JP6842304B2 - Medical image processing equipment, medical image processing methods, and medical image processing programs - Google Patents

Description

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

As a non-invasive method for measuring blood flow, the blood flow velocity and time velocity curve can be obtained by the Phase Contrast method (hereinafter, also simply referred to as "PC method") using an MRI (Magnetic Resonance Imaging) device. It is known to be. In the PC method, the flow in the normal direction can be acquired with respect to the captured 2D cross section (hereinafter, "2D-PC method"). Further, in the PC method, a three-dimensional flow can be acquired by imaging cross sections in a plurality of directions (hereinafter, "3D-PC method"). It is also possible to measure the flow rate using the Pulse Doppler method in an ultrasonic diagnostic apparatus.

A Fontan procedure that connects the vena cava and the pulmonary artery may be performed on a patient who is congenitally single ventricle to create a Fontan circulation. In the Fontan circulation, venous blood returned from systemic organs is supplied to the lungs without passing through the ventricles (Fontan route). The Fontan procedure improves the prognosis and symptoms in the short postoperative period, but it has been found that complications due to its special circulatory state occur in various organs over a long period of time. Recently, congestive liver and hepatic fibrosis associated with increased central venous pressure, and the development of liver cirrhosis and hepatocellular carcinoma due to their progression have been attracting attention as important factors that influence the prognosis. However, there are no blood or biomarkers that make them predictable.

Riki Terada, Yasuo Takehara, "Hemodynamic Analysis and Blood Flow Quantification Method Using MR Images", INNERVISION, September 26, 2011, p23-p27

The 2D-PC method has a drawback that it is difficult to measure vortices (including turbulent flow), although laminar flow with uniform flow can be measured. Therefore, it is not possible to distinguish between vortices and retention. The result also depends on the Velocity Encoding setting. Moreover, only the flow component in the normal direction can be measured with respect to the cross-sectional image. Further, since the image obtained by the PC method is a functional image, it is necessary to separately capture a morphological image, and in some cases, alignment is required. For this reason, the PC method has been used to measure the flow rate of arteries with clear flow and high straightness. Further, in the improved 3D-PC method, the imaging time is long, and the data analysis software is expensive and requires specialized knowledge and operation. In addition, it takes a long time to take an image, and it is necessary to limit the movement of the patient such as respiratory synchronization.

Due to these characteristics, the PC method was not suitable for flow evaluation of vena cava, atrium, dilated ventricle, Fontan route, etc., which have a relatively slow flow and are prone to vortex generation.

In addition, the Pulse Doppler method has a limit in the evaluation of blood flow in the deep part of the patient such as the right heart system and the atrium due to the limited visual field. In addition, there was a possibility of technical differences depending on the operator.

The present disclosure has been made in view of the above circumstances, and is a medical image processing apparatus and a medical image processing capable of easily improving the estimation accuracy of the blood flow at the observation site of the subject in which the flow is relatively slow and vortices are likely to occur. Provide a program.

The medical image processing apparatus of the present disclosure includes a port and a processor. The port contains a blood flow having at least one of a vortex and a retention, and acquires a plurality of cross-sectional morphological images arranged in chronological order. The processor acquires motion information regarding the deformation of the cross-sectional morphological image based only on a plurality of cross-sectional morphological images, and indicates the state of blood flow based on at least one of the movement amount, the speed, and the strain amount in the motion information. Information is acquired, and information indicating the state of blood flow is superimposed on the cross-sectional morphological image to generate a display image. The information indicating the state of blood flow includes at least one piece of information on a vortex and a retention having in-plane components parallel to the cross section of the cross-sectional morphological image.

The medical image processing method of the present disclosure is a medical image processing method in a medical image processing apparatus, which includes a region of blood flow having at least one of a vortex and a retention, and acquires a plurality of cross-sectional morphological images arranged in a time series. Motion information regarding deformation of the cross-sectional morphological image is acquired based only on a plurality of cross-sectional morphological images, and information indicating the state of blood flow is acquired based on at least one of the movement amount, speed, and strain amount in the motion information. Then, information indicating the state of blood flow is superimposed on the cross-sectional morphological image to generate a display image. The information indicating the state of blood flow includes at least one piece of information on a vortex and a retention having in-plane components parallel to the cross section of the cross-sectional morphological image.

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

According to the present disclosure, it is possible to evaluate the flow of veins, atriums, aortic aneurysms, varicose veins, venous dilations, and shunts, which are relatively slow in flow and prone to vortex formation. In particular, we have found that the disappearance of the Fontan root vortex in Fontan circulating patients can predict future liver fibrosis and cirrhosis. Therefore, the present disclosure can be an epoch-making minimally invasive diagnostic imaging and biomarker. In addition, the flow including the in-plane component of the cross section can be evaluated with respect to the cross section image. In addition, the flow can be evaluated for the image acquired by imaging by the Gradient Echo method, and it is not necessary to separately capture the morphological image or align it with the morphological image. In addition, vortices, laminar flows, and retention contained in the bloodstream can be evaluated. Moreover, the imaging time is short.

