Medical information processing device, medical image diagnosis device, and medical image processing method

Abstract

To estimate "an unknown component existing in a living body but not representable by a physical model", of "noise components" and "an unknown component existing in a living body but not representable by a physical model" in component estimation using a physical model in consideration of noise components and unknown components.SOLUTION: A medical information processing device comprises an estimation unit, a first evaluation unit, a second evaluation unit, and an updating unit. The estimation unit estimates component amounts of components in a living body depicted in a medical image, and component amounts of noise components and unknown components, using a physical model in consideration of noise components and unknown components. The first evaluation unit calculates, as a first evaluation value, a likelihood of the component amounts for the physical model. The second evaluation unit calculates a second evaluation value based on morphological characteristics of the components in the living body, the noise components, and the unknown components. The updating unit updates the component amounts of the components in the living body and the component amounts of the noise components and the unknown components on the basis of the first evaluation value and the second evaluation value.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] An embodiment of the present invention relates to a medical information processing apparatus, a medical image diagnostic apparatus, and a medical image processing method. [Background technology]

[0002] Imaging information plays an important role in doctors' diagnoses. X-ray CT (Computed Tomography) visualizes the internal structure of an object based on the amount of transmitted X-rays. Optical imaging visualizes an object based on reflected light. While the amount of a component in a living body is important in diagnosis, doctors cannot accurately determine the amount of a component based on images alone. For example, doctors estimate the amount of a component to some extent by comparing the brightness of the area of ​​the object for which the amount of the component is to be estimated with the brightness of other areas.

[0003] A method has been proposed for estimating biological components from the observed photon counts in photon-counting CT. This method uses maximum likelihood estimation based on a physical model to estimate the amount of biological components at each location from the photon count. However, if there are unknown components not considered in the physical model, the estimation may fail. One approach to address this issue is to use a physical model that takes unknown components into account, but this increases the number of estimation targets and increases the likelihood of multiple optimal solutions. As a result, it is not possible to distinguish between "noise components due to external factors" and "unknown components contained in the biological body but not represented by the physical model," and these components may be displayed as noise when imaged. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-161693 [Patent Document 2] Special Publication No. 2009-512528 Summary of the Invention [Problem to be solved by the invention]

[0005] The problem to be solved by the present invention is to estimate "unknown components that are contained in a living body but cannot be expressed by a physical model" out of "noise components" and "unknown components that are contained in a living body but cannot be expressed by a physical model" in component estimation using a physical model that takes noise components and unknown components into consideration. [Means for solving the problem]

[0006] A medical image processing device according to an embodiment includes an acquisition unit, an estimation unit, a first evaluation unit, a second evaluation unit, and an update unit. The acquisition unit acquires a medical image depicting a living body. The estimation unit estimates the component amounts of in-vivo components depicted in the medical image, as well as the component amounts of noise components and unknown components, using a physical model that takes noise components and unknown components into account. The first evaluation unit calculates the likelihood of the component amounts relative to the physical model as a first evaluation value. The second evaluation unit calculates a second evaluation value based on morphological characteristics of the in-vivo components, noise components, and unknown components. The update unit updates the component amounts of in-vivo components, as well as the component amounts of noise components and unknown components, based on the first evaluation value and the second evaluation value. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of a medical information processing system according to the first embodiment. [Figure 2] FIG. 2 is a diagram showing an example of a case where a plurality of components are depicted in a CT image, the component amounts of the respective components are estimated, and a plurality of component maps are generated by performing reconstruction on the estimated component amounts. [Figure 3] FIG. 3 is a diagram showing an example of a case where a plurality of components are depicted in a CT image, the component amounts of the respective components are estimated, and a plurality of component maps are generated by performing reconstruction on the estimated component amounts. [Figure 4] FIG. 4 is a diagram for explaining an example of processing executed by the morphological feature evaluation function according to the first embodiment. [Figure 5] FIG. 5 is a diagram for explaining an example of processing executed by the morphological feature evaluation function according to the first embodiment. [Figure 6] FIG. 6 is a diagram for explaining another example of the processing executed by the morphological feature evaluation function according to the first embodiment. [Figure 7] FIG. 7 is a diagram showing an example of a display image according to the first embodiment. [Figure 8] FIG. 8 is a flowchart showing the flow of an example of processing executed by the medical image processing apparatus according to the first embodiment. [Figure 9] FIG. 9 is a flowchart showing the flow of another example of the processing executed by the medical image processing apparatus according to the first embodiment. [Figure 10] FIG. 10 is a diagram showing an example of the configuration of an X-ray CT apparatus according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] Hereinafter, embodiments of a medical information processing apparatus, a medical image diagnostic apparatus, and a medical image processing method will be described in detail with reference to the accompanying drawings.

