JP2024061240A - Medical image processing device, medical image processing method and program - Google Patents

Abstract

[Problem] To improve the accuracy of medical simulation. [Solution] A medical image processing apparatus according to an embodiment includes an acquisition unit, an identification unit, and an estimation unit. The acquisition unit acquires medical image data. The identification unit identifies a first structure included in the medical image data. The estimation unit estimates a connection position of a second structure relative to the first structure. The estimation unit estimates the connection position based on the first mechanical parameter at a position on the first structure and a second mechanical parameter including the first mechanical parameter. [Selected Figure] Figure 1

Description

The embodiments disclosed in this specification and the drawings relate to a medical image processing device, a medical image processing method, and a program.

Research is being conducted into mitral valve simulation to predict the blood flow around the mitral valve and its surroundings based on physical principles, with the aim of predicting the outcome of treatment for mitral regurgitation. Mitral valve simulation makes it possible to predict, for example, the extent to which a particular type of surgery will improve the condition.

Here, when performing a mitral valve simulation, the chordae tendineae contribute greatly to the movement of the valve, so modeling the chordae tendineae is important in terms of improving the accuracy of the mitral valve simulation. However, because the chordae tendineae are thin structures, it can be difficult to obtain their position, thickness, and accurate number from CT images, ultrasound diagnostic images, etc., due to issues with spatial resolution.

JP 2020-502681 A

One of the problems that the embodiments disclosed in this specification and the drawings attempt to solve is to improve the accuracy of medical simulations. However, the problems that the embodiments disclosed in this specification and the drawings attempt to solve are not limited to the above problem. Problems that correspond to the effects of each configuration shown in the embodiments described below can also be positioned as other problems.

The medical image processing device according to the embodiment includes an acquisition unit, an identification unit, and an estimation unit. The acquisition unit acquires medical image data. The identification unit identifies a first structure included in the medical image data. The estimation unit estimates a connection position of a second structure relative to the first structure. The estimation unit estimates the connection position based on the first mechanical parameter at a position on the first structure and a second mechanical parameter including the first mechanical parameter.

FIG. 1 is a diagram showing an example of the configuration of a medical image processing apparatus 100 according to the first embodiment. FIG. 2 is a diagram illustrating the process according to the first embodiment. FIG. 3 is a flowchart illustrating the flow of processing performed by the medical image processing apparatus 100 according to the first embodiment. FIG. 4 is a diagram illustrating the process according to the first embodiment. FIG. 5 is a diagram illustrating the process according to the first embodiment. FIG. 6 is a flowchart illustrating the process of step S140 in FIG. 3 in more detail. FIG. 7 is a diagram showing an example of a screen displayed by the medical image processing apparatus 100 according to the first embodiment. FIG. 8 is a flowchart illustrating the flow of processing performed by the medical image processing apparatus 100 according to the third embodiment.

Below, the embodiments of the medical image processing device, medical image processing method, and program will be described in detail with reference to the drawings.

First Embodiment
First, a configuration example of a medical image processing apparatus 100 according to the first embodiment will be described with reference to Fig. 1. Note that the medical image processing apparatus 100 may be incorporated in other medical image diagnostic apparatuses, such as an ultrasound diagnostic apparatus, a magnetic resonance imaging apparatus, or a PET diagnostic apparatus.

In FIG. 1, the medical image processing device 100 includes a memory 132, an input device 134, a display 135, and a processing circuit 150. The processing circuit 150 includes an acquisition function 150a, an estimation function 150b, a display control function 150c, a reception function 150d, a specification function 150e, and a setting function 150f.

In the embodiment, each processing function performed by the acquisition function 150a, the estimation function 150b, the display control function 150c, the reception function 150d, the identification function 150e, and the setting function 150f is stored in the memory 132 in the form of a program executable by a computer. The processing circuit 150 is a processor that realizes the function corresponding to each program by reading the program from the memory 132 and executing it. In other words, the processing circuit 150 in a state in which each program has been read has each function shown in the processing circuit 150.

