JP6748866B2 - Computer program for image processing, image processing apparatus and method - Google Patents

Description

The present invention relates to a medical image processing technique, and more particularly to a technique for analyzing a motion of a heart.

Generally, the movement of the heart is analyzed by using an image obtained by dividing one cycle of the heartbeat into a plurality of time phases in synchronization with the electrocardiogram. For example, Patent Document 1 describes a technique for evaluating ventricular function using cine MRI.

JP, 2005-270491, A

Although the heart expands and contracts in synchronization, there is a dysfunction called dyssynchronism in which the timing of this synchronization is shifted. By more specifically grasping the state of dyssynchrony, it is possible to more appropriately determine the indication of treatment such as CRT (Cardiac Resynchronization Therapy).

A computer program according to an embodiment of the present invention is a computer program for image processing performed by a computer that divides one cycle of a heartbeat into a plurality of time phases and stores three-dimensional volume data of the heart in each time phase. And, for one or more regions of the wall of the atrium or ventricle in the three-dimensional volume data, the movement from the first time phase to the second time phase after the first time phase is performed for each area. A vector specifying step for specifying a vector shown, a vector for each area, and a dyssynchronization area specifying step for specifying a dyssynchronized area in the atrium or ventricle based on the barycentric position of the atrium or ventricle; is executed, the dyssynchrony region specifying step, the the direction of the vector of each region, based on the difference between the direction to the center-of-gravity position from each region, that identifies the area of the dyssynchrony.

In a preferred embodiment, the synchronization failure region specifying step specifies, for each region, the angle between the vector of each region and the vector from each region to the center of gravity position, and the angle is a region equal to or greater than a predetermined threshold value. It may be specified as the area of dyssynchrony.

In a preferred embodiment, the step of identifying a synchronization failure area identifies an area in which the angle is equal to or greater than a predetermined threshold value and a vector length of each area is equal to or greater than a predetermined threshold value as the area of the synchronization failure. May be.

In a preferred embodiment, each area may be composed of one voxel, and the vector specifying step may specify a vector for each voxel.

In a preferred embodiment, the method may further include the step of identifying a representative vector for each segment having a plurality of voxels based on the vector of each voxel. The synchronization failure area specifying step may specify the synchronization failure area in segment units based on the representative vector.

In a preferred embodiment, the dyssynchronization region specifying step specifies a vector of each voxel and an angle of a vector from each voxel to the center of gravity position for each voxel, and based on the angle of each voxel, a plurality of It is also possible to specify a representative angle for each segment having the voxel and specify a segment in which the representative angle of each segment is equal to or more than a predetermined threshold value as the area of the synchronization failure.

In a preferred embodiment, each of the regions may be a segment having a plurality of voxels. The vector identifying step may identify a vector for each segment.

In a preferred embodiment, a step of generating the three-dimensional volume data by interpolating one or more extended slice images between adjacent slice images from a plurality of slice images obtained by imaging the heart. , May further have.

1 is an overall configuration diagram of a medical image processing apparatus 1 according to an embodiment of the present invention. An example of the image based on imaging data is shown. It is a schematic diagram explaining the interpolation of imaging data. It is a schematic diagram of the left ventricle of a 1st time phase and a 2nd time phase. It is a flowchart which shows an example of the calculation procedure of a movement vector. It is explanatory drawing of a rotation angle. An example of the data structure of the motion analysis data 190 is shown. It is a display example of the motion analysis result of the left ventricle. It is a flowchart which shows the procedure of the determination process of the dyssynchrony region of a left ventricle.

Hereinafter, a medical image processing apparatus according to an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is an overall configuration diagram of a medical image processing apparatus 1 according to this embodiment.

The medical image processing apparatus 1 includes a medical image processing apparatus main body 10, an input device 2 such as a keyboard and a pointing device connected to the medical image processing apparatus main body 10, and a display device 3 such as a liquid crystal display.

The medical image processing apparatus main body 10 is configured by, for example, a general-purpose computer system, and each component or function in 10 described below is realized by executing a computer program, for example.

The medical image processing apparatus main body 10 includes an imaging data storage unit 11, an extension data storage unit 13, a movement vector data storage unit 15, a segment data storage unit 17, a motion analysis data storage unit 19, and an interpolation processing unit 31. The center-of-gravity calculation unit 33, the movement vector calculation unit 35, the representative vector calculation unit 37, the synchronization failure region extraction unit 39, and the display control unit 45. The out-of-synchronization region extraction unit 39 further includes a movement direction identification unit 41 and a vector length identification unit 43.

