrast agent is performed. When performing the above, the accuracy of the alignment process is lowered.

  Therefore, in such a case, for example, a time phase in which a region to be analyzed is highly likely to be imaged is set in advance, and the selection condition is such that a reference image is selected from images other than the image at that time phase. Is set. Alternatively, a selection condition is set so that a contrasted time phase image is automatically detected based on the pixel value histogram and a reference image is selected from images other than the detected time phase image. On the contrary, in the examination, when the time phase being contrasted is more important than other time phases, the selection condition is set so that the reference image is selected from the images of the time phase being contrasted.

  The control unit 17 includes an imaging position calculation unit 17a, a most frequent position specification unit 17b, a reference image selection unit 17c, a reference selection screen control unit 17d, an image registration unit 17e, and an image analysis unit 17f.

  The imaging position calculation unit 17a acquires a plurality of time-phase images obtained by imaging the subject P in time series, and calculates the imaging position of the subject P for each acquired image.

  For example, when a plurality of time phase images are captured by the operation of each unit illustrated in FIG. 1, the imaging position calculation unit 17a reads the captured multiple time phase images from the image storage unit 14a and reads the read images. Is stored in memory. Then, the imaging position calculation unit 17a extracts the area of the subject P from each image stored in the memory, and calculates the gravity center position of the extracted area as the imaging position. In this embodiment, the imaging position calculation unit 17a performs a smoothing process on a plurality of time phase images, and then extracts a region of the subject P from the smoothed image.

  3 and 4 are diagrams for explaining region extraction by the imaging position calculation unit 17a. As shown in FIG. 3, first, the imaging position calculation unit 17a generates a smoothed image 32 by applying a smoothing filter to the image 31 read from the image storage unit 14a. Next, the imaging position calculation unit 17a extracts a region 33 of the subject P (a hatched region illustrated in FIG. 3) by performing threshold processing on the pixel value on the image 32.

  In this way, the imaging position calculation unit 17a can appropriately extract only the region of the subject P by performing the smoothing process on the image read from the image storage unit 14a. Here, for example, when the smoothing process is omitted, as shown in FIG. 4, a region 34 other than the subject P such as a noise component region drawn around the subject P is also extracted. there is a possibility. However, even in this case, for example, the imaging position calculation unit 17a detects the number of connected pixels for each extracted region, and performs a process of removing a region with a small number of connections, so that a region other than the subject P can be obtained. It is possible not to be extracted.

  In the present embodiment, the imaging position calculation unit 17a calculates the barycentric position as the imaging position for a plurality of time-phase axial images captured in time series.

  The most frequent position specifying unit 17b analyzes the distribution of the imaging positions calculated for each image by the imaging position calculating unit 17a, thereby specifying a set of imaging positions that minimizes the variance of the imaging positions. Hereinafter, a set of imaging positions identified by the most frequent position identifying unit 17b is referred to as a most frequent position group.

  For example, the mode position specifying unit 17b specifies the mode position group by performing cluster analysis on the distribution of the image pickup positions calculated for each image by the image pickup position calculating unit 17a. It should be noted that various generally known analysis methods can be used for the cluster analysis here.

  5 and 6 are diagrams for explaining the identification of the most frequent position group by the most frequent position identifying unit 17b. 5 and 6, the horizontal axis indicates the time phase, and the vertical axis indicates the position of the center of gravity. Further, each point shown in FIGS. 5 and 6 indicates an imaging position calculated for each image by the imaging position calculation unit 17a. For example, the mode position specifying unit 17b specifies the imaging positions included in the range 51 shown in FIG. 5 or the range 61 shown in FIG. 6 as the mode position group.

  Further, after specifying the mode position group, the mode position specifying unit 17b may calculate the representative value of the imaging position included in the mode position group as the mode position. For example, the most frequent position specifying unit 17b obtains an approximate function that approximates the relationship between the barycentric position included in the most frequent position group and the time phase by using the least square method. The mode position specifying unit 17b calculates, for example, a time phase located in the middle of the time phase range included in the mode position group. Then, the most frequent position specifying unit 17b calculates the position of the center of gravity obtained by substituting the calculated time phase into the approximate function as a representative value. Alternatively, for example, the most frequent position specifying unit 17b may calculate the average value of the imaging positions as a representative value.