[0009] (First embodiment) In the first embodiment, a medical information processing system 1 including a medical information processing device 100 will be described as an example. For example, as shown in Fig. 1, the medical information processing system 1 includes the medical information processing device 100, a medical image diagnostic device 200, and a database 300. Fig. 1 is a block diagram showing an example of the configuration of the medical information processing system 1 according to the first embodiment.

[0010] 1, a medical information processing device 100, a medical image diagnostic device 200, and a database 300 are connected via a network 90. ​​Here, the network 90 may be configured as a closed local network within a hospital, or may be a network via the Internet. For example, the network 90 includes a LAN (Local Area Network) or a WAN (Wide Area Network).

[0011] The medical image diagnostic device 200 is a device that collects medical images (medical image data) depicting an examination target region of a subject. That is, the medical images are images depicting a living body. The medical image diagnostic device 200 transmits the medical images to the database 300 and the medical information processing device 100. The medical image diagnostic device 200 includes, for example, at least one of an X-ray CT device, an ultrasound diagnostic device, an MRI (Magnetic Resonance Imaging) device, and a visible light camera. When the medical image diagnostic device 200 is an X-ray CT device, the medical image diagnostic device 200 collects CT images (CT image data). Such an X-ray CT device is, for example, a device that can perform photon-counting CT. That is, such an X-ray CT device is a device that can reconstruct an X-ray CT image by counting X-ray photons that have passed through the subject using a photon-counting X-ray detector (photon-counting detector). When the medical image diagnostic apparatus 200 is an ultrasound diagnostic apparatus, the medical image diagnostic apparatus 200 collects ultrasound images (ultrasound image data). When the medical image diagnostic apparatus 200 is an MRI apparatus, the medical image diagnostic apparatus 200 collects MR images (MR image data). When the medical image diagnostic apparatus 200 is a visible light camera, the medical image diagnostic apparatus 200 collects optical images (optical image data). Below, an example will be described in which the medical image diagnostic apparatus 200 is an X-ray CT apparatus capable of performing photon-counting CT.

[0012] The database 300 is a storage device that stores various data, and is realized by computer equipment such as a server or a workstation. The database 300 may be a server of an information management system such as a Radiology Information System (RIS), a Hospital Information System (HIS), or a Picture Archiving and Communication System (PACS). For example, the database 300 includes an image storage device that stores medical images collected by the medical image diagnostic device 200. Although a single database 300 is shown in FIG. 1, the database 300 may be realized by a combination of multiple storage devices.

[0013] The medical information processing device 100 assists doctors in making diagnoses. For example, the medical information processing device 100 performs component estimation using a physical model that takes unknown components into account. In addition, in the component estimation using the physical model that takes unknown components into account, the medical information processing device 100 distinguishes between "noise" and "unknown components that are contained in the living body but cannot be expressed by the physical model," and mainly estimates the "unknown components that are contained in the living body but cannot be expressed by the physical model," and displays them as images. The medical information processing device 100 is, for example, an example of a medical image processing device.

[0014] As shown in FIG. 1, the medical information processing device 100 includes a communication interface 101, an input interface 102, a display 103, a memory 104, and a processing circuit 105.

[0015] The communication interface 101 is configured by, for example, a network card such as a LAN card, a network adapter, etc. The communication interface 101 transmits and receives various information to and from devices connected via the network 90 under the control of the processing circuit 105.

[0016] The input interface 102 accepts various input operations from a user, converts the accepted input operations into electrical signals, and outputs the electrical signals to the processing circuit 105. For example, the input interface 102 may be implemented by a mouse, keyboard, trackball, switch, button, joystick, a touchpad that performs input operations by touching the operation surface, a touchscreen that integrates a display screen and a touchpad, a non-contact input circuit using an optical sensor, a voice input circuit, or the like. The input interface 102 may also be configured as a tablet terminal or the like that can wirelessly communicate with the medical information processing device 100. The input interface 102 may also be a circuit that accepts input operations from a user using motion capture. For example, the input interface 102 can accept the user's body movements, line of sight, and the like as input operations by processing signals acquired via a tracker and images collected about the user. The input interface 102 is not limited to those that include physical operating components such as a mouse and keyboard. For example, an example of the input interface 102 also includes an electrical signal processing circuit that receives an electrical signal corresponding to an input operation from an external input device provided separately from the medical information processing device 100 and outputs this electrical signal to the processing circuit 105.