Block diagram showing a configuration example of the medical image processing apparatus according to the first embodiment Flowchart showing an operation example by a medical image processing device A diagram schematically showing an image to which information indicating the state of blood flow based on the maximum main strain component is mapped. Schematic diagram schematically showing an MRI image on which the maximum main strain component is superimposed The figure which shows the image example of the MRI image of FIG. The figure which shows the image example of the MRI image including the liver of the patient (Control) who performed Fontan operation but does not have a liver disease. The figure which shows the image example of the MRI image including the liver of a healthy person (Normal) The figure which shows the image example of the MRI image containing the liver of a person with liver cancer The figure which shows the 1st image example of the MRI image containing the liver of a person whose liver became fibrotic The figure which shows the 2nd image example of the MRI image which includes the liver of the person which the liver became fibrotic The figure for demonstrating the blood flow measurement by the conventional 2D-PC method. The figure for demonstrating the blood flow estimation of the 1st Embodiment

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

(Background to obtain one form of this disclosure)
When the blood flow is measured according to the 2D-PC method, the blood flow measuring device can measure the laminar flow in which the flow is uniform, but it is difficult to measure the vortex. Difficulty in measuring vortices makes it difficult to distinguish between a state without blood flow and a state with blood flow but vortex, with at least part of the blood flow offset and appearing to be non-blood flow. Because it is. That is, it is difficult for the blood flow measuring device to distinguish whether the blood flow is a vortex or a retention. In addition, the blood flow measurement result of the blood flow measuring device is also influenced by the Magnetic Encoding parameter for changing the gradient magnetic field strength of the MRI device according to the blood flow velocity. Further, it is difficult for the blood flow measuring device to measure a flow component other than the flow component in the normal direction with respect to the cross-sectional image captured by the MRI device. Further, since the image obtained by the PC method is a functional image, a morphological image is required separately. In this case, it may be necessary to align the image of the observation site between the image capture of the functional image and the image capture of the morphological image.

When the blood flow measuring device measures the blood flow according to the 3D-PC method, it is necessary to image the cross sections in a plurality of directions by electrocardiographic synchronization, so that the imaging time for capturing the image by the MRI device becomes long. Therefore, when a person with uneven breathing or a person with an arrhythmia takes an image with an MRI apparatus, the imaging accuracy with the MRI apparatus may deteriorate. In addition, the image quality is affected by the movement of the patient during imaging. In addition, a blood flow measuring device that measures blood flow by the 3D-PC method and an MRI device that captures an image used in the 3D-PC method are expensive.

Therefore, the PC method is often used for measuring the flow rate of an artery having a clear blood flow (laminar flow) and high straightness. On the other hand, in the PC method, blood flow is relatively slow and blood at observation sites where vortices are likely to occur (for example, veins, atrium, apex, aneurysm, varicose veins, false lumen of vascular dissociation, venous dilation). Not suitable for flow evaluation.

By the way, the Fontan procedure may be performed on a patient who is congenitally single ventricle to create a Fontan circulation. In the Fontan circulation, venous blood returned from systemic organs can be supplied to the pulmonary arteries without going through the heart. In this case, the Fontan route, which is an artificial route that supplies the returned venous blood directly to the lungs, is created. Patients with Fontan circulation need to be followed up after surgery. This is because venous blood stays in the Fontan route and the venous pressure rises, which puts a strain on the visceral tissues. In particular, it has been reported that when the venous pressure in or around the position where the right ventricle / right atrium should exist rises and the blood flow becomes stagnant, the liver is overloaded, and the liver becomes fibrotic and eventually becomes cancerous. In addition, venous blood retention also causes pulmonary infarction.

The Fontan route tends to have relatively slow or stagnant blood flow because venous blood passes near the pulmonary artery. Therefore, the PC method is not suitable for evaluating the blood flow of the Fontan route and the pulmonary artery in the Fontan circulation patient.

Hereinafter, a medical image processing apparatus, a medical image processing method, and a medical image processing program that can improve the estimation accuracy of blood flow at an observation site of a subject whose flow is relatively slow and vortices are likely to occur will be described.

(First Embodiment)
FIG. 1 is a block diagram showing a configuration example of the medical image processing apparatus 100 according to the first embodiment. The medical image processing device 100 includes a port 110, a user interface (UI: User Interface) 120, a display 130, a processor 140, and a memory 150.

An MRI apparatus 200 is connected to the medical image processing apparatus 100. The medical image processing device 100 acquires an MRI image as a cross-sectional image from the MRI device 200, and processes the acquired MRI image. The medical image processing apparatus 100 may be composed of a PC (Personal Computer) and software installed in the PC. Further, the medical image processing device 100 may be built in the MRI device 200.

The MRI apparatus 200 captures an image (MRI image) by imaging information inside the living body by utilizing a nuclear magnetic resonance (NMR) phenomenon. At the time of imaging of the MRI apparatus 200, the contrast medium may or may not be added to the living body. Examples of the living body include the human body. The living body is an example of a subject. The MRI apparatus 200 images an arbitrary cross section in a living body and obtains an MRI image as a cross section image. Multiple MRI images may be captured in chronological order in the same cross section. This cross section includes at least a portion of the body including blood flow (eg, vein, atrium, apex, aneurysm, varicose vein, false lumen of vascular dissociation, venous dilation, or fontan root). By capturing the MRI image, the pixel value (MRI value) of each pixel (pixel or voxel) in the MRI image can be obtained. The MRI apparatus 200 transmits an MRI image to the medical image processing apparatus 100 via a wired line or a wireless line.

Specifically, the MRI apparatus 200 includes a gantry (not shown) and a console (not shown). The gantry includes a mounting table on which the subject is placed, a static magnetic field magnet, a gradient magnetic field coil, an RF (Radio Frequency) transmitting coil, and an RF receiving coil. The static magnetic field magnet forms a uniform magnetic field region. The mount is located in the center of the uniform magnetic field region. The gradient magnetic field coils are three pairs of coils that generate gradient magnetic fields in three directions (x direction, y direction, and z direction) orthogonal to each other in a three-dimensional space. The RF transmission coil is arranged between the imaging site of the subject and the static magnetic field magnet, and pulsates the subject with a high-frequency magnetic field. The RF receiving coil receives a weak MR signal immediately after the pulse irradiation of the high frequency magnetic field. The gradient magnetic field may be applied in a pulse shape during RF transmission or RF reception. The RF transmitting coil and the RF receiving coil may be used in combination.