  In this embodiment, the imaging position specifying unit 17b specifies the most frequent position group for a plurality of time-phase axial images captured in time series.

  The reference image selection unit 17c selects at least one image as a reference image from images of a plurality of time phases based on the imaging positions included in the mode position group specified by the mode position specifying unit 17b.

  For example, the reference image selection unit 17c specifies the earliest time phase among the time phases of the imaging positions included in the most frequent position group, and selects an image of the specified time phase as a reference image. In addition, when the most frequent position is calculated by the most frequent position specifying unit 17b, the reference image selecting unit 17c selects, for example, an image whose imaging position of the subject P is closest to the most frequent position as the reference image. .

  The reference image selection unit 17c selects an image based on a selection condition determined according to the type of image or the type of examination when selecting a reference image. For example, the reference image selection unit 17c specifies the type of captured image or the type of inspection based on the imaging conditions set by the operator during the inspection. Then, the reference image selection unit 17c acquires a selection condition corresponding to the specified image type or examination type from the reference selection condition storage unit 14b, and selects a reference image according to the acquired selection condition.

  For example, the reference image selection unit 17c selects a reference image from images other than a time phase in which a region to be analyzed is highly likely to be contrasted among time phase images at an imaging position included in the most frequent position group. Alternatively, the reference image selection unit 17c automatically detects a contrasted time-phase image based on a histogram of pixel values in a time-phase image at an imaging position included in the most frequent position group, and A reference image is selected from images other than the phase image. Or, in the examination, when the time phase being contrasted is more important than the other time phases in the examination, the reference image selection unit 17c, among the time phase images of the imaging position included in the most frequent position group, A reference image is selected from the time phase images being contrasted.

  When the reference image selection unit 17c selects the reference image, the reference image selection unit 17c notifies the reference selection screen control unit 17d described later of information indicating the selected reference image. In this embodiment, the reference image selection unit 17c selects a reference image for each of a plurality of time-phase axial images captured in time series, a sagittal image generated from the axial image, and a coronal image, and each reference image is selected. Is sent to the reference selection screen control unit 17d.

  The reference selection screen control unit 17d displays the reference image selected by the reference image selection unit 17c on the display unit 15, and also selects an arbitrary image from the images of a plurality of time phases via the input unit 16. Is received from the operator.

  For example, when the reference image is selected by the reference image selection unit 17c, the reference selection screen control unit 17d causes the display unit 15 to display a reference selection screen having various GUIs (Graphical User Interfaces). The reference selection screen control unit 17d displays various images and accepts various operations via the reference selection screen.

  FIG. 7 is a diagram illustrating an example of a reference selection screen displayed by the reference selection screen control unit 17d. As shown in FIG. 7, for example, the reference selection screen 70 includes an axial image display area 71, a sagittal image display area 72, a coronal image display area 73, an imaging position display area 74, an Acpt button 75, and a Manual button 76.

  The reference selection screen 70 includes a vertical scroll bar 71 a provided beside the axial image display area 71 and a horizontal scroll bar 71 b provided below the axial image display area 71. The reference selection screen 70 includes a vertical scroll bar 72 a provided beside the sagittal image display area 72 and a horizontal scroll bar 72 b provided below the sagittal image display area 72. The reference selection screen 70 has a vertical scroll bar 73 a provided beside the coronal image display area 73 and a horizontal scroll bar 73 b provided below the coronal image display area 73.

  When the reference selection screen control unit 17d is notified of the information indicating the reference image from the reference image selection unit 17c, the reference selection screen control unit 17d receives the reference images of the axial image, the sagittal image, and the coronal image from the memory based on the notified information. read out. Subsequently, the reference selection screen control unit 17 d displays an axial image reference image in the axial image display area 71. In addition, the reference selection screen control unit 17 d displays a reference image of the sagittal image in the sagittal image display area 72. In addition, the reference selection screen control unit 17 d displays a coronal image reference image in the coronal image display area 73.

