JP2017093729A - Medical image processing device, medical image processing method, and medical image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a medical image processing device capable of easily and visually recognizing a correspondence relation of cross-sectional images at each point of a path of a tissue.SOLUTION: The medical image processing device acquires volume data of a living body, extracts a path of a tissue included in the volume data, determines a main part and a non-main part in the path, makes a running direction of the path a normal direction, and generates a first cross-sectional image with a first plane including a first point included in the main part of the path as a cross section and a second cross-sectional image with a second plane including a second point included in the non-main part of the path as a cross section. A roll of the first cross-sectional image is uniquely determined on the basis of a roll of a prescribed cross section, and a depth direction of the first cross-sectional image is uniquely determined on the basis of a depth direction of the prescribed cross section.SELECTED DRAWING: Figure 7

Description

  The present disclosure relates to a medical image processing apparatus, a medical image processing method, and a medical image processing program.

  A conventional medical image processing apparatus can generate a path as a center line of various tissues included in volume data related to a medical image. Various tissues include blood vessels and bronchi. As a conventional medical image processing apparatus, an image display apparatus that displays a traveling state of a path of various tissues is known (see Patent Document 1).

  This image display device discloses a simple method for creating a path as a center line of a tissue included in volume data obtained by stacking a plurality of CT tomographic images by a user operation. Further, the image display device uses an MPR (Multi Planer Reconstruction) method or an MPR as an image of a cross section including the attention point and having the tangent direction of the path at the attention point as a normal line as the attention point progresses on the path. Sequential display by improved method.

US Patent Application Publication No. 2008/0008368

  In the image display device described in Patent Document 1, when a cross-sectional image is displayed in order while advancing a point of interest on a path, if the direction of a blood vessel path changes, the cross-sectional image is rotated by 180 degrees or horizontally reversed. May occur. Therefore, when the user confirms the display of the image, it is difficult to recognize the correspondence between the images before and after inversion, and there is a possibility that misdiagnosis and mishandling by the user may occur.

  The present disclosure has been made in view of the above circumstances, and provides a medical image processing apparatus, a medical image processing method, and a medical image processing program that make it easy to visually recognize the correspondence of cross-sectional images at each point of a tissue path. To do.

  The medical image processing apparatus according to the present disclosure includes a port and a processor. The port acquires biological volume data. The processor extracts a path of the organization included in the volume data and determines a main part and a non-main part in the path. Based on the volume data, the processor generates a first cross-sectional image in which the traveling direction of the path is a normal direction and the first surface including the first point included in the main part of the path is a cross-section. Based on the volume data, the processor generates a second cross-sectional image in which the traveling direction of the path is a normal direction and a second surface including a second point included in a non-main part of the path is a cross-section. The roll of the first cross-sectional image is uniquely determined based on the roll of the predetermined cross-section. The depth direction of the first cross-sectional image is uniquely determined based on the depth direction of the predetermined cross-section.

  A medical image processing method of the present disclosure is a medical image processing method in a medical image processing apparatus, which acquires volume data of a living body, extracts a path of a tissue included in the volume data, and extracts a main part and a non-main part in the path And based on the volume data, generate a first cross-sectional image in which the traveling direction of the path is the normal direction and the first surface including the first point included in the main part of the path is a cross-section. Based on the volume data, a second cross-sectional image is generated in which the traveling direction of the path is a normal direction and a second surface including a second point included in a non-main part of the path is a cross-section. The roll of the first cross-sectional image is uniquely determined based on the roll of the predetermined cross-section, and the depth direction of the first cross-sectional image is uniquely determined based on the depth direction of the predetermined cross-section.

  The medical image processing program of the present disclosure is a program for causing a computer to execute the medical image processing method.

  According to the present disclosure, it is possible to easily visually recognize the correspondence relationship of the cross-sectional images at each point of the tissue path.

1 is a block diagram illustrating a configuration example of a medical image processing apparatus according to a first embodiment. The flowchart which shows the operation example by the medical image processing apparatus at the time of determining the main part of a path | pass Schematic diagram showing an example of each point on the path and the sum of angles at each point Schematic diagram showing an example of each point on the path and the difference in the sum of angles at two adjacent points including each point Schematic diagram showing an example of each point on the path and the segment to which each point belongs Schematic diagram showing an example of each point of the path and the segment to which each point belongs when the threshold is set to double The flowchart which shows the operation example by the medical image processing apparatus at the time of determining the roll reference | standard direction and flip reference | standard direction of main parts Explanatory diagram for determining the roll reference direction and flip reference direction in the main part The flowchart which shows the operation example by the medical image processing apparatus at the time of determining the roll reference | standard direction and flip reference | standard direction of a non-main part Explanatory drawing for determining roll reference direction and flip reference direction in non-main part The schematic diagram which shows the image which shows the path | pass of a tubular structure | tissue, and the SA image cut | disconnected by the SA cross section in a path | pass. Conceptual diagram showing the rotation direction of roll, pitch, and yaw Explanatory drawing when the downward direction in the SA section is always the direction close to the vertical downward direction as the roll reference direction Explanatory drawing when flip reference orientation is always the orientation when looking at the head from the foot of a living body (A)-(D) Explanatory drawing of a roll (A), (B) explanatory drawing of flip

  Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

(Background to obtaining one form of the present disclosure)

  An image of a vertical cross section at an arbitrary position in a path as a center line of various tissues included in the volume data is also referred to as an SA (Short Axis) image. The SA image is generated from volume data generated by a CT (Computed Tomography) apparatus. The SA image may be generated by an MPR (Multi Planar Reconstruction) method. An image generated by MPR is an image reconstructed by cutting out volume data in an arbitrary cross section.

  FIG. 11 is a schematic diagram showing an image GZ11 showing the path Pa1 representing the travel of the tubular tissue t1, and an SA image GZ12 in the SA cross section cs1 in the path Pa1. That is, the SA image is an image of the SA cross section. Tubular tissue may include blood vessels, bronchi, and gastrointestinal lumens.