[0017] The display 103 displays various types of information and images. For example, the display 103 displays various images such as medical images (medical images based on medical image data) collected by the medical image diagnostic apparatus 200 under the control of the processing circuitry 105. In addition, for example, the display 103 displays a GUI (Graphical User Interface) for receiving various instructions, settings, etc. from a user via the input interface 102. For example, the display 103 is a liquid crystal display or a CRT (Cathode Ray Tube) display. The display 103 may be a desktop type, or may be configured as a tablet terminal or the like capable of wireless communication with the main body of the medical information processing apparatus 100. The display 103 is an example of a display unit.

[0018] The medical information processing device 100 may include a projector instead of or in addition to the display 103. The projector can project onto a screen, wall, floor, etc. under the control of the processing circuitry 105. For example, the projector can project onto any plane, object, space, etc. by projection mapping. Such a projector is an example of a display unit.

[0019] The memory 104 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory), a flash memory, a hard disk, an optical disk, or the like. For example, the memory 104 stores various data (medical image data) such as various medical images transmitted from the medical image diagnostic apparatus 200. The following description will be given taking as an example a case where the memory 104 stores CT images transmitted from the medical image diagnostic apparatus 200. The CT images stored in the memory 104 are two-dimensional or three-dimensional image data. The memory 104 also stores programs that enable circuits included in the medical information processing apparatus 100 to realize various functions. The memory 104 may be realized by a group of servers (cloud) connected to the medical information processing apparatus 100 via a network 90.

[0020] The processing circuit 105 includes an acquisition function 105a, a biological component estimation function 105b, a physical model evaluation function 105c, a morphological feature evaluation function 105d, a display image generation function 105e, and a display control function 105f. The acquisition function 105a is an example of an acquisition section. The biological component estimation function 105b is an example of an estimation section and an example of an update section. The physical model evaluation function 105c is an example of a first evaluation section. The morphological feature evaluation function 105d is an example of a second evaluation section. The display image generation function 105e is an example of a generation section. The display control function 105f is an example of a display control section.

[0021] 1, each processing function is stored in the form of a computer-executable program in the memory 104. The processing circuitry 105 is a processor that realizes the function corresponding to each program by reading and executing each program from the memory 104. In other words, the processing circuitry 105 in a state in which a program has been read has various functions corresponding to the read program.

[0022] 1, the acquisition function 105a, biological component estimation function 105b, physical model evaluation function 105c, morphological feature evaluation function 105d, display image generation function 105e, and display control function 105f are described as being realized by a single processing circuit 105, but the processing circuit 105 may be configured by combining multiple independent processors, and each processor may execute a program to realize the functions. Furthermore, each processing function of the processing circuit 105 may be realized by being appropriately distributed or integrated into a single or multiple processing circuits.

[0023] The processing circuitry 105 may also realize its functions by using a processor of an external device connected via the network 90. ​​For example, the processing circuitry 105 reads and executes a program corresponding to each function from the memory 104, and realizes each function shown in Fig. 1 by using a group of servers (cloud) connected to the medical information processing device 100 via the network 90 as a computational resource.

[0024] Here, when executing various processes, the processing circuit 105 specifies an address in the memory space of the memory 104, and writes various data, such as data used in the various processes and data generated in the various processes, to the specified address (an area of ​​the memory 104 corresponding to the specified address). Furthermore, when executing various processes, the processing circuit 105 specifies an address in the memory space of the memory 104 to which the data used in the various processes has been written, reads the data written to the specified address from the memory 104, and executes various processes using the read data.

[0025] 2 and 3 are diagrams illustrating an example in which, when components 50a, 50b, and 50c are depicted in a CT image 50, the component amounts of the components 50a, 50b, and 50c are estimated, and reconstruction is performed on the estimated component amounts to generate three component maps 51a to 51c. Note that the component 50c in FIG. 2 is different from the component 50c in FIG. 3. In FIG. 2, the component 50c is a noise component, whereas in FIG. 3, the component 50c is an unknown component contained in the living body (inside the subject) but not represented by a physical model. Furthermore, the components 50a and 50b are components that can be represented in the physical model used to estimate the component amounts (components assumed in the physical model). Furthermore, the component 50c is a component that is taken into account in the physical model used to estimate the component amounts.