  Thereafter, when the reference selection screen control unit 17d receives an operation of moving the knob of the vertical scroll bar 71a up and down, the reference selection screen control unit 17d reads out an axial image having a different slice position from the memory according to the amount of movement of the knob and reads the axial image. It is displayed in the display area 71. In addition, when the reference selection screen control unit 17d receives an operation of moving the knob of the horizontal scroll bar 72b to the left or right, the reference selection screen control unit 17d reads out an axial image having a different time phase from the memory according to the amount of movement of the knob. It is displayed in the display area 71.

  Here, when displaying the reference image of the axial image, the reference selection screen control unit 17d displays an icon 71c indicating that the displayed axial image is the reference image, as shown in FIG. Display overlaid on top. Further, the reference selection screen control unit 17d displays the slice number of the displayed axial image so as to be superimposed on the axial image. For example, “Sleece 05/23” shown in FIG. 7 indicates that the fifth axial image among the 23 axial images is displayed. In addition, the reference selection screen control unit 17d displays the time phase of the axial image being displayed so as to overlap the axial image. For example, “Phase 15/167” shown in FIG. 7 indicates that an axial image of the 15th time phase of the 167 time phases is displayed.

  Note that the reference selection screen control unit 17d performs the same display control as the axial image display area 71 for the sagittal image display area 72 and the coronal image display area 73. Thus, the operator can easily confirm which image is selected as the reference image for each of the axial image, the sagittal image, and the coronal image by operating the scroll bar.

  The reference selection screen control unit 17d displays an axial image, a sagittal image, and a coronal image, and displays the imaging position calculated by the imaging position calculation unit 17a in the imaging position display area 74 for any of the images. In the example illustrated in FIG. 7, a case where the imaging position of the axial image is displayed in the imaging position display area 74 is illustrated.

  For example, as illustrated in FIG. 7, the reference selection screen control unit 17d displays two-dimensional coordinates (axi_x, axi_y) indicating the center of gravity position in the axial image in the imaging position display area 74 for each time phase (FIG. 7). Diamond points shown in Fig. 7).

  Here, as shown in FIG. 7, the reference selection screen control unit 17d gives a color different from the other imaging positions to the imaging positions included in the most frequent position group specified by the most frequent position specifying unit 17b. Are displayed (dots in the shaded area shown in FIG. 7). Further, the reference selection screen control unit 17d displays the imaging position of the reference image selected by the reference image selection unit 17c with a color different from the imaging position included in the most frequent position group (shown in FIG. 7). Black diamond point).

  Further, the reference selection screen control unit 17d displays information that is a basis for specifying the most frequent position group and the reference image in the imaging position display area 74. For example, as illustrated in FIG. 7, the reference selection screen control unit 17 d displays a curve graph indicating a change over time in the concentration of the contrast agent on the imaging position display region 74. At this time, the reference selection screen control unit 17d displays a range of time phases included in the most frequent position group with a different color from other time phases in the curve graph (shown in FIG. 7). Hatched area). Further, the reference selection screen control unit 17d displays the time phase of the reference image with a different color in the range indicating the most frequent position group (a grid-like shaded range shown in FIG. 7). .

  Here, for example, when the reference selection screen control unit 17d receives an operation of pressing the Manual button 76 from the operator, the reference selection screen control unit 17d selects an arbitrary time phase from the time phases displayed in the imaging position display area 74. Make the selected operation acceptable. When an operation for selecting a time phase is received, the reference selection screen control unit 17d reads out the image of the selected time phase from the memory and displays it. For example, when the imaging position of the axial image is displayed in the imaging position display area 74, the reference selection screen control unit 17 d reads out the axial image of the selected time phase from the memory and displays it in the axial image display area 71. To do.