  In the SA cross section, a tangent line at a point of interest on the path can be defined as a normal line. On the other hand, a reference direction (hereinafter also referred to as a roll reference direction) for determining a rotation angle (roll angle) about the normal line in the SA image is roughly determined in a conventional medical image processing apparatus. (For example, in the vertical direction). In addition, a normal direction reference direction (hereinafter also referred to as a flip reference direction) for determining the depth direction of an image is roughly determined in a conventional medical image processing apparatus (for example, a head direction). When the depth direction of the image is reversed, the left and right sides of the image are reversed.

  Therefore, if the roll reference direction and the flip reference direction are easily fixed in accordance with the bending of the path, when the SA image along the path is displayed in the forward direction, the SA image is discontinuously rotated or horizontally reversed. There is. In this case, it becomes difficult for the user to track the tissue to be observed.

  FIG. 12 is a conceptual diagram showing the rotation direction of the roll, pitch, and yaw. In FIG. 12, each rotation direction is shown by taking an aircraft body as an example. Roll refers to the aircraft body rotating clockwise or counterclockwise about the body axis of the aircraft body. The pitch refers to rotation in the vertical direction along a cross section having the body axis of the aircraft body as a normal line. Yaw refers to rotation in the left-right direction along a cross section with the body axis of the aircraft body as the normal.

  The expressions of roll, pitch, and yaw are often used for setting the viewpoint (camera) in a three-dimensional image. In the cross-sectional image, although the viewpoint cannot be directly considered, the roll, pitch, and yaw can be considered as rotations around the normal of the cross-section and the normal of the cross-sectional image. Therefore, if the camera advances the tissue path, rotating (twisting) the camera clockwise or counterclockwise around the path corresponds to a roll.

  In medical 3D images, rolls have been neglected with respect to pitch and yaw. This is because the cross-section of the cross-sectional image does not change due to the roll, and thus the amount of information included in the image does not change essentially even if the roll changes. For example, it is convenient to define the roll angle as the angle of the roll with respect to the reference direction (that is, the roll reference direction) of the cross-sectional image (SA image) with the path as a normal line.

  When the roll reference direction is always close to the vertical downward direction (downward direction in the SA image), the downward direction in the SA image cannot be defined when the path direction (normal direction with respect to the SA cross section) becomes the vertical direction. Further, when the path is folded back through a state parallel to the vertical direction, as a result of maintaining the downward direction in the SA image, the contents of the SA image displayed in forward rotation are rotated approximately 180 degrees around the center point of the SA image. Sometimes. In the case of an aircraft body, this means that the aircraft is turned upside down when it is turned over, but this is equivalent to correcting the top and bottom of the aircraft by making an immerman turn. However, when the SA image is displayed in order, such a 180 degree rotation is not preferable. Note that the direction close to the downward direction in the vertical direction is for considering a case where the direction is not parallel to the downward direction in the vertical direction.

  FIG. 13 is an explanatory diagram when the downward direction (roll reference direction) in the SA image of the SA cross section cs2 is always set to a direction close to the vertical downward d1. In FIG. 13, the range D1 of the path Pa2 that has no problem even if the downward direction of the SA image of the SA section cs2 is close to the vertical downward direction d1, and the downward direction of the SA image of the SA cross section cs2 is the downward direction of the vertical direction d1. Then, there is a range D2 of the problematic path Pa2. In FIG. 13, in the section where the path Pa2 extends along the vertical direction (vertical direction), the downward direction of the SA image of the SA cross section cs2 cannot be defined, the roll of the SA image becomes unstable, and further, the path Pa2 turns. The SA image is rotated 180 degrees.

  In addition, when the flip reference direction is always close to the direction of viewing the head from the foot of the living body, the head is viewed from the foot when the SA cross section with the path as the normal is parallel to the flip reference direction. The direction cannot be defined. For example, when the direction of the path is folded from the direction from the foot to the head to the direction from the head to the foot, the position of the point of interest on the path for viewing the SA image changes. ) In addition, it describes that it is the direction close | similar to the direction which looked at the head from the foot part of the biological body in order to consider the case where it is not parallel to the direction which looked at the head from the foot part of the biological body.

  FIG. 14 is an explanatory diagram in the case where the flip reference direction (the depth direction of the SA image of the SA cross section cs3) is always the direction close to the direction d2 when the head is viewed from the foot of the living body. In FIG. 14, there is no problem even if the flip reference direction is close to the direction d2 when the head is viewed from the foot of the living body, and the flip reference direction is close to the direction d2 when the head is viewed from the foot of the living body. There is a range D4 where there is a problem if it is oriented. In FIG. 14, when the path Pa3 is folded back from the body axis direction (up and down direction) and becomes the direction perpendicular to the body axis direction (left and right direction), the SA image is reversed left and right. Therefore, the SA image G13 in the unaffected range D3 and the SA image G14 in the disturbed range D4 are symmetric images. In addition, direction d2 which looked at the head from the leg part of the biological body is along the body axis direction.

  15A to 15D are explanatory diagrams of the roll. 16A and 16B are explanatory diagrams of flipping. That is, when the path rolls, the SA image rotates about the center point O1 of the SA image as the rotation center. When flipped, the SA image is horizontally reversed with respect to the reference line L1 of the SA image.

  Therefore, when the SA images are sent in order along the path and observed, the SA image is rotated 180 degrees by changing the downward direction of the SA image, or the SA image is inverted horizontally by changing the depth direction of the SA image. Sometimes. Therefore, it is difficult to recognize the correspondence between the cross-sectional images before and after the image rotation or the image inversion occurs, and there is a possibility that misdiagnosis or erroneous treatment by the user may occur.

  Hereinafter, a medical image processing apparatus, a medical image processing method, and a medical image processing program that make it easy to visually recognize the cross-sectional image correspondence at each point of the tissue path in the volume data will be described.