In FIG. 1, the processing functions performed by the acquisition function 150a, the estimation function 150b, the display control function 150c, the reception function 150d, the identification function 150e, and the setting function 150f are described as being realized by a single processing circuit 150, but the processing circuit 150 may be configured by combining multiple independent processors, and each processor may execute a program to realize the function. In other words, each of the above-mentioned functions may be configured as a program, and one processing circuit 150 may execute each program. As another example, a specific function may be implemented in a dedicated independent program execution circuit. In FIG. 1, the acquisition function 150a, the estimation function 150b, the display control function 150c, the reception function 150d, the identification function 150e, and the setting function 150f are examples of an acquisition unit, an estimation unit, a display control unit, a reception unit, an identification unit, and a setting unit, respectively. Also, the display 135 is an example of a display unit.

The term "processor" used in the above description refers to circuits such as a CPU (Central Processing Unit), a GPU (Graphical Processing Unit), or an application specific integrated circuit (ASIC), a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)). The processor performs functions by reading and executing programs stored in memory 132.

In addition, instead of storing the program in memory 132, the program may be directly embedded in the circuitry of the processor. In this case, the processor realizes the function by reading and executing the program embedded in the circuitry.

The processing circuit 150 acquires various information using the acquisition function 150a. The processing circuit 150 also performs various processes, which will be described later, using the estimation function 150b, the identification function 150e, and the setting function 150f.

The processing circuitry 150 controls the generation, display, etc. of images using the display control function 150c. As an example, the processing circuitry 150 causes the display control function 150c to display various images generated on the display 135. In addition, the processing circuitry 150 may use the display control function 150c to perform overall control of the medical image diagnostic device from which the medical image processing device 100 acquires data.

The processing circuit 150 uses the reception function 150d to receive various processes from the user, for example, via the input device 134.

Note that the above functions are merely examples, and the processing circuit 150 does not need to have all of the functions listed here.

The memory 132 stores data acquired from the medical image diagnostic device, image data generated by the processing circuit 150, etc. For example, the memory 132 is a semiconductor memory element such as a RAM (Random Access Memory), a flash memory, a hard disk, an optical disk, etc.

The input device 134 accepts various instructions and information input from the operator. The input device 134 is, for example, a pointing device such as a mouse or a trackball, a selection device such as a mode change switch, or an input device such as a keyboard. The display 135 displays a GUI (Graphical User Interface) for accepting input of imaging conditions, images generated by the processing circuit 150, etc. under the control of the processing circuit 150 having a display control function 150c. The display 135 is, for example, a display device such as a liquid crystal display.

Next, we will explain the background to the embodiment.

Research is being conducted into mitral valve simulation to predict the blood flow around the mitral valve and its surroundings based on physical principles, with the aim of predicting the outcome of treatment for mitral regurgitation. Mitral valve simulation makes it possible to predict the extent to which a particular type of surgery will improve the condition.

For example, the structure of a heart 10 is shown in FIG. 2. The mitral valve 1 is a valve provided, for example, between the left ventricle 4 and the left atrium 5. The opening and closing of the mitral valve 1 is controlled by chordae tendineae 2 extending from the papillary muscles 3.

Here, when performing a mitral valve simulation, the chordae tendineae 2 contribute significantly to the movement of the valve, so modeling the chordae tendineae 2 is important in terms of improving the accuracy of the simulation of the mitral valve 1. However, because the chordae tendineae 2 are thin structures, it can be difficult to obtain their position from CT images, ultrasound diagnostic images, etc., due to issues with spatial resolution.

Therefore, in the conventional technology, for example, the number, hardness, and joint positions of representative chordae 2 were obtained from literature values and the chordae were modeled without estimating the patient-specific chordae positions.

In contrast, in the medical image processing device 100 according to the embodiment, the position of the chordae tendineae 2 is estimated. Specifically, the medical image processing device 100 includes an acquisition unit, an identification unit, and an estimation unit. The acquisition unit acquires medical image data. The identification unit identifies a first structure included in the medical image data. The estimation unit estimates the connection position of the second structure relative to the first structure based on a first mechanical parameter at the position of the first structure and a second mechanical parameter including the first mechanical parameter. Here, the first structure is, for example, the mitral valve 1, and the second structure is, for example, the chordae tendineae 2.

In addition, in the medical image processing method according to the embodiment, medical image data is acquired, a first structure contained in the medical image data is identified, and a connection position of a second structure relative to the first structure is estimated based on a first mechanical parameter at a position on the first structure and a second mechanical parameter including the first mechanical parameter.