[0026] As shown in Figures 2 and 3, components 50a and 50b, which are contained in a living organism and can be represented by a physical model, have morphological characteristics in component maps 51a and 51b. However, as shown in Figure 2, component 50c, which is a noise component considered in the physical model but is instead noise, does not have morphological characteristics in component map 51c. On the other hand, as shown in Figure 3, component 50c, which is an unknown component contained in a living organism and considered in the physical model, does have morphological characteristics in component map 51c. In other words, because unknown component 50c exists in a living organism, it has spatial characteristics such as being distributed locally or along blood vessels.

[0027] Therefore, as explained with reference to Figures 2 and 3, based on the fact that noise components and unknown components contained in a living body have different characteristics, the medical information processing device 100 according to this embodiment distinguishes between "noise components" and "unknown components contained in a living body but which cannot be expressed by a physical model" in component estimation using a physical model that takes noise components and unknown components into consideration, and mainly estimates and images "unknown components contained in a living body but which cannot be expressed by a physical model" among these components, as will be explained below.

[0028] Next, an example of processing executed by the acquisition function 105a, biological component estimation function 105b, physical model evaluation function 105c, morphological feature evaluation function 105d, display image generation function 105e, and display control function 105f will be described.

[0029] The acquisition function 105a acquires a CT image 60 (see FIG. 7) stored in the memory 104. The CT image 60 is used in various processes described below.

[0030] The in vivo component estimation function 105b estimates the component amounts in the living body using a physical model that takes into account noise components and unknown components. For example, the in vivo component estimation function 105b estimates the component amounts at each position in the living body using the physical model (a physical model that takes into account noise components and unknown components) shown in the following equation (1).

[0031]

number

[0032] In formula (1), "l" is a value indicating the position of the detector elements arranged in the detector of the X-ray CT device. "j" is a value indicating the wavelength band (channel) of the X-ray. "E" is the energy of the X-ray. "λ l j (a) is the expected value of the number of X-ray photons in the wavelength band indicated by "j" that are incident on the detector element at the position indicated by "l". j" is a weight (coefficient) corresponding to the X-ray of the wavelength band indicated by "j". "m" is a value indicating the type of component. "a m " is the component amount of the component of the type indicated by "m". "τ m " is a coefficient corresponding to the type of component indicated by "m." "ε" is the component amount of the noise component and the unknown component.

[0033] A specific method for estimating the amount of a component in a living body will be described. l j (a) is calculated in advance, and the biological component estimation function 105b calculates "λ l j (a)" and calculate the expected value of the number of X-ray photons, λ, according to the following equation (2): j By fitting the Poisson distribution to (a), the observed number of photons y j get.

[0034]

number

[0035] In addition, in equation (2), "y j " is the number of observed X-ray photons in the band of wavelengths indicated by "j".

[0036] The biological component estimation function 105b calculates the objective function L shown in the following equation (3): p By minimizing the value of (a, ε), the objective function L p The component amount a when the value of (a, ε) is minimized m is estimated as the amount of the component in the living body. m is a vector that summarises a=(a1, a2, …,a M In this embodiment, the biological component estimation function 105b uses likelihood estimation / gradient descent as an optimization method to calculate the component amount a m However, the optimization method is not limited to this. For example, the in vivo component estimation function 105b estimates the component amount a using Bayesian optimization or a genetic algorithm. mmay be estimated.

[0037]

number

[0038] Here, equation (3) represents the objective function obtained from equation (2). In equation (3), L p (a, ε) is the component amount a for the physical model shown in Eq. (1). m is the likelihood (degree of likelihood).

[0039] The biological component estimation function 105b calculates the l j (a, ε)" and component amount a m The component amount ε of the noise component and the unknown component is calculated from equation (1) using the above equation, and the component amount ε of the noise component and the unknown component is estimated.

[0040] Note that "estimating various component amounts" is synonymous with "decomposing a plurality of component amounts into various component amounts." In the following description, the expression "decomposing a plurality of component amounts into various component amounts" may be used instead of the expression "estimating various component amounts."