(First embodiment)
[Configuration]
FIG. 1 is a block diagram illustrating a configuration example of a medical image processing apparatus 100 according to the first embodiment. The medical image processing apparatus 100 includes a port 110, a UI (User Interface) 120, a display 130, a processor 140, and a memory 150. A CT apparatus 200 is connected to the medical image processing apparatus 100. The medical image processing apparatus 100 acquires volume data from the CT apparatus 200 and performs processing on the acquired volume data.

  The CT apparatus 200 irradiates a living body with X-rays and captures an image (CT image) using the difference in X-ray absorption by tissues in the body. A plurality of CT images may be taken in time series. The CT image forms volume data including information on an arbitrary location inside the living body. By capturing a CT image, a pixel value (CT value) of each pixel (voxel) in the CT image is obtained. The CT apparatus 200 transmits volume data as a CT image to the medical image processing apparatus 100 via a wired line or a wireless line.

  The port 110 acquires volume data as a CT image. The acquired volume data may be immediately sent to the processor 140 and processed in various ways, or may be stored in the memory 150 and then sent to the processor 140 and processed in various ways when necessary.

  The UI 120 may include a touch panel, a pointing device, a keyboard, and voice input. The UI 120 receives an arbitrary input operation from the user of the medical image processing apparatus 100. The UI 120 may accept a region designation operation. The user may include, for example, a doctor, a radiologist, and an interpretation doctor.

  The display 130 may include an LCD (Liquid Crystal Display). The display 130 displays various information. Various types of information may include a three-dimensional image or a two-dimensional image, and may include a volume rendering image, a surface rendering image, an MPR image, and an SA image.

  The memory 150 includes various ROMs (Read Only Memory) and RAMs (Random Access Memory). The memory 150 stores various information and programs. The various types of information may include volume data acquired by the port 110, an image generated by the processor 140, and setting information set by the processor 140.

  The processor 140 includes a CPU (Central Processing Unit), a DSP (Digital Signal Processor), or a GPU (Graphics Processing Unit). The processor 140 performs various processes and controls by executing a medical image processing program stored in the memory 150. The processor 140 supervises each unit of the medical image processing apparatus 100.

  The processor 140 extracts the path of the tissue in the body from the volume data acquired by the port 110 and the processor 140 by a known method. Thereby, the user can grasp | ascertain the running state of various structures | tissues (blood vessel, stomach, intestines, etc.). The known method may include obtaining a tissue center line by extracting various tissue regions and thinning the extracted regions.

  The processor 140 determines main parts and non-main parts (parts other than the main part) described later in the extracted path, and generates an SA image at each point in the main part and the non-main part. The main part of the path has less change in the traveling direction of the path than the non-main part, that is, the path is less bent. The method for determining the roll reference direction and flip reference direction of the SA image differs between the main part and the non-main part. The SA image may be an MPR image (cross-sectional image) obtained from volume data. The SA image may include bones, blood vessels, and organs, and may include a cross-sectional image of tubular tissue.

  Further, the processor 140 determines a roll of the SA image based on a roll having a predetermined cross section (for example, an axial cross section, a sagittal cross section, or a coronal cross section) described later. The roll of the SA image refers to rotating the SA image around the normal line of the SA image. In addition, a roll of a cross-sectional image with respect to the normal line of the predetermined cross-section is determined for the predetermined cross-section, and the cross-sectional image of the predetermined cross-section is normally displayed according to the roll of the cross-sectional image. This is called a roll having a predetermined cross section. In the SA image, the comparison of the degree of rotation of the image with the normal line of the SA image as an axis from the roll of a predetermined section is also referred to as the roll angle of the SA image. When a reference direction (for example, a roll reference direction) is obtained using a roll having a predetermined cross section, the roll angle of the SA image can be expressed using the reference direction. Therefore, when the roll reference direction is changed, the degree of roll is different. The roll can also be defined by giving a twist to the path.

  Further, the processor 140 determines the depth direction (flip reference direction) of the SA image based on the depth direction of the predetermined cross section. The processor 140 may determine the depth direction of the SA image based on the positive / negative of the inner product of a vector and a predetermined cross section described later.

[Operation etc.]
Next, an operation example of the medical image processing apparatus 100 will be described.
First, an example of an outline of the operation of the medical image processing apparatus 100 will be described.

  In the medical image processing apparatus 100, the processor 140 derives the main part of the path. The processor 140 determines a roll reference direction and a flip reference direction based on any one of an axial image, a sagittal image, and a coronal image in at least a main part of the path, and on the path. An SA image of the SA cross section including any of the points is generated. Arbitrary points on the path are sequentially moved along the path to generate a plurality of SA images.

  Axial images, sagittal images, and coronal images are often used for diagnosis. An axial image is an image of an axial cross section. The sagittal image is an image of a sagittal section. A coronal image is an image of a coronal cross section.

  The processor 140 sets the same flip reference direction as that of the main part for the SA image including the points arranged in the front and rear along the path in the non-main part (parts other than the main part) in the path. In addition, the processor 140 sets a roll reference direction in which a difference in the downward direction of the SA image at an adjacent point on the path (for example, an angle formed by both SA cross sections) is equal to or less than a predetermined value (for example, minimum). The processor 140 generates an SA image of the SA cross section including an arbitrary point on the path according to the determined roll reference direction and flip reference direction. Even in the non-main part, arbitrary points on the path are sequentially moved along the path, and a plurality of SA images are generated.

  Accordingly, the medical image processing apparatus 100 can suppress a sudden change (discontinuous roll such as 180 ° rotation) of the SA image by sequentially creating and displaying the SA image along the path, and can perform left and right etc. Can be suppressed. Further, the medical image processing apparatus 100 can generate an SA image that approximates an image familiar to the user in the main part of the path.

  In addition, when there are a plurality of paths whose shapes are approximated or whose running directions are approximated, a specific one of the axial image, the sagittal image, and the coronal image is used as a reference. Therefore, in the SA image in each pass, the roll angle of the SA image and the depth direction (flip reference direction) of the SA image substantially coincide. Therefore, comparative observation of these plurality of SA images is facilitated. The path that approximates a plurality of shapes or approximates the running direction may include left and right femoral arteries and blood vessels on both legs.