The gantry obtains an MR signal by taking an image at a predetermined timing instructed by the console. The console is connected to the medical image processing device 100. The console acquires an MR signal from the gantry and generates an MRI image as a cross-sectional image based on the MR signal. The console transmits the generated MRI image to the medical image processing apparatus 100.

The MRI apparatus 200 may acquire a plurality of MRI images of a cross section to be imaged and generate a moving image by continuously taking images. This moving image may be a moving image of a plurality of two-dimensional images. In addition, the MRI apparatus 200 can capture an image by various imaging methods. The MRI apparatus 200 images according to, for example, the Gradient Echo method. The Gradient Echo method enables high-speed imaging as compared with the general Spin Echo method, and is suitable for capturing images captured in a plurality of phases. The captured images of the plurality of phases may function as moving images. The Gradient Echo method includes a rewinder type, a spoiler type, a coherent type, a non-coherent type, and other improved methods.

The MRI apparatus 200 captures a plurality of MRI images at different timings (plural phases) in time series. The imaging interval of the MRI image is arbitrary, but may be an interval of 0.1 seconds or another time interval. The imaging intervals may be equal or non-equal intervals. The MRI apparatus 200 may collectively transmit a predetermined number (for example, 10 or 20) of MRI images to the medical image processing apparatus 100, or may transmit one MRI image for each imaging.

The port 110 in the medical image processing device 100 includes a communication port and an external device connection port, and acquires an MRI image. The acquired MRI image may be immediately sent to the processor 140 for various processing, or may be stored in the memory 150 and then sent to the processor 140 for various processing when necessary.

The MRI image acquired by the port may be an image captured according to the Gradient Echo method, or may be a morphological image captured according to another imaging method. For example, it may be imaged by an improved sufficiently fast Spin Echo method or the Fast Spin Echo method. It can be said that the image captured according to the Phase Contrast method is a functional image and not a morphological image. A functional image is an image of the physiological activity of a tissue taken by an apparatus. Examples of the functional image include an image obtained by a PET (Positron Emission Tomography) device and a diffusion-weighted image obtained by MRI. On the other hand, the morphological image is an image in which the morphology of the tissue is taken by the device. Examples of the morphological image include an image obtained by a CT device, a T1-weighted image, a T2-weighted image, and a T2 * (T2 star) -enhanced image obtained by an MRI image.

The UI 120 may include a touch panel, a pointing device, a keyboard, or a microphone. The UI 120 accepts an arbitrary input operation from the user of the medical image processing device 100. The user may include a doctor, a radiologist, or other Paramedic Staff.

The UI 120 accepts operations such as designation of a region of interest (ROI: Region of Interest) in an MRI image and setting of luminance conditions. Areas of interest may include areas of tissue (eg, blood vessels, bronchi, organs, bone). The tissue may broadly include living tissue such as lesioned tissue, normal tissue, organ, organ, and the like.

The display 130 may include an LCD (Liquid Crystal Display) and displays various information. The various information includes an MRI image and information added to or superimposed on the MRI image (for example, information indicating the state of blood flow).

The memory 150 includes various ROM (Read Only Memory) and RAM (Random Access Memory) primary storage devices. The memory 150 may include a secondary storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The memory 150 may include a USB memory or a tertiary storage device of an SD card. The memory 150 stores various information and programs. The various information may include an MRI image acquired by the port 110, an image generated by the processor 140, and setting information set by the processor 140.

The processor 140 may include 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. In addition, the processor 140 controls each part of the medical image processing device 100.

The processor 140 may extract a part of the area of the MRI image from the acquired MRI image. In this case, the UI 120 receives the instruction from the user, and the information of the instruction is sent to the processor 140. Processor 140 may segment the region of interest from the MRI image based on the indicated information. Further, the region of interest may be set manually according to a detailed instruction from the user. If the observation site is predetermined, the processor 140 may extract the region of interest including the observation site from the MRI image without user instruction. The observation site or region of interest may include a region through which blood flowing as blood flow passes (for example, a blood vessel, a heart, or other organ containing blood flow) (hereinafter, also referred to as a blood flow region). The blood flow region is extracted by existing segmentation means such as RegionGrow, LevelSet method, GraphCut, DeepLearning and the like. In the following, it is mainly illustrated that the entire MRI image is image-processed without extracting a part of the MRI image, but even if a part of the MRI image (for example, a vascular region) is extracted and image-processed. Good.

The processor 140 acquires a plurality of phases of MRI images in any cross section including at least a part of the observation site from the port 110 and the memory 150.

The processor 140 performs motion analysis on the deformation of the MRI image between the plurality of phases based on the MRI image of the plurality of phases, and acquires the motion information in the MRI image. A specific method of motion analysis is described in, for example, Reference Patent Document 1.
(Reference Patent Document 1: US Pat. No. 8311300)

The processor 140 may acquire information related to the amount of movement of an arbitrary point in the MRI image and information related to the speed as motion information. When the method of Reference Patent Document 1 is applied, the processor 140 divides the MRI image into two-dimensional grids node (k, l) and divides the two-dimensional coordinates in the phase t grid node (k, l, t) of the two-dimensional grid. When (x, y) is set, information on the amount of movement related to the grid points of the node (k, l) based on the differences of a plurality of nodes (k, l, t) obtained by changing the value of the phase t. May be calculated. Further, the processor 140 may calculate the speed information by time-differentiating the movement amount information. Information on the amount of movement and speed may be indicated by a vector.

When the processor 140 interpolates the motion information of the two-dimensional lattice with respect to each point of the entire MRI image, the motion information of each point of the MRI image is obtained. When the motion information of the predetermined point is applied to each point in the region including the observation site, the motion information of each point in the region including the observation site can be obtained. The processor 140 is not limited to the method of Reference Patent Document 1, and motion analysis may be performed using other known registration methods.