  Further, when the reference selection screen control unit 17d receives an operation of pressing the Accept button 75, the reference selection screen control unit 17d sets an image of a phase selected at that time as a reference image. When the reference selection screen control unit 17d sets the reference image, the reference selection screen control unit 17d notifies the image registration unit 17e described later of information indicating the set reference image. Here, when the reference image is not changed by the operator, the reference selection screen control unit 17d notifies the image registration unit 17e of information indicating the reference image selected by the reference image selection unit 17c.

  In the present embodiment, the reference selection screen control unit 17d notifies the reference selection screen control unit 17d of information indicating the reference image for each of the axial image, the sagittal image, and the coronal image.

  As described above, in this embodiment, the reference selection screen control unit 17d sets a reference image serving as a reference in motion correction performed by the image alignment unit 17e based on an operation by the operator. Therefore, the operator can reselect the reference image even after the reference image is selected by the reference image selection unit 17c. For example, the operator can reselect the reference image when a noise component is included in the image for some reason. Further, when the operator reselects the reference image, if the reference image is selected within the range of the most frequent position group displayed in the imaging position display area 74, the load and time required for the motion correction process are greatly increased. It is possible to reselect the reference image without making it.

  Here, the case has been described in which the reference selection screen control unit 17d displays in the imaging position display area 74 information that is the basis for specifying the most frequent position group and the reference image. However, for example, when a selection condition depending on the analysis content or the like is used, information related to the selection condition may be further displayed in the graph. For example, the reference selection screen control unit 17d may display pixel values for each time phase so that the time phase stained with the contrast agent can be distinguished from the time phase not stained. Thereby, for example, when the operator wants to select a reference image from the contrasted time phases, the operator can easily select a desired reference image.

  The image alignment unit 17e performs motion correction of the subject P by changing the position of an image other than the reference image so as to match the imaging position of the subject P in the reference image.

  For example, when information indicating the reference image is notified from the reference selection screen control unit 17d, the image alignment unit 17e reads the reference image from the internal memory based on the notified information. Further, the image alignment unit 17e reads an arbitrary image other than the reference image from the memory as a target image, and calculates the degree of coincidence between the read target image and the reference image. Here, the degree of coincidence is an evaluation criterion indicating how similar the comparison target images are. For example, the degree of coincidence is a mutual information amount.

  Subsequently, the image alignment unit 17e sequentially performs a process of moving the target image while searching for a moving direction in which the degree of coincidence increases. When the degree of coincidence between both images reaches a predetermined value, the position of the target image is determined. The alignment process is terminated. The image alignment unit 17e performs such alignment processing for all images other than the reference image. Then, the image alignment unit 17e stores all images subjected to the alignment process in the memory as motion-corrected images.

  The image analysis unit 17f performs image analysis processing on the image that has been subjected to motion correction by the image alignment unit 17e. For example, after motion correction is performed by the image alignment unit 17e, an image analysis application corresponding to the type of image to be inspected is started. The image analysis application here is, for example, an application for analyzing an fMRI image or an application for analyzing a PWI (Perfusion Weighted Imaging) image. Then, the activated application performs various image analysis processes using the motion-corrected image stored in the memory.

  Next, a processing procedure of motion correction processing by the computer system 10 according to the present embodiment will be described.

  FIG. 8 is a flowchart illustrating the processing procedure of the motion correction processing by the computer system 10 according to the present embodiment. As shown in FIG. 8, in the MRI apparatus 100 according to the present embodiment, when a plurality of time-phase images regarding the subject P are captured (step S <b> 101, Yes), the imaging position calculation unit 17 a performs each captured image. Then, the imaging position of the subject P is calculated (step S102).

  Subsequently, the most frequent position specifying unit 17b analyzes the distribution of the imaging positions calculated for each image by the imaging position calculating unit 17a, so that the most frequent position is a set of imaging positions in which the variance of the imaging positions is minimized. A group is specified (step S103). Then, the reference image selection unit 17c selects a reference image from a plurality of time-phase images based on the imaging positions included in the mode position group specified by the mode position specifying unit 17b (step S104).