  In addition, when the shape approximates or the traveling direction approximates in different paths extracted from different volume data, when there are a plurality of paths that approximate the shape or approximate the traveling direction, all are axial images, sagittal images, And a specific one in the coronal image. Therefore, in the SA image in each pass, the roll angle of the SA image and the depth direction (flip reference direction) of the SA image substantially coincide. Therefore, comparative observation of these plurality of SA images is facilitated. Furthermore, it becomes easier for the user to grasp a routine for sequentially diagnosing the patient. Different paths extracted from different volume data may include paths of the same tissue of different patients and paths taken at different times of the same patient.

  Next, an operation example of the medical image processing apparatus 100 when determining the main part of the path will be described. FIG. 2 is a flowchart showing an operation example by the medical image processing apparatus 100 when determining the main part of the path. At the start of FIG. 2, a path to be observed (also simply referred to as a path) has already been obtained from the volume data.

  First, the processor 140 determines, for each point on the path, the tangent direction (path direction) of the path and the three unit elements (1, 0, 0), (0, 1, 0), (0, 0, 1). And the sum of the angles (scalar sum) is calculated (S11).

  FIG. 3 is a schematic diagram illustrating an example of each point on the path and a sum of angles at each point. In FIG. 3, the horizontal axis indicates the distance (unit: mm) from the reference position of each point of the path (for example, the end point of the path), and the vertical axis indicates the sum of the angles (unit: radians (rad)) at each point on the path. ).

  The processor 140 selects two adjacent points on the path, and calculates the difference between the sums of the angles calculated in S11 at the two selected points (S12). The difference in the sum of angles indicates the rate of change in the normal running direction.

  FIG. 4 is a schematic diagram illustrating an example of each point on the path and a difference in the sum of angles at two adjacent points including each point. In FIG. 4, the horizontal axis indicates the distance (unit: mm) from the reference position of each point of the path, and the vertical axis indicates the difference (unit: radians) of the sum of the angles at the two points.

  The processor 140 segments each point on the path in units in which the difference in the sum of angles is equal to or less than the threshold th1 (S13). That is, when the difference in the sum of angles is larger than the threshold value th1, the processor 140 divides the segment at the two points. The threshold th1 may be a value 0.001. Therefore, it is divided into segments for each range where the change in the traveling direction of the path is small. When the traveling direction of the path is bent more than a predetermined reference, it is divided into segments and a new segment is obtained from the following points.

  FIG. 5 is a schematic diagram illustrating an example of each point on the path and a segment to which each point belongs. In FIG. 5, when a segment is updated, the segment ID (for example, segment identification number = segment number) is updated one by one. In FIG. 5, the horizontal axis indicates the position (unit: mm) of each point of the path with respect to the reference position, and the vertical axis indicates the segment ID to which each point belongs.

  The processor 140 acquires the longest segment among the separated segments (S14). In addition, even if it is not the longest, any one of the segments having a predetermined length or more may be used.

  The processor 140 determines whether or not the length of the longest segment is longer than the threshold th2 (S15). That is, the threshold th2 is a threshold compared with the longest segment. If the length of the longest segment is longer than the threshold th2, the processor 140 sets this segment as the main part of the path and sets the other segments as non-main parts of the path (S16).

  On the other hand, when the length of the longest segment is equal to or smaller than the threshold th2, the processor 140 sets the threshold th2 to double (S17). Note that double is an example, and other multiples may be used. As the threshold th2 is larger, segmentation becomes difficult, so the segment length becomes longer and the total number of segments in the path decreases. On the other hand, the smaller the threshold th2, the easier it is to segment, so the segment length becomes shorter and the total number of segments in the path increases.

  FIG. 6 is a schematic diagram illustrating an example of each point of a path and a segment to which each point belongs when the threshold th2 is set to double. It can be seen that the largest segment has become longer. In FIG. 6, when the segment is updated, the segment ID is updated one by one. The horizontal and vertical axes in FIG. 6 are the same as those in FIG.

  According to the processing in FIG. 2, the medical image processing apparatus 100 derives the angle between the path direction and the unit element, and determines the difference between the angles derived at adjacent points, thereby determining the path travel direction. The change is determined, and the path is segmented based on the change in the traveling direction of the path. Further, a segment in which the change in the direction of travel for a sufficiently long distance is smaller than a predetermined reference is a main part, and the other segments are non-main parts. Therefore, it is possible to determine a main part in which the same roll reference direction and flip reference direction are expected to be applicable.

  In addition, the medical image processing apparatus 100 can visually recognize the continuity of the path even when the SA images are sequentially displayed along the path by setting the length of the segment, which is the main part of the path, to a predetermined length or more. This makes it possible to lengthen the easy range and to make it easier for the user to visually recognize the running state of the pass and whether or not the pass is the same.

  Next, an operation example of the medical image processing apparatus 100 when determining the roll reference direction and the flip reference direction of the main part will be described. FIG. 7 is a flowchart illustrating an operation example of the medical image processing apparatus 100 when determining the roll reference direction and the flip reference direction of the main part. FIG. 8 is an explanatory diagram for determining the roll reference direction and the flip reference direction in the main part.

  In the present embodiment, the roll reference direction and the flip reference direction are determined according to which predetermined cross section (here, the axial cross section, the coronal cross section, or the sagittal cross section) the SA cross section is close to. Which predetermined cross section is close depends on the size of the inner product described later and depends on the angle formed by the SA cross section and the predetermined cross section. It can be said that the smaller this angle is, the closer the SA cross section is to the predetermined cross section. Information about the roll reference direction and the flip reference direction is held in the memory 150. The predetermined cross section does not define coordinates included in the predetermined cross section, and the predetermined cross section includes a cross section obtained by translating some specific cross section. This is because the axial cross section, coronal cross section, or sagittal cross section includes a cross section obtained by translating the axial cross section.