The program according to the embodiment also causes the computer to execute a procedure of acquiring medical image data, identifying a first structure contained in the medical image data, and estimating a connection position of a second structure relative to the first structure based on a first mechanical parameter at a position on the first structure and a second mechanical parameter including the first mechanical parameter.

This allows mitral valve simulation to be performed based on the unique chordae tendineae position that differs for each patient, improving the accuracy of the mitral valve simulation.

The process flow according to the embodiment will be described with reference to Figures 3 to 6. Figure 3 is a flowchart showing the process flow according to the embodiment.

First, in step S100, the processing circuitry 150 acquires medical image data using the acquisition function 150a. Here, medical images include CT (Computed Tomography) images, MR (Magnetic Resonance) images, ultrasound images, and blood vessel depiction images. The processing circuitry 150 acquires CT images from an X-ray CT device, MR images from MRI (Magnetic Resonance Imaging) images, and ultrasound images from an ultrasound diagnostic device using the acquisition function 150a.

In addition, the medical image data acquired by the processing circuitry 150 in step S100 is, for example, data of multiple time phases. As an example, the processing circuitry 150 acquires data of a total of two time phases, the temporary phase of the diastole in which the mitral valve, which is the first structure, is open, and the temporary phase of the systole, as data of multiple time phases, by the acquisition function 150a. That is, the processing circuitry 150 acquires medical image data in the first time phase and the second time phase by the acquisition function 150a. The former time phase is, for example, a time phase in which the cardiac cycle is in the range of 41% to 100%, and the latter time phase is, for example, a time phase in which the cardiac cycle is in the range of 1% to 40%. These medical image data are data selected from the medical image data of multiple time phases at a timing when the balance of the force due to the first mechanical parameter and the second structure is balanced, as described later.

As another example, the processing circuitry 150 may use the acquisition function 150a to acquire data for a total of three time phases as data for multiple time phases. For example, the processing circuitry 150 may use the acquisition function 150a to acquire data for three time phases with a cardiac cycle of 0%, 30%, and 60% as data for multiple time phases. As another example, the processing circuitry 150 may use the acquisition function 150a to acquire MR images or echo Doppler images to obtain velocity information on the movement of the mitral valve leaflets.

Next, in step S110, the processing circuitry 150 uses the identification function 150e to identify a first structure included in the medical image data. Here, the first structure is, for example, the mitral valve 1. For example, the processing circuitry 150 uses the identification function 150e to perform segmentation using deep learning or an energy method on the CT image acquired in step S100 to extract the shape of the mitral valve 1, which is the first structure, from the medical image data in multiple time phases. In addition, the processing circuitry 150 uses the identification function 150e to acquire velocity information of the movement of the valve leaflets from the medical image data acquired in step S100. As an example, the processing circuitry 150 uses the identification function 150e to acquire velocity information of the movement of the valve leaflets in multiple time phases from the echo Doppler image acquired in step S100.

Next, in step S120, the processing circuit 150 uses the setting function 150f to set initial values of the physical property values of the first structure and the second structure based on, for example, literature values, etc.

Here, the second structure is, for example, the chordae tendineae 2. The medical image processing device 100 according to the embodiment can include, as the second structure, a structure that is not imaged in medical image data, such as X-ray CT image data, due to, for example, spatial resolution constraints.

The processing circuit 150 uses the setting function 150f to set, for example, the Young's modulus of the mitral valve 1 and the parameters of the constitutive equation of the hyperelastic body of the mitral valve 1 as the initial values of the physical property values of the first structure. Here, since the mitral valve 1 is an anisotropic body, the processing circuit 150 may use the setting function 150f to set the parameters of the constitutive equation having anisotropy as the initial values of the physical property values of the first structure.

Next, in step S130, the processing circuit 150 uses the setting function 150f to set initial values for various boundary conditions based on literature, etc. Here, the boundary conditions are, for example, the change over time in pressure on the mitral valve 1, the amount of movement of the junction between the mitral valve 1 and the left ventricle 4, or the tension of the chordae tendineae 2, etc.

The processing circuit 150 may use the setting function 150f to calculate the initial setting values of these boundary conditions based on literature values, or based on information obtained from the image, or by combining information obtained from the image with literature values.