The imaging data storage unit 11 stores imaging data 110 obtained by imaging the heart of the subject. The imaged data 110 may be data of imaged images of each time phase obtained by dividing one cycle of the heartbeat into a plurality of time phases. The captured image data of each time phase may be three-dimensional volume data having a plurality of voxels, and may be, for example, MR (Magnetic Resonance) image data or CT (Computed Tomography) image data.

FIG. 2 shows an example of an image based on the imaging data 110 according to this embodiment. As shown in the figure, the captured image of this embodiment is an MR image in which one heartbeat is divided into n time phases. The images of each time phase in the figure are slice images obtained by tomographic imaging of the heart from the apex to the base of the heart in the short-axis direction. Each slice image of the present embodiment is an image of a region including the left ventricle and the right ventricle of the heart, and may be composed of m sheets. The captured image is not limited to the left ventricle and the right ventricle, and may be slice images of the left atrium and the right atrium.

Here, in the imaging data 110, the distance between slice images is large in each time phase and is discrete in the long axis direction. Therefore, in order to grasp the smooth three-dimensional movement of the heart, it is preferable that the interval between slices in the long axis direction is narrow.

Therefore, in the present embodiment, the motion of the heart is analyzed using the expanded data obtained by expanding the imaging data 110.

Returning to FIG. 1, the extended data storage unit 13 stores the extended data obtained by extending the imaging data 110. The interpolation processing unit 31 performs interpolation processing based on the imaged data 110 to generate extended data. The extended data is also composed of a plurality of time phases, and the image of each time phase is also three-dimensional volume data having a plurality of voxels.

FIG. 3 is a schematic diagram illustrating the interpolation of the imaging data 110.

The same figure A is a schematic diagram of the imaging data 110 of one certain time phase which has several imaging slice images 115. The captured slice image 115 schematically shows the contour of the left ventricle in the short-axis tomographic image of the left ventricle.

FIG. 11B is a schematic diagram of the extended data 130 in which the interpolation slice image 135 is inserted between the captured slice images 115 of FIG.

The interpolation processing unit 31 generates the extension data 130 from the imaged data 110 using, for example, a known image interpolation technique. For example, the interpolation processing unit 31 uses the data of two adjacent imaging slice images 115 or the imaging slice images 115 around the imaging slice images 115, and one or more interpolation slice images interpolated between the imaging slice images 115. 135 is generated. The interpolation processing unit 31 interpolates the interpolated slice image 135 in the captured slice image 115 for all time phases. It should be noted that in the following embodiments, various processing is performed using the extended data, but the imaged data may be used.

Returning to FIG. 1, the center-of-gravity calculator 33 calculates the center-of-gravity position of each ventricle or atrium of the heart based on the expanded data 130.

For example, the center-of-gravity calculating unit 33 may specify the center-of-gravity position of each of the left and right ventricles or atria in each time phase based on the expanded data 130. The position of the center of gravity may be specified as follows, for example. That is, the center-of-gravity calculation unit 33 specifies the voxels of the myocardial (wall) region that constitutes the target ventricle or atrium for each time phase based on the expanded data 130, and the coordinates of the voxels of the specified myocardial region. Based on, the center of gravity position (voxel) of the target ventricle or atrium may be specified.

The movement vector calculation unit 35 calculates a movement vector indicating the movement of the heart based on the expansion data 130.

The movement vector data storage unit 15 stores the movement vector data calculated by the movement vector calculation unit 35.

The movement vector calculation unit 35, for example, with respect to one or more regions of the wall (myocardium) of the atrium or ventricle in the dilated data 130, a certain time phase (first time phase) to another time phase after that time phase (second time phase). You may specify the vector which shows movement to (time phase).

The movement vector calculation unit 35 specifies, for example, voxels in the myocardial region in the first time phase. Further, the movement vector calculation unit 35 may specify the position (voxel) of the destination on the second time phase where each voxel in the identified myocardial region of the first time phase has moved due to the motion of the heart. The voxel position of the first time phase and the voxel position of the corresponding second time phase do not have to be on the same plane. That is, in the present embodiment, the motion of the myocardium is represented by the movement vector that also includes the component in the long axis direction of the heart, so that the three-dimensional motion close to the actual motion of the myocardium can be detected.