  Subsequently, the reference selection screen control unit 17d causes the display unit 15 to display the reference image selected by the reference image selection unit 17c via the reference selection screen 70 (step S105). When the manual button 76 is pressed by the operator (step S106, Yes), the reference selection screen control unit 17d selects an arbitrary time phase from the time phases displayed in the imaging position display area 74. A selection operation is accepted (step S107).

  Thereafter, the reference selection screen control unit 17d continues to accept an operation for selecting an arbitrary time phase until an operation for pressing the Accept button 75 is accepted from the operator (No at Step S108). When an operation of pressing the Accept button 75 is received from the operator (Yes in step S108), the reference selection screen control unit 17d sets the image of the time phase selected at that time as a reference image. (Step S109).

  On the other hand, when the Manual button 76 is not pressed (step S106, No) and the Accept button 75 is pressed (step S108, Yes), the reference selection screen control unit 17d is selected by the reference image selection unit 17c. An image is set as a reference image (step S109).

  Then, after the reference image is set, the image alignment unit 17e changes the position of the image other than the reference image so as to match the imaging position of the subject P in the set reference image. The motion correction is executed (step S110). Then, the image analysis unit 17f performs an image analysis process on the image on which the motion correction has been performed by the image alignment unit 17e (step S111).

  In this embodiment, when the reference selection screen control unit 17d displays the reference image on the display unit 15 and the accept button 75 is pressed by the operator, the image registration unit 17e performs image registration. It was. However, for example, after the reference image is selected by the reference image selection unit 17c, the image alignment unit 17e performs image alignment using the selected reference image without waiting for confirmation by the operator. May be.

  Thereby, for example, when an inspection with a fixed selection condition is performed, or when it is known before the inspection that noise is unlikely to be mixed into the image, the inspection can be completed quickly. . Even in this case, the reference selection screen controller 17d uses the reference selection screen 70 and the like after the image alignment is completed so that the operator can confirm which reference image has been selected. It is desirable to display an image.

  As described above, in the present embodiment, the imaging position calculation unit 17a acquires a plurality of time-phase images obtained by imaging the subject P in time series, and calculates the imaging position of the subject P for each acquired image. . Further, the mode position specifying unit 17b analyzes the distribution of the imaging positions calculated for each image, thereby specifying the mode position group that minimizes the variance of the imaging positions. Further, the reference image selection unit 17c selects at least one image as a reference image from a plurality of time phase images based on the imaging positions included in the identified most frequent position group. Then, the image alignment unit 17e changes the position of the image other than the reference image, that is, the target image so as to match the imaging position of the subject P in the selected reference image. This reduces the distance that the target image must be moved before the degree of coincidence between the reference image and the target image reaches a predetermined value, thereby reducing the number of sequential processing performed in the alignment process. Can do. Therefore, according to the present embodiment, it is possible to reduce the load and time required for the motion correction process of the subject P.

  In this embodiment, the mode position specifying unit 17b calculates the representative value of the imaging position included in the mode position group after specifying the mode position group. Then, the reference image selection unit 17c selects an image whose imaging position of the subject P is closest to the calculated representative value as a reference image. Therefore, according to the present embodiment, it is possible to select an optimal reference image even when there are variations in imaging positions with high frequency.

  In this embodiment, the reference image selection unit 17c selects a reference image based on a selection condition determined according to the type of image or the type of examination. Therefore, according to the present embodiment, an appropriate reference image can be selected according to the type of image and the type of examination.

  In the present embodiment, the display unit 15 displays the selected reference image. Further, the input unit 16 receives an operation for selecting an arbitrary image from images of a plurality of time phases from the operator. Then, when an operation for selecting an image is accepted, the image alignment unit 17e changes the position of an image other than the reference image using the image selected by the operator as a reference image. Therefore, according to the present embodiment, even when the optimum reference image is not selected due to the influence of noise or the like, the operator can appropriately change the reference image.