As an example, the processing circuit 150 uses the setting function 150f to provisionally determine the junction between the mitral valve 1 and the chordae tendineae 2 based on literature values, obtains the position of the papillary muscle 3 from a medical image, and further determines the natural length of the chordae tendineae 2 to calculate the tension due to the chordae tendineae 2 and set it as the initial value of the boundary condition.

In other words, the processing circuit 150 sets the initial values of the boundary conditions of the first mechanical parameter and the second mechanical parameter by the setting function 150f.

Here, the first mechanical parameter is, for example, a parameter based on blood pressure, and the second mechanical parameter is the resultant force of blood pressure and the second structure, chordae tendineae 2. That is, the processing circuit 150 uses the setting function 150f to set the initial values of the parameter based on blood pressure and the boundary condition of the resultant force of blood pressure and chordae tendineae 2.

A schematic diagram of such a situation is shown in FIG. 4. Arrow 30 indicates the tension due to the chordae tendineae 2. Arrows 31 and 32 indicate the pressure exerted on the mitral valve 1 by blood pressure. That is, arrows 31 and 32 are an example of a first mechanical parameter set by the processing circuit 150 using the setting function 150f, and the vector sum of arrows 30, 31, and 32 is an example of a second mechanical parameter set by the processing circuit 150 using the setting function 150f.

Here, the method in which the processing circuit 150 estimates the first mechanical parameter using the setting function 150f is called the first method, and the method in which the processing circuit 150 estimates the second mechanical parameter is called the second method. The second method is generally different from the first method. As an example, the processing circuit 150 calculates the first mechanical parameter, blood pressure, using the first method based on literature values using the setting function 150f. On the other hand, as described above, the processing circuit 150 uses the setting function 150f to provisionally determine the junction between the mitral valve 1 and the chordae tendineae 2 based on literature values, and then obtains the position of the papillary muscle 3 from a medical image, and further determines the natural length of the chordae tendineae 2 to calculate the tension of the chordae tendineae 2, and calculates the resultant force with the force due to blood pressure using the second method.

Next, in step S140, the processing circuit 150 uses the estimation function 150b to estimate the connection position of the second structure relative to the first structure based on the first mechanical parameter and the second mechanical parameter at the position on the first structure. At this time, the processing circuit 150 uses data assimilation techniques such as Bayesian estimation or machine learning using the estimation function 150b.

The specific processing of step S140 will be explained in detail below using the flowchart in Figure 6, but an overview of the processing of this embodiment will be explained using Figure 5.

It is generally difficult to estimate the position of the chordae tendineae 2, which is the second structure, relative to the mitral valve 1, which is the first structure, but in this embodiment, the processing circuit 150 estimates the connection position of the chordae tendineae 2 relative to the mitral valve 1 using the estimation function 150b to estimate the tension due to the virtual chordae tendineae and the force due to the virtual blood pressure that balances with the tension due to the chordae tendineae, arrow 30, and the forces due to the blood pressure, arrows 31 and 32, shown in FIG. 4. In FIG. 5, the resultant force of arrows 41 and 42 is a schematic representation of the force due to the virtual chordae tendineae, and arrow 40 is a schematic representation of the force due to the virtual pressure.

Now, regarding the virtual chordae, in FIG. 4, the resultant force of arrows 30, 31, and 32 is the resultant force calculated for the position of the chordae 2 assumed at the time of that step. Considering that these resultant forces should be zero, it can be considered that the resultant force of the arrows in FIG. 4 was not zero because the position of the chordae 2 was shifted from the true position. Therefore, the processing circuit 150 adds a virtual chordae to the system, and updates the position of the chordae 2 so that the sum of the resultant force of the tension and blood pressure forces of the chordae assumed in the previous steps and the resultant force of the tension and virtual blood pressure forces of the newly added virtual chordae is zero. In other words, the processing circuit 150 updates the position of the chordae 2, etc., by the estimation function 150b so that the resultant force of the tension and blood pressure forces of the chordae and the resultant force of the tension and virtual blood pressure forces of the virtual chordae are balanced.