[0041] The physical model evaluation function 105c calculates an evaluation value based on the physical model shown in formula (1). For example, the physical model evaluation function 105c uses formula (3) to calculate the likelihood L of the component amount a for the physical model shown in formula (1). p (a, ε) is calculated as the evaluation value.

[0042] The morphological feature evaluation function 105d calculates an evaluation value based on the morphological features. In this embodiment, the morphological feature evaluation function 105d calculates an evaluation value related to components in the living body, as well as evaluation values ​​related to noise components and unknown components. A specific example will be described below with reference to FIGS. 4 and 5. FIGS. 4 and 5 are diagrams for explaining an example of processing executed by the morphological feature evaluation function 105d according to the first embodiment. Here, a case will be described in which component amounts a1 and a2 are estimated by the living body component estimation function 105b. Note that the component amount a1 is the component amount of iodine, and the component amount a2 is the component amount of bone. As shown in FIG. 4, the morphological feature evaluation function 105d performs reconstruction on the component amount a1 to generate a component map 61a. Similarly, the morphological feature evaluation function 105d performs reconstruction on the component amount a2 to generate a component map 61b, and performs reconstruction on the component amount ε to generate a component map 61c. In this way, the morphological feature evaluation function 105d visualizes each of the component amounts a1, a2, and ε to generate the component maps 61a to 61c.

[0043] Then, morphological feature evaluation function 105d extracts morphological features for each component map. Morphological features that are less susceptible to noise and have additive characteristics are adopted as morphological features. For example, morphological feature evaluation function 105d calculates higher-order local auto-correlation (HLAC) features or cubic higher-order local auto-correlation (CHLAC) features as morphological features from each of component maps 61a to 61c. For example, morphological feature evaluation function 105d calculates the HLAC features or CHLAC features using the following equation (4):

[0044]

number

[0045] The number of types of elements (mask patterns) for correlation calculation is, for example, 35 types for zero-order features, first-order features, and second-order features in a 3x3 pixel range. HLAC features and CHLAC features are less susceptible to noise because the autocorrelation of noise is 0. Furthermore, HLAC features and CHLAC features have the additivity of the object.

[0046] Morphological feature evaluation function 105d calculates HLAC features or CHLAC features for each element in each of component maps 61a to 61c. As a result, morphological feature evaluation function 105d generates a feature vector composed of HLAC features or CHLAC features for each element in each component map. That is, morphological feature evaluation function 105d generates feature vector 62a composed of HLAC features or CHLAC features for each element from component map 61a. Similarly, morphological feature evaluation function 105d generates feature vector 62b from component map 61b and feature vector 62c from component map 61c.

[0047] Then, morphological feature evaluation function 105d calculates the sum of all feature vectors 62a to 62c of all component maps 61a to 61c. In this way, morphological feature evaluation function 105d calculates the sum 63 of feature vectors 62a to 62c by calculating the sum of feature vector 62a, feature vector 62b, and feature vector 62c.

[0048] 5, the morphological feature evaluation function 105d calculates the HLAC feature amount or the CHLAC feature amount for each element of the CT image (original image) 60 to calculate a feature vector 64. The morphological feature evaluation function 105d then compares the sum 63 of the feature vectors 62a to 62c with the feature vector 64, and calculates an error L between the sum 63 of the feature vectors 62a to 62c and the feature vector 64 using the following equation (5): m is defined as the objective function.

[0049]

number

[0050] In formula (5), "d" indicates an element, "D" indicates the number of all elements, and "f d " indicates the HLAC feature or CHLAC feature of the element indicated by "d" in the sum 63 of the feature vectors 62a to 62c, and "f ∧ d " indicates the HLAC feature or CHLAC feature of the element indicated by "d" in the feature vector 64.

[0051] The morphological feature evaluation function 105d may calculate, as the error, the Euclidean distance (L2 norm) or the L1 norm between the sum 63 of the feature vectors 62a to 62c and the feature vector 64. The morphological feature evaluation function 105d may also calculate, as the error, an optimal transportation distance that can take into account the shape of the distribution.

[0052] Here, if the estimated component has a morphological feature in the component map, the error is relatively small. On the other hand, if the estimated component does not have a morphological feature in the component map and appears as noise, the HLAC feature or CHLAC feature approaches 0, resulting in a relatively large error.

[0053] Although the morphological feature evaluation function 105d has been described as calculating an evaluation value by comparing the sum 63 of the feature vectors 62a to 62c with the feature vector 64, the morphological feature evaluation function 105d may calculate the evaluation value from the sum 63 of the feature vectors 62a to 62c without performing such a comparison. An example of a method for calculating an evaluation value from the sum 63 of the feature vectors 62a to 62c will be described with reference to FIG. 6.