In the motion analysis, the medical image processing device 100 can grasp, for example, to which position an arbitrary position in the living body has moved. In this case, the medical image processing apparatus 100 can detect and track a region of a mass having close pixel values in the MRI image of the previous and next phases. With respect to blood flow, it is possible to detect one or more regions of lumps having similar pixel values and track the movement of these regions.

The processor 140 acquires the amount of distortion included in the motion of the MRI image, for example, the maximum principal distortion component, based on the result of the motion analysis, that is, the motion information. A specific method for obtaining the maximum principal strain component is described in, for example, Reference Non-Patent Document 2.
(Reference Non-Patent Document 1: "The visualization of 3D stress and strain tensor fields", Burkhard Wuensche, 1999)

When the method of Reference Non-Patent Document 2 is applied, the processor 140 is two-dimensional based on the positions of the two points P and Q before the movement and the positions of the two points P'and Q'after the movement obtained by the motion analysis. The strain tensor of may be calculated. The strain tensor may be acquired by the spatial differentiation of the movement amount and velocity vectors acquired by the motion information. The processor 140 may calculate the two eigenvectors v1 and v2 and the two eigenvalues λ1 and λ2 related to the calculated two-dimensional strain tensor. The eigenvalues λ1 and λ2 may be eigenvalues related to the matrix transformation. The processor 140 may acquire the maximum eigenvalue among the eigenvalues λ1 and λ2 as the value of the maximum principal strain component.

When the processor 140 interpolates the acquisition of the maximum principal distortion component based on the two points P and Q for each point in the entire MRI image, information on the maximum principal distortion component included in the movement at each point of the MRI image can be obtained. When the acquisition of the maximum principal strain component based on these two points P and Q is applied to each point in the region including the observation site, information on the maximum principal strain component at each point in the region including the observation site can be obtained.

The processor 140 superimposes the acquired information on the maximum principal strain component on the MRI image to generate a display image. The processor 140 causes the display 130 to display the displayed image.

Next, an operation example by the medical image processing apparatus 100 will be described.
FIG. 2 is a flowchart showing an operation example by the medical image processing device 100.

First, the port 110 acquires a multi-phase MRI image captured by the Gradient Echo method (S11).

The processor 140 performs motion analysis based on the MRI images of the plurality of phases (S12). The processor 140 acquires the maximum principal strain component based on the result of the motion analysis (S13). The processor 140 superimposes the information based on the acquired maximum principal distortion component value on the MRI image to generate a display image, and displays the display image on the display 130 (S14). The value of the maximum principal strain component is an example of information indicating the state of blood flow.

The following can be considered as the reason why the state of blood flow appears in the maximum main strain component in S13.

Flow-voice artifacts caused by blood flow appear in the MRI image, and changes in flow-voice over time on the MRI image are captured as motion information. The flow-voice captured as motion information can also be said to be a traceable flow-voice. Therefore, the medical image processing apparatus 100 can estimate that the motion information includes at least qualitative blood flow or blood flow properties by tracking the flow-voice.

Flow-voice shows a phenomenon in which a tissue with a flow becomes no signal on an MRI image, and occurs particularly at a position where the flow velocity changes. At the position where blood flows from a small blood vessel into a lump or a hypertrophied blood vessel, the flow velocity becomes sharply slow, and at the position where the blood vessel is narrowed, the flow velocity becomes high. In addition, the closer to the blood vessel wall, the slower the flow velocity, and the faster the blood flow at the center of the blood vessel. It can be estimated that this change in flow velocity produces a traceable flow-voice.

Further, the maximum principal strain component derived from the motion information is considered to indicate the direction in which the amount of deformation at the observation site in the MRI image or the MRI image is maximum. Therefore, in the medical image processing apparatus 100, it can be estimated that the maximum main strain component well reflects the properties of the actual blood flow. In this way, it is presumed that the maximum principal strain component expresses the state of blood flow at the observation site.

By displaying such information on the maximum main strain component together with the MRI image as a morphological image, the user of the medical image processing apparatus 100 can recognize the amount of distortion of blood vessels and the like on the MRI image, and the state of blood flow. Can be grasped.

FIG. 3 is a diagram schematically showing an image to which information indicating the state of blood flow based on the maximum main strain component is mapped. FIG. 4 is a schematic diagram schematically showing an MRI image on which the maximum main strain component is superimposed. FIG. 5 is a diagram showing an image example of the MRI image of FIG. 3 to 5 show figures of subjects who have undergone the Fontan procedure and have the Fontan route 11 expanded. The Fontan route 11 may include at least a portion of the pulmonary artery, vein, or right atrial ventricle.

The subject from which the MRI image according to FIGS. 3 to 5 was imaged was subjected to the Fontan operation, and the Fontan route 11 connecting the vein (for example, the hepatic vein 13) and the pulmonary artery was formed without going through the heart. It is also presumed that the Fontan route 11 is expanding.

FIG. 3 is a diagram showing an image G11 to which information indicating a state of blood flow based on the maximum principal strain component in and around the Fontan route 11 is mapped. The nature of blood flow may include vortices, retention, and laminar flow. This property of blood flow is reflected in the information indicating the state of blood flow.

In the vortex portion 11a in the image G11, blood flow is generated, but since blood flow is generated in a plurality of (for example, many) directions, at least a part of each blood flow is canceled out and the blood flow is captured. It can be difficult. However, when blood flow is generated in many directions and the wall of the blood vessel through which the blood flows is pressed, the time change of the pressure on the blood vessel becomes large, and the blood vessel is greatly deformed. Therefore, the value of the maximum main strain component becomes large (high strain). Blood flow is affected by the characteristics of blood (for example, viscosity), the shape of blood vessels, the shapes of surrounding organs surrounding blood vessels, and the like, and the direction of travel of blood flow becomes indefinite and becomes a vortex. This vortex can be a turbulent flow. Thrombi are likely to occur in the vortex portion 11a.