  In the present embodiment, the imaging position calculation unit 17a performs the smoothing process on the images of a plurality of time phases, extracts the region of the subject P from the smoothed image, and extracts the extracted region. The center of gravity position is calculated as the imaging position. Therefore, according to the present embodiment, only the region to be analyzed can be appropriately extracted, so that the reference image can be selected more accurately.

  In the above embodiment, the imaging position calculation unit 17a calculates the imaging position for each of a plurality of time-phase axial images, sagittal images, and coronal images captured in time series. However, depending on the type of inspection, the imaging position may be calculated only for a specific cross-sectional image. For example, in fMRI, since the subject's head is fixed from the side so that the subject does not move during imaging, the subject rarely moves the neck left and right during imaging, and the neck shakes up and down. Only prone to occur. Therefore, in fMRI, the imaging position need only be calculated for the sagittal image only.

  FIG. 9 is a diagram illustrating region extraction from a sagittal image. For example, when fMRI is performed, the imaging position calculation unit 17a performs a threshold value process on the captured sagittal image 91 with respect to the pixel value, so that the region 92 of the subject P (the hatched line illustrated in FIG. 9 is attached). Extracted region). And the imaging position calculation part 17a calculates the gravity center position of the extracted area | region as an imaging position.

  FIG. 10 is a diagram illustrating a change in the center of gravity position in the sagittal image. As shown in FIG. 10, for example, when the centroid position of the part in the sagittal image is represented by two-dimensional coordinates (sag_i, sag_j), the time-series change in the centroid position is represented by the distribution of sag_i and sag_j. The In this case, the mode position specifying unit 17b specifies the mode position group of sag_i by performing cluster analysis on the distribution of sag_i. Further, the mode position specifying unit 17b specifies the mode position group of sag_j by performing cluster analysis on the distribution of sag_j.

  For example, the reference image selection unit 17c selects a reference image from temporal images included in both the sag_i mode position group and the sag_j mode position group.

  Thus, for example, when the movement of the subject is restricted to a specific direction, the imaging position calculation unit 17a may calculate the imaging position only for a specific cross-sectional image. In that case, the imaging position specifying unit 17b specifies the most frequent position group for only a specific cross-sectional image. The reference image selection unit 17c selects a reference image for only a specific cross-sectional image. As a result, it is possible to further reduce the load and time required for the motion correction processing of the subject.

  In the above-described embodiment, the case where the reference selection screen control unit 17d displays the imaging position calculated by the imaging position calculation unit 17a in the imaging position display area 74 of the reference selection screen 70 has been described. However, for example, in addition to the imaging position calculated by the imaging position calculation unit 17a, the reference selection screen control unit 17d further displays the imaging position after the motion correction is performed by the image alignment unit 17e. Also good.

  FIG. 11 is a diagram illustrating an example of a reference selection screen when the imaging position after motion correction is displayed. As shown in FIG. 11, in this case, for example, the reference selection screen 70 is provided with a pre-correction position display area 74a for displaying an imaging position before motion correction and a post-correction position display area 74b.

  Then, the reference selection screen control unit 17d first displays the imaging position calculated by the imaging position calculation unit 17a in the pre-correction position display area 74a. Further, the reference selection screen control unit 17d, after the image alignment is performed by the image alignment unit 17e, the imaging positions in the images of a plurality of time phases after the alignment is completed in the corrected position display area 74b. indicate.

  In this way, the reference selection screen control unit 17d displays the imaging positions before and after the motion correction on the display unit 15 so that the operator can easily confirm the result of the motion correction.

  In the above embodiment, the case where the brain dynamic study is performed by the MRI apparatus has been described. However, the present invention is not limited to this. For example, the present invention can be similarly applied to other diagnostic imaging apparatuses such as an X-ray CT (Computed Tomography) apparatus and a SPECT (Single Photon Emission Computed Tomography) apparatus.