[0054] 6 is a diagram illustrating another example of processing executed by the morphological feature evaluation function 105d according to the first embodiment. An element having a HLAC feature or CHLAC feature greater than 0 in the sum 63 of feature vectors 62a to 62c, which are morphological features, is considered to reflect a morphological feature. This is because the HLAC feature or CHLAC feature approaches 0 for noise. Therefore, as shown in FIG. 6, the morphological feature evaluation function 105d may calculate, as an evaluation value, the number of elements in the sum 63 of feature vectors 62a to 62c whose HLAC feature or CHLAC feature is equal to or greater than a threshold value 63a.

[0055] Furthermore, the morphological feature evaluation function 105d may calculate the degree (likelihood) that the sum 63 of the feature vectors 62a to 62c fits a predetermined distribution (for example, a mixed normal distribution) as an evaluation value.

[0056] The evaluation value L calculated by the physical model evaluation function 105c p (a, ε) and the evaluation value L calculated by the morphological feature evaluation function 105d. m (a, ε) is used when the biological component estimation function 105b updates the component amount a1, the component amount a2, and the component amount ε.

[0057] In the first embodiment, the calculation of the evaluation value by the physical model evaluation function 105c, the calculation of the evaluation value by the morphological feature evaluation function 105d, and the updating of the component amounts a1, a2, and ε by the biological component estimation function 105b are repeatedly executed until a predetermined condition is satisfied. Then, when the predetermined condition is satisfied, the final component amounts a1, a2, and ε are determined. The component amount ε determined in this manner is a component amount in which noise components are suppressed and unknown biological components are the main components. Therefore, according to the first embodiment, of the "noise components" and the "unknown components contained in a biological body but not represented by a physical model," it is possible to estimate the "unknown components contained in a biological body but not represented by a physical model."

[0058] The display image generating function 105e generates a display image. FIG. 7 is a diagram illustrating an example of a display image according to the first embodiment. For example, the display image generating function 105e reconstructs the final component amounts a1, a2, and ε to generate component maps 66a-66c. The display image generating function 105e then assigns a color to each of the component maps 66a-66c, the color intensity of which corresponds to the component amount, with a different color for each component type. The display image generating function 105e then generates a display image by superimposing each of the colored component maps 66a-66c on the CT image 60. In FIG. 7, the component map 66c is an image in which noise components are suppressed and based on the component amount ε of components primarily consisting of unknown components in the living body. Therefore, according to the first embodiment, of the “noise components” and the “unknown components contained in the living body but not represented by a physical model,” it is possible to primarily visualize the “unknown components contained in the living body but not represented by a physical model.”

[0059] The display control function 105f causes various images to be displayed on the display 103. For example, the display control function 105f causes the display 103 to display the image for display shown in FIG. 7 (CT image 60 on which each of colored component maps 66a to 66c is superimposed).

[0060] 8 is a flowchart showing an example of the process executed by the medical information processing apparatus 100 according to the first embodiment. As shown in FIG. 8, the in vivo component estimation function 105b sets a randomly generated (determined) value or an arbitrary value as an initial value to the in vivo component amount a m and set as the component amount ε of the unknown component (step S101).

[0061] The physical model evaluation function 105c then calculates an evaluation value L based on the physical model shown in equation (1). p (a, ε) is calculated (step S102). In parallel with the process of step S102, the morphological feature evaluation function 105d calculates an evaluation value L based on the morphological features. m (a, ε) is calculated (step S103).