In the retention portion 11b in the image G11, almost no blood flow is generated, the pressure on the blood vessel hardly changes with time, and the blood vessel is hardly deformed. Therefore, the value of the maximum main strain component becomes small (low strain). Thrombi are likely to occur in the retention portion 11b.

In the laminar flow portion 11c in the image G11, blood preferably flows in the blood vessel along the blood vessel, the time change of the pressure on the blood vessel is moderate, and the blood vessel is slightly deformed. Therefore, the value of the maximum main strain component is medium (medium strain).

In FIG. 4, the MRI image G12 includes the Fontan route 11, the aorta 12, the hepatic vein 13, and the liver 14. The subject in which the MRI image G12 of FIG. 4 was imaged was subjected to the Fontan operation, and a Fontan route connecting a vein (for example, hepatic vein 13) and a pulmonary artery was formed without going through the heart. The blood flowing out of the liver 14 flows into the pulmonary artery via the hepatic vein 13. In a subject who has undergone the Fontan procedure, blood is pumped into the pulmonary artery from a vein with low blood pressure, so that blood tends to accumulate in the Fontan route 11. Therefore, the Fontan route 11 tends to expand (see FIG. 4). The expanded Fontan route 11 may include a vortex portion 11a, a retention portion 11b, and a laminar flow portion 11c.

In FIG. 5, the vortex portion 11a, the retention portion 11b, and the laminar flow portion 11c of FIG. 4 are colored by the processor 140. The scale bar 21 in FIG. 5 shows information on the magnitude of the value of the maximum principal strain component. The scale bar 21 may be color-coded for each value. In this case, the processor 140 determines a color indicating the magnitude of the maximum principal strain component for each pixel of the MRI image G13 according to the magnitude of the maximum principal strain component in each pixel. This color is an example of information indicating the state of blood flow. For example, in the MRI image G13 on which the maximum principal strain component is superimposed, the value of the maximum principal strain component may be expressed by changing the color between 0 and 2.8.

When the value of the maximum principal strain component is large (for example, the vortex portion 11a), the color may be close to orange to red. When the value of the maximum main strain component is small (for example, the retention portion 11b), the color may be close to blue to purple. When the value of the maximum main strain component is medium (for example, the laminar flow portion 11c), the color may be close to green.

The processor 140 may determine the type of blood flow, that is, the state of blood flow, based on the value of the maximum principal strain component. In this case, if the value of the maximum principal strain component is 2.0 or more, it may be determined that the blood flow is a vortex. The processor 140 may determine that the blood flow is stagnant when the value of the maximum principal strain component is in the vicinity of 0.0. The processor 140 may determine that the blood flow is laminar when the value of the maximum principal strain component is around 1.0 (near 1.0). The value of the maximum principal strain component indicates a ratio of "1.0" when the same portion does not move (deform) between a plurality of phases.

Next, the MRI image on which the maximum main strain component is superimposed is shown for each case of the subject.

FIG. 6 is a diagram showing an example of an image of an MRI image G14 including the liver 14 of a patient (also referred to as Control) who does not have a disease in the liver 14 after the Fontan operation. In FIG. 6, a portion close to purple, that is, a stagnant portion 11b can be confirmed to some extent, but the vortex portion 11a does not exist. The size of the stagnant portion 11b in FIG. 6 is within an allowable range, and can be confirmed in a small range even in a healthy person in FIG. 7 described later. For Fontan circulating patients who have undergone the Fontan procedure, it is preferable that the condition shown in FIG. 6 is maintained.

FIG. 7 is a diagram showing an image example of an MRI image G15 including a liver 14 of a healthy person (also referred to as Normal). FIG. 8 is a diagram showing an image example of an MRI image G16 including the liver 14 of a patient with liver cancer. FIG. 9 is a diagram showing an example of a first image of an MRI image G17 including the liver 14 of a patient whose liver 14 has become fibrotic. FIG. 10 is a diagram showing a second image example of an MRI image G18 including the liver 14 of a patient whose liver 14 has become fibrotic. In the subject shown in FIG. 10, it can be easily understood that the Fontan route 11 is dilated and the blood flow is stagnant.

In this way, when the Control (see FIG. 6), Normal (see FIG. 7), and symptomatic patients (see FIGS. 5, 8 to 10) are compared, the vortex portion and retention on the image in the symptomatic patient. Since all of the parts are remarkably observed, it can be seen that there is a problem with blood flow dynamics.

Next, the blood flow associated with the Fontan procedure and the Fontan procedure will be supplemented.

The Fontan procedure, for example, is performed on a patient with one ventricle to form the Fontan route. In the Fontan route, blood that circulates through the arteries is sent to the pulmonary arteries without going through the heart. In this case, the vein may be directly connected to the pulmonary vein without the heart, or the vein and the pulmonary artery may be connected via the artificial blood vessel without the heart. Since the pressure to transport blood tends to be low in the veins, blood is not released from the veins and the pressure inside the veins tends to accumulate. Similarly, in the pulmonary arteries that are not connected to the heart, the pressure to transport blood tends to be low, so that blood is not released from the pulmonary arteries and the pressure inside the pulmonary arteries tends to accumulate. The same applies when the vein connected to the pulmonary artery is an artificial blood vessel. In this case, arterial blood may not be sufficiently drained from the liver or pancreas near the veins or pulmonary arteries, resulting in damage. It is preferable that this accumulation of pressure is suppressed so that blood flows smoothly (for example, in laminar flow) in veins and pulmonary arteries.