  Further, the present invention can be similarly applied to image processing apparatuses that process images taken by various image diagnostic apparatuses such as an MRI apparatus, an X-ray CT apparatus, and a SPECT apparatus. The image processing apparatus here is connected to a communication system such as a PACS (Picture Archiving and Communication System). Then, the image processing apparatus acquires image data from the image diagnostic apparatus via the network, and stores the acquired image in a storage unit in the own apparatus.

  In general, an image captured by an image diagnostic apparatus has different properties depending on the image diagnostic apparatus, the imaging method, the type of imaging region, and the like. For example, when the brain is imaged by an X-ray CT apparatus, an image in which the skull is clearly depicted is obtained. Therefore, the region of the skull can be easily and accurately extracted from the image captured by the X-ray CT apparatus.

  Therefore, for example, when the image processing apparatus processes a brain image captured by the X-ray CT apparatus, a skull region is extracted from the processing target image, and the imaging position is determined based on the extracted skull region. You may make it calculate. As a result, alignment can be performed more easily and accurately. Further, for example, in the kidney perfusion examination, the image processing apparatus may set an easily extractable part such as the pelvis included in the imaging range as a Merckmar and calculate the imaging position based on the Merckmar.

  That is, the reference part for calculating the imaging position is not limited to the part to be analyzed. Furthermore, the reference for calculating the imaging position is not necessarily a part of the subject. For example, a marker may be attached to the subject before imaging, and the image processing apparatus may extract the marker from the captured image and calculate the imaging position with reference to the marker.

DESCRIPTION OF SYMBOLS 100 MRI apparatus 10 Computer system 14 Memory | storage part 14a Image memory | storage part 14b Reference selection condition memory | storage part 17 Control part 17a Imaging position calculation part 17b Most frequent position specific | specification part 17c Reference image selection part 17d Reference selection screen control part 17e Image alignment part 17f Image analysis unit

Claims (6)

Calculating means for acquiring images of a plurality of time phases obtained by imaging the subject in time series, and calculating the imaging position of the subject for each acquired image;
Identifying means for identifying a set of imaging positions that minimizes variance in imaging positions by analyzing the distribution of imaging positions calculated for each image by the calculating means;
Selection means for selecting at least one image as a reference image from among the images of the plurality of time phases based on imaging positions included in the set of imaging positions specified by the specifying means;
An image processing apparatus comprising: an alignment unit configured to change a position of an image other than the reference image so as to match an imaging position of the subject in the reference image selected by the selection unit. The specifying means, after specifying the set of imaging positions, calculates a representative value of the imaging positions included in the set,
The image processing apparatus according to claim 1, wherein the selection unit selects, as the reference image, an image whose imaging position of the subject is closest to the representative value calculated by the specifying unit.   The image processing apparatus according to claim 1, wherein the selection unit selects the reference image based on a selection condition determined according to the type of the image or the type of examination. Display means for displaying a reference image selected by the selection means;
Receiving means for receiving an operation of selecting an arbitrary image from the images of the plurality of time phases from an operator;
When the operation for selecting an image is received by the receiving unit, the alignment unit changes the position of an image other than the reference image using the image selected by the operator as the reference image. The image processing apparatus according to claim 1, 2, or 3.   The calculation means performs a smoothing process on the images of the plurality of time phases, extracts the region of the subject from the smoothed image, and calculates the gravity center position of the extracted region as the imaging position The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. Imaging means for imaging the subject in time series using a magnetic resonance phenomenon;
Calculating means for acquiring images of a plurality of time phases imaged by the imaging means, and calculating an imaging position of the subject for each acquired image;
Identifying means for identifying a set of imaging positions that minimizes variance in imaging positions by analyzing the distribution of imaging positions calculated for each image by the calculating means;
Selection means for selecting at least one image as a reference image from among the images of the plurality of time phases based on imaging positions included in the set of imaging positions specified by the specifying means;
A magnetic resonance imaging apparatus comprising: an alignment unit configured to change a position of an image other than the reference image so as to match an imaging position of the subject in the reference image selected by the selection unit.

Images