  When the SA cross section is close to the axial cross section, the downward direction of the SA image is a direction close to the vertical downward direction. A direction close to the vertical direction can be set as the roll reference direction. When the SA cross section is close to the axial cross section, the depth direction (flip reference direction) of the SA image is one of the traveling directions of the path, and the main part of the path is directed from the foot to the head. . This is because this is a standard axial image display. The case where the SA cross section is close to the axial cross section includes the case where the SA cross section and the axial cross section are parallel.

  When the SA cross section is close to the coronal cross section, the downward direction of the SA image is a direction close to the direction of the foot along the body axis direction. A direction close to the direction of the foot along the body axis direction can be set as the roll reference direction. When the SA cross section is close to the coronal cross section, the depth direction (flip reference direction) with respect to the SA image is one of the traveling directions of the path, and in the main part of the path, from the left side of the living body to the right side of the living body approximately orthogonal to the body axis direction. It is said that the direction is toward. This is because this is a standard coronal image display. When the SA section is close to the coronal section, the SA section and the coronal section are parallel.

  When the SA cross section is close to the sagittal cross section, the downward direction of the SA image is close to the direction of the foot along the body axis direction. A direction close to the direction of the foot along the body axis direction can be set as the roll reference direction. When the SA cross section is close to the sagittal cross section, the depth direction (flip reference direction) with respect to the SA image is one of the traveling directions of the path, and in the main part of the path, from the front of the living body to the back of the living body approximately perpendicular to the body axis direction. It is said that the direction is toward. This is because this is a standard sagittal image display. In addition, the case where the SA cross section is close to the sagittal cross section includes the case where the SA cross section and the sagittal cross section are parallel.

  First, the processor 140 calculates a vector Ps_Pe connecting both ends (Ps, Pe) of the main segment (S21). The vector Ps_Pe is a vector from the start end Ps to the end Pe. Note that the direction from the end point Ps to the end point Pe on the path is a direction arbitrarily given to the path. The direction given to the path is, for example, the order in which the user has arbitrarily clicked, the order of the positions of the data in the memory, or the order in which the tracking algorithm has progressed. However, this order is often not essential in the observation of lesions.

  The processor 140 calculates the absolute value of the inner product of the vector Ps_Pe and the depth direction of the normal to the axial section, coronal section, and sagittal section (S22). Therefore, three calculation results are obtained.

  The processor 140 selects a cross section in which the absolute value of the inner product is maximum among the three calculation results (S23). That is, the processor 140 selects an axial section, a coronal section, or a sagittal section. In FIG. 8, the axial cross section is selected. It can be said that the cross section in which the absolute value of the inner product is the maximum is the cross section in which the angle formed with the SA cross section is the minimum.

  The processor 140 determines whether or not the calculated inner product is a negative value (S24). If the inner product is negative, the processor 140 sets the flip reference direction to be opposite to the direction from the end point Ps to the end point Pe on the path, that is, the direction from the end point Pe to the end point Ps on the path (S25). .

  Therefore, when the inner product is a positive value, in FIG. 8, the flip reference direction coincides with the direction given to the path, and the direction from the foot to the head (the direction from Ps to Pe) in the main part is approximately the same. Become. When the inner product is a negative value, the processor 140 indicates that the flip reference direction is opposite to the direction given to the path, and is also in a direction close to the direction from the foot to the head (Pe to Ps). Direction). Therefore, by giving the depth direction of the SA image according to the flip reference direction, an SA image in a direction in which the head is viewed from the foot is obtained regardless of the direction of the path arbitrarily given to the path. This is the same depth orientation as a standard axial section.

  The processor 140 projects the downward direction (downward vector) of the selected axial section, coronal section, or sagittal section onto the plane defined by the vector Ps_Pe, sets the projected vector as the roll reference direction, and uses the SA image. Is set downward (S26). Note that the plane defined by the vector Ps_Pe is a plane in which the direction of the vector Ps_Pe is the normal direction, and is an SA cross section. The setting information is held in the memory 150.

  In FIG. 8, since the SA section cs11 with the path Pa11 as a normal line is along the direction close to the axial section, the downward direction dr11 of the SA image of the SA section cs11 is projected from the vertically downward dr12. Therefore, the roll reference direction of this SA image is close to the vertical downward direction dr12.

  According to the processing in FIG. 7, the processor 140 determines the roll reference direction and the flip reference direction based on the vector Ps_Pe corresponding to the average of the traveling directions of the points in the main part. The roll reference direction and the flip reference direction are commonly used at each point in the main part. Since the traveling direction of the path in the main part is little changed, by using the roll reference direction and the flip reference direction derived based on the average of the traveling direction of the path, each SA image along the path can be rotated 180 degrees or horizontally reversed. Can be suppressed.

  Next, an operation example of the medical image processing apparatus 100 when determining the roll reference direction and the flip reference direction of the non-main part will be described. FIG. 9 is a flowchart illustrating an operation example performed by the medical image processing apparatus 100 when determining the roll reference direction and the flip reference direction of the non-main part. FIG. 10 is an explanatory diagram for determining the roll reference direction and the flip reference direction in the non-main part.

  First, the processor 140 selects a point included in the non-major part in the path (S31). In the first S31, a point adjacent to the main part whose roll reference direction and flip direction have already been determined is selected. From the second time onward, a point that is adjacent to the previously selected point and whose roll reference direction and flip reference direction are undetermined is selected. Therefore, after the second time, the point is selected by moving in a direction away from the main part.

  The processor 140 sets the variable n to a value 1 (S32).

  The processor 140 acquires the downward vector v_downn and the path direction pvn of the SA image at the point Pn on the path for which the roll reference direction and the flip direction have already been determined (S33).

  In FIG. 10, at the point P1 on the path Pa21, a vector (downward vector v_down1) indicating the downward direction dr21 of the SA image of the SA section cs21 and the direction pv21 of the path Pa21 are acquired. The downward vector v_down1 of the SA image of the SA section cs21 is the value set in S26 of FIG.