Further, the observation of the patient's blood flow after the Fontan operation is an example of blood flow estimation, and other blood flow may be observed. For example, the medical image processing apparatus 100 may estimate the blood flow by using a general vein as an observation site. In addition, blood flow estimation by the medical image processing device 100 is also effective in patients who have aneurysms or stenosis in blood vessels.

Sudden changes in blood vessel diameter can cause blood to stagnate and stagnate or generate vortices in the dilated blood vessels. In the liver around the veins and artificial blood vessels connected to the pulmonary artery by the Fontan procedure, blood stagnation and vortices in the Fontan route correlate with the increase in blood pressure in the large vein, and the liver is directly connected to the inferior vena cava. Therefore, it is strongly influenced by it. Therefore, stagnation, vortex, etc. are likely to induce liver cirrhosis and liver fibrosis. Therefore, it is very meaningful to discover stagnation of blood flow, which is one of the causes of liver cirrhosis and fibrosis of the liver.

In addition, thrombi are likely to occur in places where vortices are present. If a blood clot moves, for example, to the head, it can induce a stroke. In the Fontan route, blood clots cause pulmonary infarction. Therefore, it is very meaningful that the place where the vortex is generated can be specified by the medical image processing device 100.

Also, the right atrium is the first place to enter when blood returns around the veins. Therefore, in the right atrium as well as veins and artificial blood vessels, the blood flow tends to be stagnant. According to the medical image processing apparatus 100, the state of blood flow in the right atrium can also be estimated with high accuracy.

FIG. 11 is a diagram for explaining blood flow measurement by the conventional 2D-PC method. A blood flow measuring device that measures blood flow by a conventional 2D-PC method acquires an MRI image with a cross section in the minor axis direction of a blood vessel or the like. Then, the blood flow measuring device expresses the flow in the normal direction with respect to the MRI image as the cross-sectional image as the flow in the long axis direction of the blood vessel. Further, since the image obtained by the PC method is a functional image, the blood flow measuring device needs to separately capture a morphological image and perform alignment.

On the other hand, FIG. 12 is a diagram for explaining the blood flow estimation of the present embodiment. The medical image processing apparatus 100 displays an MRI image in which information indicating the state of blood flow based on the maximum main strain component is superimposed. The medical image processing apparatus 100 acquires an MRI image having a cross section in the long axis direction of a blood vessel or the like to be observed, and expresses the state of blood flow as an in-plane component of the MRI image. Therefore, the user can understand the state of blood flow along the blood vessel to be confirmed by confirming the MRI image derived by the medical image processing apparatus 100, and thus when confirming the flow in the normal direction of the cross section. By comparison, blood flow can be recognized intuitively.

Further, the medical image processing apparatus 100 seems to have no blood flow and a vortex having blood flow, and at least a part of the blood flow is canceled out and there is no blood flow, based on the information indicating the state of blood flow. The state can be identified. Therefore, the medical image processing apparatus 100 has a flow (blood) in veins, atriums, aneurysms, varicose veins, false cavities of vascular dissociation, vein dilation, shunts, cardio-ear, etc. Flow) can be evaluated. In particular, the medical image processing apparatus 100 can suitably evaluate the flow of the pulmonary artery in a Fontan circulation patient. Further, the medical image processing apparatus 100 can evaluate the flow including the in-plane component of the cross section (the in-plane component parallel to the cross section) with respect to the cross-sectional image. Further, the medical image processing apparatus 100 can evaluate the flow of the image acquired by the imaging by the Gradient Echo method, and it is not necessary to separately capture the morphological image or align it with the morphological image. Further, the medical image processing apparatus 100 can evaluate the vortex, the laminar flow, and the retention contained in the blood flow. In addition, the imaging time of the MRI image used for blood flow estimation can be shortened.

As described above, in the medical image processing apparatus 100 of the present embodiment, the port 110 includes a blood flow having at least one of a vortex and a retention, and acquires a plurality of cross-sectional morphological images arranged in a time series. The processor 140 acquires motion information regarding deformation of the cross-sectional morphological image based on the plurality of cross-sectional morphological images. The processor 140 acquires information indicating the state of blood flow based on at least one of the movement amount, the velocity, and the strain amount in the motion information. The processor 140 generates a display image by superimposing information indicating the state of blood flow on the cross-sectional morphological image. The information indicating the state of blood flow includes at least one piece of information about a vortex and a retention having an in-plane component parallel to the cross section of the cross-sectional morphological image. The plurality of images arranged in chronological order may be images of a plurality of phases. The cross-sectional morphological image may be a cross-sectional image showing a cross section of the morphological image.

As a result, the medical image processing apparatus 100 derives information indicating the blood flow state based on the motion information related to the deformation in the cross-sectional image of the plurality of phases, and qualitatively derives the blood flow state (for example, vortex, laminar flow, stagnation). Can be estimated. For example, when the amount of strain is large, the medical image processing apparatus 100 includes a portion where the blood flow becomes a vortex (including turbulent flow), and it can be inferred that the blood and surrounding organs are adversely affected. Further, when the amount of strain is medium, it can be inferred that the medical image processing apparatus 100 has an appropriate laminar flow. When the amount of distortion is small, it can be inferred that the medical image processing apparatus 100 has almost no blood flow and is stagnant. Then, the medical image processing device 100 can visualize the state of blood flow from the displayed image. Further, in the medical image processing apparatus 100, the cross section of the cross-sectional image is a long-axis plane which is a plane in the long-axis direction with respect to the blood vessel, and the components in this plane can be expressed by one image, so that the state of the entire blood vessel can be grasped. It can be done easily. As described above, the medical image processing apparatus 100 can easily improve the estimation accuracy of the blood flow at the observation site of the subject in which the flow is relatively slow and vortices are likely to be generated.