The above-described movement destination voxel position on the second time phase may be specified by, for example, a matching process using the voxel value of each voxel. That is, the movement vector calculation unit 35 searches for a pattern within the predetermined search range by pattern matching, and detects a voxel value of the second temporal phase having substantially the same voxel value as the voxel of the voxel on the first temporal phase (pre-movement voxel). A voxel (destination voxel) may be specified. The search for the destination voxel may be performed by a block matching method in which the matching is performed for each block including a plurality of voxels including the position of the voxel before the movement. The shape of the block used here may be, for example, a rectangular parallelepiped or a cube. The block size is arbitrary.

If there is no matching voxel within the search range (the destination voxel cannot be specified), the operator may manually specify the destination voxel.

The second time phase may be a time phase next (temporally adjacent) to the first time phase or a time phase later than the next time phase.

FIG. 4 is a diagram in which the schematic diagram At of the left ventricle at the first time phase (time T=t) and the schematic diagram At+1 of the left ventricle at the second time phase (time T=t+1) are overlapped and displayed. The center of gravity position G in the figure is the center of gravity position of the first time phase and/or the second time phase.

The figure shows that the movement vector calculation unit 35 detects that the voxel Bt in the first time phase has moved to the voxel Bt+1 in the second time phase. At this time, the vector V1 from the voxel Bt to the voxel Bt+1 becomes the movement vector of the voxel Bt. The movement vector calculation unit 35 may calculate such movement vectors for all voxels in the myocardial region of the first time phase.

It should be noted that in the above-described embodiment, the movement vector is calculated with voxels as one area, but a plurality of voxels may be combined into one area and the movement vector may be calculated for each area. For example, the movement vector calculation unit 35 may calculate the movement vector for each segment described below. In this case, the movement vector calculated for each segment is equivalent to a representative vector that is determined for each segment to be described later.

FIG. 5 is a flowchart showing an example of the procedure for calculating the movement vector.

The movement vector calculation unit 35 specifies the processing target time phase Phx (S101).

The movement vector calculation unit 35 specifies a voxel in the myocardial region of the processing target time phase Phx in the expansion data 130 as a processing target voxel vx (S103).

The movement vector calculation unit 35 searches the next time phase Phx+1 of the processing time phase Phx of the extension data 130 by the box matching method, and specifies the movement destination voxel vx+1 of the target voxel vx (S105).

The movement vector calculation unit 35 calculates a three-dimensional movement vector of the target voxel vx from the three-dimensional coordinates of the target voxel vx before movement and the three-dimensional coordinates of the movement destination voxel vx+1 (S107).

The movement vector calculation unit 35 repeats the processing of steps S103 to S107 until the movement vectors are generated for all voxels in the myocardial region of the processing target time phase Phx (S109).

The movement vector calculation unit 35 further repeats the processing of steps S101 to S109 until the movement vectors are generated for all the time phases (S111). Thereby, the movement vector is calculated for each voxel.

Returning to FIG. 1, the segment data storage unit 17 stores segment data for dividing the heart into a plurality of segments. A segment is a region in which each ventricle or each atrium of the heart is divided into a plurality of regions. Each segment contains one or more voxels. The method of dividing the segments is arbitrary, but for example, the segments may be divided into anatomically meaningful segments or may be divided into anatomically meaningless segments. The anatomically meaningful segmentation may be, for example, an AHA (American Heart Association) segment model such as a polar map (bullseye map). The segmentation of a shape that does not have anatomical meaning may be a cube or a rectangular parallelepiped.

In the segment data stored in the segment data storage unit 17, for example, segment identification information and position coordinates forming each segment may be associated with each other.

The representative vector calculation unit 37 may calculate the representative vector of each segment based on the movement vector data 150 of the movement vector data storage unit 15. For example, the representative vector calculation unit 37 targets either the left or right atrium or ventricle and divides the target atrium or ventricle into a plurality of segments based on the segment data in the segment data storage unit 17. The representative vector calculation unit 37 may specify the representative vector for each segment based on the movement vector for each voxel. For example, an average vector of movement vectors of voxels included in each segment may be used as the representative vector. Alternatively, the median value, the maximum value, or the mode value of the lengths of the voxel movement vectors included in each segment may be set as the length of the representative vector. Further, regarding the direction of the vector, the center of the direction of the movement vector of the voxel included in each segment may be the direction of the representative vector.

The representative vector calculation unit 37 may calculate the representative vector of each segment in all time phases by performing the above-mentioned processing on the movement vectors of all time phases.

The representative vector for each segment calculated by the representative vector calculation unit 37 is stored in the motion analysis data storage unit 19.