  The path direction pvn is the traveling direction of the path Pa21 at the point Pn, and corresponds to the tangential direction at the point Pn. The path direction pvn is acquired by a known method.

  The processor 140 acquires the path direction pv (n + 1) at the point P (n + 1) selected in S31 (S34). Point P (n + 1) is a point adjacent to point Pn. In FIG. 10, the path direction pv2 at the point P2 selected in S31 is acquired.

  The processor 140 sets the flip reference direction to the same direction as the flip reference direction derived in FIG. 7 (S35). That is, the processor 140 does not depend on the direction of the path at any point in the non-main part (the traveling direction of the path), and as the flip reference direction, the direction in which the position of the point of interest on the path advances as in the main part or The direction of attention point position on the path is set in the opposite direction.

  As a result, the medical image processing apparatus 100 changes the reference flip direction in a non-major part of the path depending on the traveling direction of the path, and is discontinuously displayed when the SA image is confirmed along the path. Can be suppressed.

  As for the flip reference orientation, the non-major part is the same as the principal part, that is, the direction in which the point of interest on the path advances, but unlike the principal part, it should not be oriented from the foot to the head. There is also.

The processor 140 calculates the downward vector v_down (n + 1) of the SA image at the point P (n + 1) (S36). In this case, the following (Formula 1) is satisfied.
v_Down (n + 1) × pv (n + 1) = v_Downn × pvn (Expression 1)

  Therefore, the path direction pv (n + 1) at the point P (n + 1) and the downward vector v_down (n + 1) of the SA image are determined. The processor 140 sets the downward vector v_down (n + 1) of the SA image as the roll reference direction.

  In FIG. 10, assuming that the SA vector of the SA cross section cs22 is the downward vector v_Down2, v_Down2 × pv2 = v_Down1 × pv1 is satisfied.

  The processor 140 determines whether or not a point whose roll reference direction and flip reference direction are undetermined remains in a non-major part of the path (S37). When the roll reference direction and the flip reference direction are all set in the non-main part, the processing in FIG. 9 is terminated.

  When a point where the roll reference direction and the flip reference direction are not set remains in the non-main part, the value 1 is added to the variable n (the variable n is incremented) (S38), and the process proceeds to S33. That is, the processor 140 moves to the next point in the path, and similarly sets the roll reference direction and the flip reference direction.

  According to the processing in FIG. 9, the processor 140 sets the roll reference direction so that the vector product of the downward direction of the SA image and the direction of the path at adjacent points in the non-major part of the path is constant. Accordingly, the medical image processing apparatus 100 can sequentially set the change in the roll reference direction of adjacent SA images, can smoothly reproduce the tissue roll change, and can suppress the SA image from being rotated 180 degrees. In addition, the processor 140 matches the flip reference direction of the main part and the non-main part in the pass so as to make the pass direction and the flip reference direction consistent. This can suppress left-right reversal of the SA image.

  The processor 140 generates an SA image based on the roll reference direction and the flip reference direction of the SA image determined at each point in each segment of the path by the processes of FIGS. The display 130 displays the generated SA image under the control of the processor 140.

  The processor 140 may generate an SA image at each point after each point of the path including the main part and the non-main part is determined. The processor 140 generates an SA image at each point of the main part when the roll reference direction and the flip reference direction at each point of the main part are determined, and the roll reference direction and the flip reference at each point of the non-main part. When the orientation is determined, an SA image at each point of the main part may be generated. The processor 140 may determine the roll reference direction and the flip reference direction for each point or a plurality of points in the path, and generate an SA image each time.

[Effects]
As described above, the medical image processing apparatus 100 includes the port 110 and the processor 140. The port 110 acquires biological volume data. The processor 140 extracts a path of the organization included in the volume data, and determines a main part and a non-main part in the path. Based on the volume data, the processor 140 generates a first cross-sectional image in which the traveling direction of the path is a normal direction and the first surface including the first point included in the main part of the path is a cross-section. Based on the volume data, the processor 140 generates a second cross-sectional image in which the traveling direction of the path is a normal direction and the second surface including the second point included in the non-main part of the path is a cross-section. . The roll of the first cross-sectional image is uniquely determined based on the roll of the predetermined cross-section, and the depth direction of the first cross-sectional image is uniquely determined based on the depth direction of the predetermined cross-section.

  The first surface may be an SA cross section. The first cross-sectional image may be an SA image. The second surface may be an SA cross section. The second cross-sectional image may be an SA image. The predetermined cross section may include an axial cross section, a coronal cross section, and a sagittal cross section.

  As a result, the medical image processing apparatus 100 has a small change in the traveling direction in the main part of the path. Therefore, the rolls in the main part are uniquely determined, so that the SA image at each point of the main part rotates 180 degrees. This can be suppressed. In addition, since the traveling direction in the main part of the path has little change, the medical image processing apparatus 100 can uniquely determine the depth direction of the cross-sectional image in the main part, so that the SA image at each point of the main part is left and right. Inversion (flip) can be suppressed. Since the main portion accounts for a large proportion of the entire path, the medical image processing apparatus 100 causes the SA image to become discontinuous, and the visibility is lowered, such as the user misidentifying the positional relationship of the body tissue in each SA image. Can be suppressed. In addition, since the roll and depth direction of the SA image are determined in relation to the predetermined cross section, the user can easily recognize the SA image.

  Thus, the medical image processing apparatus 100 can suppress the 180-degree rotation and the horizontal reversal when displaying the SA image, and can easily recognize the correspondence of the cross-sectional images at each point of the tissue path. Accordingly, it is possible to reduce the possibility of misdiagnosis and mishandling by the user.

  The medical image processing apparatus 100 may uniquely determine the roll of the second cross-sectional image based on the roll of the already determined cross-sectional image in the vicinity of the second point and the travel of the path. The depth direction of the second cross-sectional image may be uniquely determined based on the depth direction of the already determined cross-sectional image in the vicinity of the second point.