At this time, the processing circuit 150 uses the estimation function 150b to identify the position where there is a difference between the first mechanical parameter and the second mechanical parameter as the connection position of the second structure to the first structure. In other words, the processing circuit 150 uses the estimation function 150b to identify, for example, a position where the force exerted by the second structure, the chordae tendineae 2, on the first structure, the mitral valve 1, is high as the connection position of the second structure to the first structure, in a state where the balance between the force exerted by the chordae tendineae 2, the second structure, on the mitral valve 1, the first structure, and the forces generated in the blood flow and blood pressure is balanced.

The process of step S140 will be described in more detail with reference to FIG. 6. FIG. 6 is a diagram illustrating the process of step S140 in FIG. 3 in more detail.

First, in step S141, the processing circuit 150 starts the calculation from the first phase. Here, the first phase is, for example, the diastole.

Next, in step S142, the processing circuit 150 uses the estimation function 150b to calculate the external and internal forces acting on the first structure using the boundary conditions and physical properties set in steps S130 and S140, with the shape of the first structure in the first time phase as the initial shape. As an example, the processing circuit 150 uses the estimation function 150b to calculate the external and internal forces acting on the first structure using the boundary conditions and physical properties set in steps S130 and S140, with the shape of the mitral valve 1 in the diastolic phase as the initial shape of the first structure, for example, using the finite element method, finite difference method, ALE (Arbitrary Lagrangian and Eularian) method, etc. In other words, the processing circuit 150 uses the estimation function 150b to calculate the balance of forces in the first time phase.

Next, in step S143, the processing circuit 150 predicts the shape of the first structure after a short time by deforming the first structure based on the force calculated in step S142 using the estimation function 150b.

Next, in step S144, the processing circuit 150 advances the time phase by a small amount of time.

Next, in step S145, the processing circuit 150 repeats the processes of steps S142 to S145 until the time phase reaches the second time phase (step S145 No). Here, the second time phase is, for example, the cardiac systole. When the time phase reaches the second time phase (step S145 Yes), the process proceeds to step S146.

In the above example, the first time phase is diastolic and the second time phase is systolic, but the embodiment is not limited to this. That is, for example, the first time phase may be systolic and the second time phase may be diastolic.

Next, in step S146, the processing circuitry 150 estimates the shape of the first structure in the second time phase using the estimation function 150b. As described above, the processing circuitry 150 estimates the shape of the first structure in the second time phase using the estimation function 150b based on the force balance calculated in the first time phase and the medical image data in the first time phase.

Next, in step S147, the processing circuitry 150 calculates a loss function based on the shape of the first structure estimated by the estimation function 150b and the shape of the first structure of the medical image data in the second phase. That is, the processing circuitry 150 performs optimization using the deviation between the shape of the first structure in the second phase estimated by the estimation function 150b and the actual shape of the second phase obtained from the medical image data as the loss function. The optimization parameters optimized here are, for example, physical property values such as the position, size, and tension of the chordae tendineae, as well as various boundary conditions such as the junction position of the chordae tendineae and blood pressure.

That is, the processing circuitry 150 uses the estimation function 150b to compare the estimated shape of the first structure with the shape of the first structure in the medical image data at the second time phase, thereby estimating, for example, the connection position of the second structure relative to the first structure.

Next, in step S148, if the loss function calculated in step S147 is not within a certain value (step S148 No), the process proceeds to step S149, where the processing circuit 150 modifies the connection position of the second structure, etc., using the estimation function 150b, and the process returns to step S141. In step S149, the processing circuit 150 modifies, for example, the position, size, number, tension, and joint position of the tendon, using the estimation function 150b, and as a result, the process from step S141 onwards is started again.

On the other hand, if the calculated loss function is within a certain value (Yes in step S149), the processing in step S140 is completed.

Returning to FIG. 3, next, in step S150, the processing circuit 150 causes the display control function 150c to display the connection position of the second structure in the display area of the display 135, for example, as shown in FIG. 7. For example, the processing circuit 150 uses the display control function 150c to display the connection position of the second structure at a corresponding position on a 2D or 3D image under display conditions that are set in advance. The display conditions include color conditions such as RGB values, transparency, and saturation, thickness, and shape. The processing circuit 150 may also display the second structure by changing the texture instead of the color, for example, using the display control function 150c. The processing circuit 150 may also change the display based on the estimation result in step S140 or the reliability of the estimation using the display control function 150c.