The dyssynchrony region extraction unit 39 may specify the dyssynchrony region in the atrium or ventricle based on the representative vector or movement vector of each region and the barycentric position of the left and right atriums or ventricles. The synchronization failure area extraction unit 39 may use either the movement vector of the movement vector data storage unit 15 or the representative vector of the segment data storage unit 17. The following description may be described by taking the case of determining the synchronization failure in the segment unit using the representative vector as an example, but it is also possible to determine the synchronization failure in the voxel unit using the movement vector in the same method. it can.

The out-of-synchronization region extraction unit 39 has a motion direction identification unit 41 and a vector length identification unit 43.

The movement direction identification unit 41 may identify the direction of movement of the segment or voxel. The motion direction identifying unit 41 may identify the motion direction of the segment or voxel, for example, based on the difference between the direction of the representative vector or the movement vector of each region and the direction from each region to the position of the center of gravity. Alternatively, the motion direction identifying unit 41 identifies, for each region, a rotation angle indicating the angle between the representative vector or movement vector of each region and the vector from each region to the position of the center of gravity, and identifies the motion direction of the segment or voxel. May be.

The movement direction identification unit 41 identifies the direction from each segment or voxel toward the center of gravity position (eg, the direction from the center of the segment or voxel toward the center of gravity position) based on the center of gravity position calculated by the center of gravity calculation unit 33, for example. To do. Furthermore, the motion direction identification unit 41 identifies the deviation between the direction of the representative vector of each segment or voxel and the direction from each segment or voxel toward the center of gravity position. The deviation may be represented by, for example, an angle (rotation angle), or may be a difference vector with respect to the vector toward the center of gravity (the vector length is fixed at a predetermined length).

FIG. 6 is an explanatory diagram of the rotation angle.

In the same figure, the position St of the time phase Pht and the position St+1 of the time phase Pht+1 of a certain segment are shown. The angle θ between the direction from the segment St of the time phase Pht toward the center of gravity G and the representative vector V2 of the segment St is the rotation angle.

The vector length specifying unit 43 calculates the vector length (Euclidean distance) of the representative vector or the movement vector.

The processing results of the motion direction specifying unit 41 and the vector length specifying unit 43 are stored in the motion analysis data storage unit 19.

FIG. 7 shows an example of the data structure of the motion analysis data 190. The motion analysis data 190 is data indicating the motion characteristics of each heart phase and each segment. In the example of the figure, an example of motion analysis data 190 in segment units is shown.

As shown in the figure, the motion analysis data 190 has a time phase 191, a segment ID 192, a representative vector 193, a rotation angle 194, and a vector length 195 as data items.

The representative vector 193 is a representative vector of each segment calculated by the representative vector calculation unit 37.

The rotation angle 194 is a rotation angle between the direction of the representative vector of each segment calculated by the motion direction specifying unit 41 and the direction toward the center of gravity.

The vector length 195 is the vector length of the representative vector of each segment calculated by the vector length specifying unit 43.

The out-of-synchronization region extraction unit 39 may specify the out-of-synchronization region based on the movement direction or the rotation angle specified by the movement direction specifying unit 41, or the amount of deviation of the vector.

For example, the out-of-synchronization region extraction unit 39 may identify the outlier of the rotation angle as described below, and the segment or voxel of the outlier may be set as the out-of-synchronization region. That is, when the rotation angle is larger than the angle that is the predetermined threshold value, the synchronization failure region extraction unit 39 may extract the segment or voxel having the rotation angle as the synchronization failure region. The angle serving as the threshold may be predetermined, for example, or a plurality of segments or voxels (all segments (voxels) or target segments (voxels) around the same time phase or a plurality of time phases may be used. It may be determined based on the average value and standard deviation of the rotation angle of the segment (voxel).

The rotation angle representing each segment is calculated from the representative vector as described above, and based on the rotation angle obtained for each voxel, the rotation angle representing each segment (for example, the average rotation angle) is determined. May be.

The out-of-synchronization region extraction unit 39 may specify the outlier of the difference vector as follows, and use the segment or voxel of the outlier as the out-of-synchronization region. That is, when the length of the difference vector is larger than the length that becomes the predetermined threshold value, or when the direction of the difference vector is out of the range of the predetermined direction, the synchronization failure region extraction unit 39 causes the difference vector You may extract the segment or voxel which has. The range of the length or the direction to be the threshold value may be, for example, predetermined, or a plurality of segments or voxels of the same time phase or multiple time phases (entire segment (voxel) or target segment (voxel) periphery) May be determined based on the average value and standard deviation of the lengths of the difference vectors of some of the segments (voxels), or the average value and standard deviation of the directions.