Further, the processor 140 may extract a region through which blood flowing as blood flow included in the cross-sectional morphological image passes, superimpose information indicating the state of blood flow in the region through which blood passes, and generate a display image. ..

As a result, the user can observe the fluid inside the subject only in the blood vessels, which makes it easier to observe the blood flow. Further, the medical image processing apparatus 100 does not need to perform image processing such as motion analysis and derivation of the maximum main distortion component on the entire acquired MRI image, and can reduce the processing load related to the image processing. Further, the medical image processing apparatus 100 can limit the range in which motion analysis is performed by limiting the area, and can be expected to improve the processing accuracy of motion analysis.

Further, the cross-sectional morphological image may be an MRI image captured by the MRI apparatus 200.

As a result, the medical image processing apparatus 100 can non-invasively observe the state of blood flow.

Further, the MRI image may be an image captured according to the Gradient Echo method.

As a result, the medical image processing apparatus 100 can take an MRI image at high speed, shorten the phase interval, and acquire a moving image of an arbitrary cross section. Therefore, the medical image processing apparatus 100 can easily recognize the deformation in the cross-sectional image in a short time and improve the accuracy of the motion information, so that the estimation accuracy of the blood flow state can be improved.

Further, the medical image processing device 100 does not need to take a special image for observing blood flow. For example, when an MRI image around the heart is captured, a positioning image is captured before the image used for individual diagnosis is captured. At this time, in order to obtain anatomical information of the entire heart, a two-dimensional cine image is often taken in each cross section by the Gradient Echo method. The medical image processing apparatus 100 can estimate the state of blood flow based on at least a part of this positioning. Therefore, the medical image processing device 100 can also estimate the state of blood flow based on, for example, the positioning images accumulated in the past.

In addition, the diameter of the flow path through which blood moving as blood flow passes may change depending on the stenosis or aneurysm.

That is, the stenosis or aneurysm disturbs the blood flow that was flowing as a laminar flow before and after the stenosis or aneurysm, causing vortices and retention, and the vortices and retention are reflected in the flow-voice and detected as motion information. Therefore, when the medical image processing device 100 presents information indicating blood flow information to the user by display, the user can observe the blood flow contained in the stenosis or aneurysm, or the nature of the blood flow, and make a highly accurate diagnosis. You can get the material. In particular, for aneurysms and atrial appendages, CT images do not distinguish between stagnation with a small influx of new blood and stagnation simply in the absence of blood. On the other hand, the medical image processing apparatus 100 can observe the blood flow in the aneurysm and the atrial appendage, or the nature of the blood flow.

Although various embodiments have been described above with reference to the drawings, it goes without saying that the present disclosure is not limited to such examples. It is clear that a person skilled in the art can come up with various modifications or modifications within the scope of the claims, which naturally belong to the technical scope of the present disclosure. Understood.

In the first embodiment, an image is captured by the MRI apparatus 200 to generate an MRI image including information inside the living body, but an image may be captured by another apparatus to generate a cross-sectional image. .. Other devices include angiography devices (Angiografy devices) or other modality devices. Moreover, the PET apparatus may be used in combination with other modality apparatus. For example, in an angiography apparatus, the contrast medium is imaged as a smoke-like texture, and the flow can be seen by the change in the texture. Therefore, since the angiography apparatus can acquire information indicating the state of blood flow by motion analysis, the above embodiment can be applied. The angiography device is suitable for observing coronary aneurysms and pulmonary aneurysms of Kawasaki disease. For example, the medical image processing apparatus 100 can acquire an image captured by the angiography apparatus and evaluate the vortex of the coronary artery flow, which is one of the factors of calcification.

In the first embodiment, it is mainly exemplified that it is used for blood flow estimation of the Fontan route, but it may be used for other blood flow estimation. For example, the method of the first embodiment may be used for the analysis of blood flow in the atrium, atrial appendage, aneurysm, valve regurgitation, aortic aneurysm, deep vein thrombus, pulmonary thromboembolism, apex thrombus and the like. That is, the medical image processing device 100 can be used for evaluating the flow at a place where the flow path diameter of the blood flow changes significantly, including stenosis and aneurysm.

In the first embodiment, the medical image processing apparatus 100 mainly exemplifies the estimation of the state of blood flow (for example, vortex, retention, laminar flow) according to the value of the maximum main strain component, but the maximum main strain The state of blood flow may be estimated using parameters other than strain. The state of blood flow may be estimated by a combination of a plurality of parameters. Parameters may include "Displacement" (movement, displacement), "Velocity" (velocity). The "displacement" may be included in the motion information or acquired based on the motion information. The "velocity" may be included in the motion information or acquired based on the motion information. “Displacement” may correspond to the amount of movement obtained in the motion analysis of Reference Patent Document 1. "Velocity" may correspond to the speed obtained in the motion analysis of Reference Patent Document 1. Therefore, in this case, the derivation of the maximum principal strain component may be omitted.

In the first embodiment, the medical image processing apparatus 100 exemplifies that the blood flow stagnation or vortex is generated due to the maximum principal strain component, but the center is focused on a specific center point. The occurrence of stagnation or vortex may be estimated from the amount of strain in the circumferential direction or the radial direction centered on the point. The medical image processing device 100 may have a user interface (for example, UI 120) in which the center point is designated by the user. Further, the medical image processing apparatus 100 focuses on the fact that the movement amount and speed are small at the center point of the vortex and the movement amount and speed are high around the center point of the vortex, and the center point is set based on the motion information. It may be specified. Further, the medical image processing device 100 pays attention to the fact that the moving directions of the blood flow of the vortex are continuous in a substantially circular shape, and may determine that the vortex is a vortex centered on the specified center point based on the movement information. Good. Further, the medical image processing apparatus 100 pays attention to the fact that the blood flow has a plurality of moving directions in the turbulent flow having no center, and the directions are indefinite or irregular, and the turbulent flow is based on the motion information. And a clear vortex at the center may be distinguished.