As described above, the medical image processing apparatus 100 according to the first embodiment estimates the connection position of the second structure relative to the first structure. This makes it possible to perform a simulation that takes into account the connection position of the chordae tendineae 2 for each patient, for example, and improves the accuracy of the simulation of the chordae tendineae 2.

(First Modification of the First Embodiment)
The method of displaying the simulation results described in the first embodiment is not limited to the above-mentioned example. For example, the processing circuit 150 may display the estimation results in various forms such as the following using the display control function 150c.

As an example, the processing circuit 150 may change the color of the chordae 2 displayed according to the estimated length of the chordae 2 by using the display control function 150c. For example, the processing circuit 150 may display chordae 2 with a length of more than 40 mm in red and chordae with a length of less than 40 mm in blue by using the display control function 150c. As another example, the processing circuit 150 may display the chordae 2 in different colors for each region by using the display control function 150c. As an example, the processing circuit 150 may display the region from the connection with the mitral valve 1 up to 40 mm in red and the region exceeding 40 mm in blue by using the display control function 150c. Also, the processing circuit 150 may gradually change the displayed color from red to blue according to the length of the chordae 2 by using the display control function 150c.

As another example, the processing circuit 150 may use the display control function 150c to change the color of the chordae for each patient according to the average length of the chordae 2. For example, the processing circuit 150 may use the display control function 150c to calculate the average length of all chordae 2 for one patient, and display all chordae 2 in red for patients whose length exceeds 40 mm, and display the other chordae 2 in blue.

As another example, the processing circuit 150 may use the display control function 150c to change the color of the tendon 2 depending on the stiffness of the tendon 2. As one example, the processing circuit 150 may use the display control function 150c to display in red those tendon 2 whose estimated stiffness exceeds 500 kPa, and in blue those whose estimated stiffness is less than 500 kPa.

As another example, the processing circuit 150 may change the color of the chordae 2 according to the number of chordae 2. For example, the processing circuit 150 may use the display control function 150c to display in red those with more than 50 chordae 2 connected to the valve of the patient being estimated, and in blue those with fewer than 50 chordae 2. As another example, the processing circuit 150 may use the display control function 150c to classify the regions and change the color of the chordae according to the number of chordae 2 in each region. For example, the processing circuit 150 may use the display control function 150c to display in red those with more than 30 chordae 2 connected to the anterior leaflet, and in blue those with fewer than 30 chordae 2.

As another example, the processing circuit 150 may change the color of the chordae 2 depending on the type of the chordae 2. As one example, the processing circuit 150 uses the display control function 150c to display what is estimated to be a basal chordae in red, what is estimated to be a marginal chordae in blue, and what is estimated to be a struct chordae in green.

The processing circuit 150 may also use the estimation function 150b to calculate the confidence level of the classification, and the display control function 150c to display in gray those tendons 2 whose confidence level falls below a certain value.

As another example, the processing circuit 150 may use the estimation function 150b to calculate the likelihood of the position of the tendon 2 and change the color depending on the likelihood. As an example, the processing circuit 150 may use the display control function 150c to display in red an area that is thought to be the position of the tendon 2 with a 90% likelihood, and in yellow an area that is thought to be the tendon 2 with a likelihood of 80% to 90%. The processing circuit 150 may further display in red any one tendon 2 with a likelihood exceeding 90%, and in yellow any one tendon 2 with a likelihood exceeding 80% to 90%. The processing circuit 150 may also use the display control function 150c to change the color, brightness, thickness, etc. depending on the number of tendon 2 with a likelihood exceeding 90%.

As another example, the processing circuitry 150 may use the estimation function 150b to display both tendon and artificial tendon extracted directly from CT images, echo images, etc., and change the colors of the extraction and estimation results. In this case, the processing circuitry 150 may use the display control function 150c to display the directly extracted tendon 2 in red and the estimated tendon 2 in blue. Furthermore, if the extraction and estimation results indicate the same tendon 2, the processing circuitry 150 may use the display control function 150c to display them in different colors.

As another example, the processing circuit 150 may allow the user to change the position of the tendon 2, and may change the color of the tendon 2 that has been changed from that of the other tendon 2. For example, the tendon 2 that has been changed may be shown in red, and the other tendon 2 in blue. The processing circuit 150 may change the color depending on the number of changes by using the display control function 150c.