The out-of-synchronization region extraction unit 39 may specify the region in which the length of the representative vector or movement vector of each region is equal to or greater than a predetermined threshold value as the in-synchronization region, based on the processing result of the vector length identification unit 43. ..

The out-of-synchronization region extraction unit 39 may specify, for example, an outlier of the length of the representative vector or the movement vector, and use the segment or voxel of the outlier as the out-of-synchronization region. That is, when the length of the representative vector or the movement vector is larger than the length that becomes the predetermined threshold value, the synchronization failure region extraction unit 39 may extract the segment or voxel having the representative vector or the movement vector as the synchronization failure region. The length to be the threshold value may be predetermined, for example, or a plurality of segments or voxels of the same time phase or multiple time phases (all segments (voxels) or a part of the periphery of the target segment (voxel)). It may be determined based on the average value and standard deviation of the length of the representative vector or movement vector of the segment (voxel).

As described above, the synchronization failure region extraction unit 39 may specify the synchronization failure region based on (1) the motion direction of the segment or voxel, or (2) based on the vector length of the representative vector or the movement vector. The dyssynchronized region may be specified, or the region that is dyssynchronized in (3) and (1) and the dyssynchrony in (2) may be the dyssynchrony region.

The display control unit 45 causes the display device 3 to display various display screens. For example, the display control unit 45 causes the display device 3 to display the motion state of the ventricle, the atrium, or the like for which the motion analysis of the heart has been performed, divided into segments.

FIG. 8 is a display example of the motion analysis result of the left ventricle displayed on the display device 3. The example in the figure shows the state of motion for each segment in a certain time phase. In the same figure, the motion direction (direction of the representative vector) of each segment is indicated by the arrow direction, and the strength of motion (vector length of the representative vector) is indicated by the thickness (or length) of the arrow. Each segment may be color-coded and displayed based on the direction and intensity of movement.

The display control unit 45 may display the display of the ventricular motion analysis results side by side for each time phase as shown in the figure, or may switch the display in the order of the time phases and display them as a moving image.

According to this embodiment, the movement and the strength of the myocardium are visually and quantitatively determined by tracking the movements of the three-dimensional voxels by utilizing the clear contrast of the living tissue of the MR image and the high spatial resolution. Can be displayed on.

FIG. 9 is a flowchart showing an example of the determination process of the dyssynchrony region of the left ventricle.

The out-of-synchronization region extraction unit 39 identifies the processing target time phase Phx (S201).

The center-of-gravity calculator 33 specifies the center-of-gravity position of the left ventricle based on the expansion data 130 of the processing target time phase Phx (S203).

The movement direction identification unit 41 refers to the segment data storage unit 17, calculates the average value and the standard deviation of the rotation angle of the segment of the processing target time phase Phx, and determines the rotation angle threshold value (S205).

The out-of-synchronization region extraction unit 39 compares the threshold value determined in step S205 with the rotation angle of each segment and sets a segment with a rotation angle larger than the threshold value as a synchronization failure candidate region (S207).

The vector length specifying unit 43 refers to the segment data storage unit 17, calculates the average value and standard deviation of the vector length of the representative vector of the segment of the processing target time phase Phx, and determines the vector length threshold value (S209). ).

The vector length identifying unit 43 compares the threshold determined in step S209 with the vector length of the representative vector of each segment, and sets the segment having a vector length larger than the threshold as a synchronization failure candidate area (S211).

The synchronization failure region extraction unit 39 specifies the synchronization failure region from the synchronization failure candidate regions in steps S207 and S211 (S213). For example, the synchronization failure region extraction unit 39 may set the region that is a candidate in both steps S207 and S211 or the region that is a candidate in one of steps S207 and S211 as the synchronization failure region.

By repeating steps S201 to S213 for all the time phases, it is possible to specify in which region of which time phase the synchronization failure has occurred over one cycle of the heartbeat.

According to the present embodiment, not only when there is a deviation in the timing of contraction of the heart, but also when there is no deviation in the timing of contraction, the region where the direction of contraction (exercise) is displaced or the strength of contraction (exercise) ( A region whose vector length) deviates from other regions can also be extracted as a synchronization failure region.