  Thereby, in the non-major part of the path, the medical image processing apparatus 100 can reduce the change of the roll of the SA image relative to the roll of the nearby SA image, can smoothly reproduce the roll of the tissue, and rotates the SA image by 180 degrees. This can be suppressed. Further, the processor 140 makes the path direction and the flip reference direction consistent by matching the depth direction of the SA image of the main part and the non-main part in the path with one of the traveling directions of the path. Can be suppressed. In other words, since the left-right direction of the SA image of the non-main part depends on the left-right direction of the SA image of the main part, the user travels along the path (from the foot to the head, from the head to the foot. Etc.).

  In the medical image processing apparatus 100, the predetermined cross section may be a cross section close to a first surface among a plurality of predetermined cross sections.

  Thereby, the user can confirm the SA image based on the familiar direction and direction according to the traveling direction of the path.

  Further, the downward direction or upward direction in the first cross-sectional image and the second cross-sectional image may be a downward direction in the vertical direction or a direction close to upward in the vertical direction.

  When the axial section is selected as the predetermined section close to the SA section, the medical image processing apparatus 100 can set the roll reference direction of the SA section based on the vertical direction downward or the vertical direction upward. Thereby, the user can observe the tissue from a familiar direction.

  The depth direction of the predetermined cross section may be a direction in which the head is viewed from the foot portion of the living body or a direction in which the foot portion is viewed from the head of the living body.

  When the axial cross section is selected as the predetermined cross section close to the SA cross section, the medical image processing apparatus 100 is oriented so that the head is viewed from the foot of the living body, or the foot is viewed from the head of the living body. Based on this, the depth direction of the SA cross section can be set. Thereby, the user can observe the tissue from a familiar direction.

(Other embodiments)
The present disclosure is not limited to the configuration of the above-described embodiment, and any configuration that can achieve the functions shown in the claims or the functions of the configuration of the present embodiment is possible. Is also applicable.

  In addition to the medical image processing apparatus 100, the above embodiment can be expressed as a medical image processing method or a medical image processing program.

  In the above embodiment, in the operation example of the medical image processing apparatus 100, it is exemplified that two adjacent points are selected for various kinds of processing. However, two points adjacent to each other on the path even if they are not strictly adjacent to each other. If it is.

  In the above-described embodiment, the method for determining the roll reference direction and the flip reference direction is divided according to whether it is a main part or a non-main part in the path, but may be divided by other methods. The processor 140 determines any point in the main part in the path by the above-described method of determining the roll reference direction and the flip reference direction in the main part, and points other than the arbitrary point in the main part in the path and the non-main part. May be determined by the method for determining the roll reference direction and the flip reference direction in the non-main part. Note that there may be two arbitrary points in the main part.

  Even in this case, if the roll reference direction and the flip reference direction at an arbitrary point in the main part are determined, the roll reference direction and the flip reference direction at the arbitrary point are the same at other points in the main part. It is expected that. Therefore, the medical image processing apparatus can hardly cause the 180-degree rotation and the left / right inversion.

  Thus, the main part of the path may be an arbitrary range in the path.

  The main part of the path may be at least one arbitrary point in the path.

  Thereby, even if it is difficult to imagine the range of the main part of a complicated and winding path, the user can observe the tissue from a familiar direction at least for a part of the path. Further, the medical image processing apparatus 100 can suppress 180-degree rotation and left-right inversion. For example, when observing the large intestine, it is possible to observe the large intestine starting from the SA image close to the axial image by providing the main part in the anus where the colonoscope is inserted.

  In the above-described embodiment, it is not assumed that the path branches. However, when the path branches, the path may have a common part (trunk part) and a branch part (branch part). In this case, the processor 140 causes the main part of the path to be included in the common part of the path. That is, the processor 140 may extract a common part from a path having a branch, and set the longest segment in the common part as the main part of the path.

  As described above, the path may include a branch part where the path is branched into a plurality of parts and a trunk part where the path is not branched. The trunk portion of the path may be included in the main portion of the path.

  Accordingly, when the user alternately compares the SA images of the branched paths, the user can obtain the SA images having similar angles for each of the branched paths, so that the comparison becomes easy.

  In the above-described embodiment, it is described that the path is divided into the main part and the non-main part. However, a part of the third path other than the main part and the non-main part may exist. For example, the processor 140 may set the third path portion when the segment length is equal to or smaller than a predetermined value. The portion of the third path is, for example, a location where the path is bent largely.

  In the above-described embodiment, an example in which an appropriate image is selected from an axial image, a sagittal image, and a coronal image is shown as a reference. However, a specific image may be set as a reference in advance. For example, in the case of a protocol for observing the aorta, it is most appropriate to observe an axial image, so it is conceivable that the axial image is used as a reference in advance. In transcatheter aortic valve implantation (TAVI), when the transfemoral approach and the transapical approach are selected, a reference image may be determined.

  In the above-described embodiment, the medical image processing apparatus 100 may acquire volume data of the entire living body and perform extraction of the entire path of the living body and generation of an SA image. Further, the medical image processing apparatus 100 may acquire volume data of the entire living body, extract a part of the living body, and generate an SA image. In addition, the medical image processing apparatus 100 may acquire volume data of a part of the living body, extract a part of the path of the living body, and generate an SA image. The part of the living body may be a specific tissue or organ (for example, blood vessel) to be observed. The observation target may be a region of interest (ROI) that is noticed by the user. The designation of a part of the living body may be performed by the UI 120.

  In the above embodiment, the lower direction of the SA image is exemplified as a direction close to any one of the roll reference directions, but the other direction (for example, the upper direction) of the SA image is set as any one of the roll reference directions. Also good.

  In the above embodiment, when the SA cross section is close to the axial cross section, it is exemplified that the roll reference direction is close to the vertical downward direction, but the roll reference direction may be close to the vertical upward direction.

  In the above embodiment, when the SA cross section is close to the axial cross section, it is exemplified that the flip reference direction is close to the direction of viewing the head from the foot, but the flip reference direction is viewed from the head to the foot. The direction may be close to that direction.