As another example, the processing circuit 150 may use the display control function 150c to determine whether the chordae 2 will affect treatment and change the color according to the result of the determination. As an example, when placing a mitraclip, the chordae 2 may become entangled in the catheter. Therefore, the processing circuit 150 may use the reception function 150d to receive the position for placing the mitraclip from a user input, and color the chordae 2 that are present within a 5 mm radius of the received position.

As described above, in the first embodiment, the processing circuit 150 uses the display control function 150c to display the estimation results to the user in various forms. This can improve user convenience.

Second Embodiment
In the first embodiment, a case has been described in which the first structure is a valve, and the second structure for estimating the connection position with respect to the first structure is a chordae tendineae. However, the embodiment is not limited to this. In the second embodiment, a case will be described in which the first structure is a facial muscle, and the second structure is a ligand that is bonded to the facial muscle.

First, in step S100, the processing circuitry 150 acquires image data using the acquisition function 150a. The image data here may be medical images such as CT images and MR images, or may be ordinary photographic data including the subject's face, for example. The processing circuitry 150 acquires image data at multiple time phases using the acquisition function 150a.

Next, in step S110, the processing circuit 150 uses the specification function 150e to extract the shape of the facial muscles, which is the first structure, from the image data extracted in step S100. As an example, the processing circuit 150 uses the specification function 150e to extract the shapes of the facial muscles at multiple time phases from multiple images. For example, at this time, it is desirable for the facial expressions to be different depending on each time point.

Next, in step S120, the processing circuit 150 uses the setting function 150f to set initial values of the physical property values of the first structure and the second structure based on, for example, literature values, etc.

Here, the physical property value is, for example, the Young's modulus of the facial muscle or the parameters of the constitutive equation of a hyperelastic body. Since muscles are known to be anisotropic bodies, the processing circuit 150 may use the setting function 150f to set the initial values of the physical property values using an anisotropic constitutive equation.

Next, in step S130, the processing circuit 150 sets various initial values of boundary conditions based on literature values, etc., using the setting function 150f. Here, the boundary conditions refer to, for example, pressure from bones or fat, pressure from an object pressed against the face, etc. Note that the embodiment is not limited to the case where the boundary conditions are set based on literature values, etc., and may accept input from a user and set the boundary conditions based on the accepted input result.

Next, in step S140, the processing circuit 150 uses the estimation function 150b to estimate the position of the ligand, which is a second structure, relative to the mimetic muscle, which is a first structure. As an example, the processing circuit 150 uses the estimation function 150b to estimate the shape of the mimetic muscle in the second time phase based on the shape of the mimetic muscle in the first time phase, and compares it with the shape of the mimetic muscle in the actual image in the second time phase to calculate a loss function, thereby estimating the position of the ligand relative to the mimetic muscle.

Next, in step S150, the processing circuit 150 causes the display control function 150c to display the position of the ligand relative to the mimetic muscles in the display area of the display 135.

As described above, in the second embodiment, the method of the first embodiment is applied to the mimetic muscles. This makes it possible to simulate the position of the ligand relative to the mimetic muscles.

Third Embodiment
In the third embodiment, a case will be described in which, rather than performing modeling directly based on medical image data acquired by the processing circuitry 150, a simulation is performed based on the results of a different model set up based on anatomical knowledge.

This situation is shown in FIG. 8. First, in step S100, the processing circuitry 150 acquires a medical image using the acquisition function 150a. As an example, the processing circuitry 150 acquires an X-ray CT image of the heart using the acquisition function 150a.

Next, in step S300, the processing circuitry 150 uses the specific function 150e to extract a first structure and a second structure in a first model from the medical image acquired in step S100. Here, the first model is a model of a patient who is the subject of the simulation. As an example, the processing circuitry 150 uses the specific function 150e to acquire the first structure, the mitral valve 1 structure, and the second structure, the chordae tendineae 2 structure, in multiple phases from the X-ray CT image acquired in step S100.

Next, in step S310, the processing circuit 150 uses the setting function 150f to set the physical property values of the first structure and the second structure in the second model based on the literature values. Here, the second model is, for example, a chordae model set based on literature values, etc., and is a model different from the first model, which is a model of the patient that is the subject of the simulation described above.