Furthermore, if a technique for quantitatively identifying a dyssynchronized region is established by the present embodiment, a new index of dyssynchrony such as CRT (cardiac resynchronization therapy), valve replacement surgery, or adaptation criteria for heart transplant surgery will be provided. It becomes possible to do.

The embodiments of the present invention described above are examples for explaining the present invention, and are not intended to limit the scope of the present invention only to those embodiments. Those skilled in the art can implement the present invention in various other modes without departing from the gist of the present invention.

1 Medical Image Processing Device 10 Medical Image Processing Device Main Body 11 Imaging Data Storage Unit 13 Extended Data Storage Unit 15 Movement Vector Data Storage Unit 17 Segment Data Storage Unit 19 Motion Analysis Data Storage Unit 31 Interpolation Processing Unit 33 Centroid Calculation Unit 35 Movement Vector Calculation Unit 37 representative vector calculation unit 39 synchronization failure region extraction unit 41 motion direction identification unit 43 vector length identification unit 45 display control unit

Claims (10)

A computer program for image processing performed by a computer that divides one cycle of a heartbeat into a plurality of time phases and stores three-dimensional volume data of the heart in each time phase,
For at least one region of the wall of the atrium or ventricle in the three-dimensional volume data, a vector indicating the movement from the first time phase to the second time phase after the first time phase is specified for each area. Vector specifying step
Based on the vector of each region and the barycentric position of the atrium or ventricle, a dyssynchronization region identifying step of identifying a region of dyssynchrony in the atrium or ventricle, causing the computer to execute :
The dyssynchrony region specifying step, the the direction of the vector of each region, based on the difference between the direction to the center-of-gravity position from the respective regions, the computer program that identifies the area of the dyssynchrony. The dyssynchronization region identification step,
Wherein the vector of each area, the angle was specified for each region of the vector directed to the center-of-gravity position from the respective regions, said angle identifying a region less than a predetermined threshold as a region of the dyssynchrony claim 1 Computer program. The dyssynchronization region identification step,
The computer program according to claim 2 , wherein a region in which the angle is equal to or greater than a predetermined threshold value and a vector length of each region is equal to or greater than a predetermined threshold value is specified as the synchronization failure region. Each of the regions consists of one voxel,
The vector specifying step specifies a vector for each voxel, a computer program according to any of claims 1-3. Further comprising the step of identifying a representative vector for each segment having a plurality of voxels based on the vector of each voxel,
The computer program according to claim 4 , wherein the synchronization failure region specifying step specifies the synchronization failure region in segment units based on the representative vector. The dyssynchronization region identification step,
The vector of each voxel and the angle between the vector from each voxel to the center of gravity position is specified for each voxel,
Based on the angle of each voxel, specify a representative angle for each segment having a plurality of voxels,
The computer program according to claim 4 , wherein a segment whose representative angle of each segment is equal to or larger than a predetermined threshold value is specified as the area of the synchronization failure. Each of the regions is a segment having a plurality of voxels,
The vector specifying step specifies a vector for each of the segments, computer program according to any of claims 1-3. The method further comprising: interpolating one or more extended slice images between adjacent slice images from a plurality of slice images obtained by imaging the heart to generate the three-dimensional volume data. The computer program according to any one of to 7 . A storage unit that divides one cycle of the heartbeat into a plurality of time phases and stores three-dimensional volume data of the heart in each time phase,
For at least one region of the wall of the atrium or ventricle in the three-dimensional volume data, a vector indicating the movement from the first time phase to the second time phase after the first time phase is specified for each area. Vector specifying means to
Based on the vector of each region and the barycentric position of the atrium or ventricle, a dyssynchrony region identifying means for identifying a region of dyssynchrony in the atrium or ventricle ,
The dyssynchrony area specifying means, said the direction of the vector of each region, based on the difference between the direction to the center-of-gravity position from each region, the image processing apparatus that identifies an area of the dyssynchrony. An image processing method performed by a computer which divides one cycle of a heartbeat into a plurality of time phases and stores three-dimensional volume data of the heart in each time phase,
The vector identifying means moves from the first time phase to the second time phase after the first time phase for each of one or more areas of the wall of the atrium or ventricle in the three-dimensional volume data. Identifying a vector indicating
Dyssynchrony area specifying means, and vector of each region, based on the gravity center position of the atrial or ventricular, have rows, identifying a region of the synchronization failure in the atrium or ventricle,
The step of identifying the area of dyssynchrony is a method of identifying the area of dyssynchrony based on the difference between the direction of the vector of each area and the direction from each area to the position of the center of gravity .