In the first embodiment, the medical image processing apparatus 100 exemplifies that an vortex of blood flow is generated, but the vortex is focused on a specific center point and centered on the center point. The swirling vortices may be individually identified. For example, the medical image processing apparatus 100 focuses on the fact that the movement amount and speed are small at the center point of such a vortex and the movement amount and speed are high around the center point of the vortex, and the center is based on the motion information. It may be determined that the vortex is centered on a point. Further, the medical image processing apparatus 100 pays attention to the fact that in the case of such a vortex, the moving directions are continuous in a substantially circular shape, and even if it determines that the vortex is centered on the center point based on the motion information. Good. Further, the medical image processing apparatus 100 pays attention to the fact that the blood flow has a plurality of moving directions in the turbulent flow having no center, and the directions are indefinite or irregular, and the turbulent flow is based on the motion information. And a clear vortex at the center may be distinguished. Further, the medical image processing device 100 may have a user interface (for example, UI 120) in which the center point is designated by the user.

In the first embodiment, it is illustrated that the captured MRI image is transmitted from the MRI apparatus 200 to the medical image processing apparatus 100. Instead, the MRI image may be transmitted to a server or the like on the network and stored in the server or the like so that the MRI image is temporarily accumulated. In this case, the port 110 of the medical image processing apparatus 100 may acquire the MRI image from the server or the like via a wired line or a wireless line when necessary, or obtains the MRI image via an arbitrary storage medium (not shown). You may.

In the first embodiment, it is illustrated that the captured MRI image is transmitted from the MRI apparatus 200 to the medical image processing apparatus 100 via the port 110. This includes the case where the MRI apparatus 200 and the medical image processing apparatus 100 are substantially combined into one product. It also includes the case where the medical image processing device 100 is treated as the console of the MRI device 200.

Further, in the above embodiment, the human body is exemplified as a living body, but an animal body may also be used.

Further, the present disclosure can also be expressed as a medical image processing method that defines the operation of the medical image processing device. Further, in the present disclosure, a program that realizes the functions of the medical image processing device of the above embodiment is supplied to the medical image processing device via a network or various storage media, and is read and executed by a computer in the medical image processing device. The program to be used is also the scope of application.

The present disclosure applies to medical image processing devices, medical image processing methods, medical image processing programs, etc., which can easily improve the estimation accuracy of blood flow at the observation site of a subject whose flow is relatively slow and vortices and retention are likely to occur. It is useful.

100 Medical image processor 110 Port 120 User interface (UI)
130 Display 140 Processor 150 Memory 200 MRI device 11 Fontan route 12 Aorta 13 Hepatic vein 14 Liver

Claims (9)

A medical image processor equipped with a port and a processor.
The port contains a blood flow having at least one of a vortex and a retention, and obtains a plurality of cross-sectional morphological images arranged in chronological order.
The processor
Based only on the plurality of cross-sectional morphological images, motion information regarding the deformation of the cross-sectional morphological image is acquired.
Based on at least one of the movement amount, the velocity, and the strain amount in the movement information, information indicating the state of blood flow is acquired, and information indicating the state of blood flow is acquired.
A display image is generated by superimposing information indicating the state of blood flow on the cross-sectional morphological image.
The information indicating the state of blood flow includes at least one information of a vortex and a retention having an in-plane component parallel to the cross section of the cross-sectional morphological image.
Medical image processing equipment. The medical image processing apparatus according to claim 1.
The processor acquires motion information regarding deformation of the cross-sectional morphological image in blood flow based only on the plurality of cross-sectional morphological images.
Medical image processing equipment. The medical image processing apparatus according to claim 1 or 2.
The blood flow includes vortices and retention
The information indicating the state of blood flow can distinguishably include information on vortices and retention having in-plane components parallel to the cross section of the cross-sectional morphological image.
Medical image processing equipment. The medical image processing apparatus according to any one of claims 1 to 3.
The processor
The region through which the blood flowing as the blood flow, which is included in the cross-sectional morphological image, passes is extracted.
Information indicating the state of the blood flow in the region through which the blood passes is superimposed to generate the display image.
Medical image processing equipment. The medical image processing apparatus according to any one of claims 1 to 4.
The cross-sectional morphological image is an MRI image captured by an MRI apparatus.
Medical image processing equipment. The medical image processing apparatus according to claim 5.
The MRI image is an image taken according to the Gradient Echo method.
Medical image processing equipment. The medical image processing apparatus according to any one of claims 1 to 6.
The diameter of the flow path through which the moving blood becomes the blood flow changes depending on the stenosis or aneurysm.
Medical image processing equipment. A medical image processing method in a medical image processing device.
Acquire multiple cross-sectional morphological images in chronological order, including regions of blood flow with at least one of vortices and retention.
Based only on the plurality of the cross-sectional morphological images, motion information regarding the deformation of the cross-sectional morphological images is acquired.
Based on at least one of the movement amount, the velocity, and the strain amount in the movement information, information indicating the state of blood flow is acquired, and information indicating the state of blood flow is acquired.
A display image is generated by superimposing information indicating the state of blood flow on the cross-sectional morphological image.
The information indicating the state of blood flow includes at least one information of a vortex and a retention having an in-plane component parallel to the cross section of the cross-sectional morphological image.
Medical image processing method. A medical image processing program for causing a computer to execute the medical image processing method according to claim 8.