Next, in step S320, the processing circuitry 150 uses the setting function 150f to set initial values of the boundary conditions of the first structure and the second structure in the second model based on the literature values. That is, the processing circuitry 150 uses the setting function 150f to set initial values of the boundary conditions of the mitral valve 1 and the chordae tendineae 2 in the chordae tendineae model, which is the second model.

Next, in step S330, the processing circuit 150 uses the setting function 150f to set a model of the second structure in the second model. As an example, the processing circuit 150 uses the setting function 150f to set chordae tendineae 2 in the chordae tendineae model.

Next, in step S340, the processing circuit 150 uses the estimation function 150b to integrate the first model and the second model to estimate the connection position of the second structure relative to the first structure. As an example, the processing circuit 150 uses the estimation function 150b to integrate the first model and the second model after removing the second structure, i.e., the chordae tendineae 2, from the first model. As an example, the processing circuit 150 uses the estimation function 150b to integrate the first model and the second model so that, for example, (1) a specific point on the left ventricle side of the mitral valve 1 is joined to one side of the chordae tendineae 2 of the chordae tendineae model, which is the second model, and (2) the papillary muscle 3 is joined to one end of the string structure. The processing circuit 150 also estimates the parameters of the mitral valve using the estimation function 150b, for example, by performing data assimilation.

As described above, in the third embodiment, the processing circuit 150 uses a first model, which is a patient model, and a second model different from the first model, and integrates them to estimate the chordae tendineae 2. This makes it possible to estimate the parameters of the mitral valve 1 without modeling the chordae tendineae 2 of the patient.

According to at least one of the embodiments described above, the accuracy of medical simulations can be improved.

Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are within the scope of the invention and its equivalents as set forth in the claims, as well as the scope and spirit of the invention.

150 Processing circuit 150a Acquisition function 150b Estimation function 150c Display control function 150d Reception function 150e Identification function 150f Setting function

Claims (12)

An acquisition unit for acquiring medical image data;
An identification unit that identifies a first structure included in the medical image data;
an estimation unit that estimates a connection position of a second structure relative to the first structure,
The estimation unit estimates the connection position based on a first mechanical parameter at the first structural position and a second mechanical parameter including the first mechanical parameter.
Medical imaging equipment. The medical image processing device according to claim 1, wherein the estimation unit identifies the position where there is a difference between the first mechanical parameter and the second mechanical parameter as the connection position. The medical image processing device according to claim 1, wherein the first mechanical parameter is a parameter estimated by a first method, and the second mechanical parameter is a parameter estimated by a second method different from the first method. The medical image processing device according to claim 1, wherein the second structure is a structure that is not imaged in the medical image data. The medical image processing device according to claim 1, wherein the first structure is a valve and the second structure is a chordae tendineae. The medical image processing device according to claim 1, wherein the first structure is a facial muscle, and the second structure is a ligand that binds to the facial muscle. The medical image processing device according to claim 1, wherein the medical image data is data selected from medical image data of multiple time phases at a timing when the balance between the first mechanical parameter and the force due to the second structure is balanced. The medical image processing device according to claim 3, wherein the first mechanical parameter is a parameter based on blood pressure, and the second mechanical parameter is a resultant force of the blood pressure and a force due to the second structure. The acquisition unit acquires the medical image data in a first time phase and a second time phase,
2. The medical image processing device according to claim 1, wherein the estimation unit calculates a force balance in the first time phase, estimates a shape of the first structure in the second time phase based on the calculated force balance and the medical image data in the first time phase, and estimates the connection position by comparing the estimated shape of the first structure with a shape of the first structure in the medical image data in the second time phase. The medical image processing device according to claim 1, wherein the second mechanical parameter is a resultant force of blood pressure and a force due to the second structure. Acquire medical image data;
Identifying a first structure included in the medical image data;
estimating a connection position of a second structure relative to the first structure based on a first mechanical parameter at a position on the first structure and a second mechanical parameter including the first mechanical parameter;
Medical image processing method. Acquire medical image data;
Identifying a first structure included in the medical image data;
A program that causes a computer to execute a procedure for estimating a connection position of a second structure relative to the first structure based on a first mechanical parameter at a position on the first structure and a second mechanical parameter including the first mechanical parameter.

Images