  In the above embodiment, when the SA cross section is close to the coronal cross section, it is exemplified that the roll reference direction is a direction close to the foot direction along the body axis direction. The direction may be close to the direction.

  In the above embodiment, when the SA cross section is close to the coronal cross section, it is exemplified that the flip reference direction is a direction close to the direction of viewing the right side of the living body from the left side of the living body orthogonal to the body axis direction. The direction may be close to the direction of viewing the left side of the living body from the right side of the living body orthogonal to the axial direction.

  In the above embodiment, when the SA cross section is close to the sagittal cross section, it is exemplified that the roll reference direction is a direction close to the foot direction along the body axis direction. The direction may be close to the direction.

  In the above embodiment, when the SA cross section is close to the coronal cross section, it is exemplified that the flip reference direction is a direction close to the direction when the back of the living body is viewed from the front of the living body orthogonal to the body axis direction. The direction may be a direction close to the direction of viewing the front of the living body from the back of the living body orthogonal to the axial direction.

  In the above embodiment, the predetermined cross section is an axial cross section, a coronal cross section, or a sagittal cross section. . Moreover, arbitrary cross sections may be sufficient. The predetermined cross section is sufficient if the roll reference direction of the cross section and the depth direction of the cross section can be derived, and coordinates including the surface are not required. It may also be some parameter sufficient to define a predetermined cross section. For example, the predetermined cross section may be defined by the combination of the roll reference direction and the depth direction of the cross section.

  In the above embodiment, the MPR image is shown as the SA image. For example, an MPR image with a thickness is conceivable, and visualization with an average value method, a maximum value projection method, and a ray cast method is conceivable in the presence of a thickness. Further, visualization may be performed as disclosed in US Patent Application Publication No. 2006/0214930.

  In the above embodiment, the volume data as a captured CT image is transmitted from the CT apparatus 200 to the medical image processing apparatus 100. Instead, since the volume data is once accumulated, it may be transmitted to a server on the network and stored. In this case, when necessary, the port 110 of the medical image processing apparatus 100 may acquire volume data via a wired line or a wireless line, or may acquire via an arbitrary storage medium (not shown). .

  In the above embodiment, the volume data as the captured CT image is transmitted from the CT apparatus 200 to the medical image processing apparatus 100 via the port 110. This includes a case where the CT apparatus 200 and the medical image processing apparatus 100 are substantially combined into one product. This includes the case where the medical image processing apparatus 100 is handled as a console of the CT apparatus 200.

  In the above-described embodiment, it is exemplified that an image is captured by the CT apparatus 200 and volume data including information inside the living body is generated. However, an image may be captured by another apparatus to generate volume data. Other devices include an MRI (Magnetic Resonance Imaging) device, a PET (Positron Emission Tomography) device, an angiography device (Angiography device), or other modality devices. In addition, the PET apparatus may be used in combination with other modality apparatuses.

  The present disclosure is useful for a medical image processing apparatus, a medical image processing method, a medical image processing program, and the like that make it easy to visually recognize the correspondence between cross-sectional images at each point of a tissue path.

100 Medical Image Processing Device 110 Port 120 UI
130 Display 140 Processor 150 Memory 200 CT Device

Claims (10)

A medical image processing apparatus comprising a port and a processor,
The port acquires biological volume data,
The processor is
Extract the organization path included in the volume data,
Determine the main and non-main parts in the path;
Based on the volume data, a first cross-sectional image having a first surface including a first point included in the main portion of the path as a normal direction as a traveling direction of the path is generated,
Based on the volume data, a second cross-sectional image having a second surface including a second point included in a non-main part of the path as a normal direction as a traveling direction of the path is generated,
The roll of the first cross-sectional image is uniquely determined based on a roll of a predetermined cross-section,
The medical image processing apparatus, wherein a depth direction of the first cross-sectional image is uniquely determined based on a depth direction of the predetermined cross-section. The medical image processing apparatus according to claim 1,
The roll of the second cross-sectional image is uniquely determined based on the already determined cross-sectional image roll in the vicinity of the second point and the travel of the path,
The medical image processing apparatus, wherein the depth direction of the second cross-sectional image is uniquely determined based on the depth direction of the already determined cross-sectional image in the vicinity of the second point. The medical image processing apparatus according to claim 1 or 2,
The medical image processing apparatus, wherein the predetermined cross section is a cross section close to the first surface among a plurality of predetermined cross sections. The medical image processing apparatus according to any one of claims 1 to 3,
In the path, the path branches from a trunk part to a plurality of branch parts,
A medical image processing apparatus in which the main portion is included in the trunk portion. The medical image processing apparatus according to any one of claims 1 to 4,
The medical image processing apparatus in which the downward direction or the upward direction in the first cross-sectional image and the second cross-sectional image is a downward direction in the vertical direction or a direction close to upward in the vertical direction. The medical image processing apparatus according to any one of claims 1 to 4,
The depth direction of the predetermined cross section is a medical image processing apparatus in which the head is viewed from the foot of the living body or the foot is viewed from the head of the living body. The medical image processing apparatus according to any one of claims 1 to 6,
The medical image processing apparatus, wherein the main part of the path is an arbitrary range in the path. The medical image processing apparatus according to any one of claims 1 to 6,
The medical image processing apparatus, wherein the main part of the path is at least one arbitrary point in the path. A medical image processing method in a medical image processing apparatus,
Get the volume data of the living body,
Extract the organization path included in the volume data,
Determine the main and non-main parts in the path;
Based on the volume data, a first cross-sectional image having a first surface including a first point included in the main portion of the path as a normal direction as a traveling direction of the path is generated,
Based on the volume data, a second cross-sectional image having a second surface including a second point included in a non-main part of the path as a normal direction as a traveling direction of the path is generated,
The roll of the first cross-sectional image is uniquely determined based on a roll of a predetermined cross-section,
The medical image processing method, wherein the depth direction of the first cross-sectional image is uniquely determined based on the depth direction of the predetermined cross-section.   A medical image processing program for causing a computer to execute the medical image processing method according to claim 9.

Images