[0062] The biological component estimation function 105b calculates the evaluation value L based on the physical model. p (a, ε) and the evaluation value L based on morphological features m The evaluation value L(a, ε) is calculated by integrating (a, ε), and the amount of the component a in the body is calculated based on the calculated evaluation value L(a, ε) using an optimization method such as gradient descent or Bayesian optimization. m and update the component amount ε of the unknown component (step S104). For example, the in vivo component estimation function 105b updates the in vivo component amount a so that the evaluation value L(a, ε) becomes smaller. m and the amount of unknown component ε is corrected to obtain the amount of component a in the living body. m and the component amount ε of the unknown component. For example, in the gradient descent method, the gradient direction is calculated based on the differentiation of the variables a and ε of the evaluation value (evaluation function) L(a, ε), and the variables a and ε are updated based on the gradient direction so that L(a, ε) becomes smaller. Since the evaluation value L(a, ε) calculated by the sum of equations (3) and (5) is differentiable with respect to the variables a and ε, the biological component estimation function 105b can update the variables a and ε using the gradient descent method. Note that even if the evaluation value L(a, ε) is not differentiable with respect to the variables a and ε, the biological component estimation function 105b can estimate it using an optimization method such as Bayesian optimization. Note that the integrated evaluation value can be calculated, for example, as follows: L(a, ε) = L p (a, ε)+L m (a, ε). If you want to emphasize one of the two, such as when placing importance on the evaluation value based on a physical model, you can use L(a, ε) = w p L p (a, ε)+ w m L m The weight w as (a, ε) p ,w m may be used for weighting.

[0063] The biological component estimation function 105b determines whether a predetermined condition is met (step S105). For example, the predetermined condition is that the processing of steps S102, S103, and S104 is repeated a predetermined number of times. In this case, the biological component estimation function 105b determines whether the processing of steps S102, S103, and S104 has been repeated a predetermined number of times. If these processing have been repeated the predetermined number of times (step S105: Yes), the biological component estimation function 105b proceeds to step S106. On the other hand, if these processing have not been repeated the predetermined number of times (step S105: No), the biological component estimation function 105b proceeds to steps S102 and S103. Then, returning to steps S102 and S103, in step S102, the physical model evaluation function 105c evaluates the updated component amount a m and the evaluation value L based on the component amount ε p In step S103, the morphological feature evaluation function 105d calculates the updated component amount a m and the evaluation value L based on the component amount ε m Calculate (a, ε).

[0064] The predetermined condition may be that the evaluation value L(a, ε) calculated in step S104 is equal to or less than a threshold. In this case, the biological component estimation function 105b determines whether the evaluation value L(a, ε) calculated in step S104 is equal to or less than the threshold, and if the evaluation value L(a, ε) is equal to or less than the threshold (step S104: Yes), the biological component estimation function 105b proceeds to step S106. On the other hand, if the evaluation value L(a, ε) is greater than the threshold (step S104: No), the biological component estimation function 105b proceeds to step S105.

[0065] Here, it is assumed that if a substance is depicted in a CT image, the CT image has some morphological characteristics. Therefore, it is desirable that the images representing each component (component images) have morphological characteristics. If component estimation is performed without taking morphological characteristics into account, the component amounts are not updated to reflect the morphological characteristics. Note that the component amount update direction can be either a direction that does not reflect the morphological characteristics or a direction that reflects the morphological characteristics. When component estimation is performed without taking morphological characteristics into account, the update direction is not limited to a direction that reflects the morphological characteristics, so a large amount of calculation is required to generate a component image that reflects the morphological characteristics. For example, the process of step S104 in FIG. 8 must be updated multiple times. Furthermore, when component estimation is performed without taking morphological characteristics into account, the morphological characteristics may not be reflected because the update is performed without considering the morphological characteristics.

[0066] On the other hand, in the medical image processing apparatus 100 according to the first embodiment, the evaluation value L m By using the evaluation value L(a, ε) that takes (a, ε) into consideration, the update direction of the component amount is limited to an update direction that reflects the morphological features. Therefore, the medical image processing apparatus 100 according to the first embodiment can estimate a component image that reflects the morphological features with a small number of updates.

[0067] In step S106, the biological component estimation function 105b determines the final component amount. For example, the biological component estimation function 105b determines the component amount a corresponding to the smallest evaluation value L(a, ε) among the evaluation values ​​L(a, ε) calculated in step S104. m and the component amount ε is the final component amount a m and the component amount ε. To explain this by taking an example, when the processing of step S102, the processing of step S103 and the processing of step S104 are repeated a predetermined number of times, a predetermined number of evaluation values ​​L(a, ε) are calculated for the predetermined number of times in step S104. Then, the biological component estimation function 105b determines the component amount a corresponding to the smallest evaluation value L(a, ε) among the predetermined number of evaluation values ​​L(a, ε). mand the component amount ε is the final component amount a m and the component amount ε is determined. Then, the biological component estimation function 105b ends the process shown in FIG.

[0068] 8, the process in which the physical model evaluation function 105c calculates an evaluation value and the process in which the morphological feature evaluation function 105d calculates an evaluation value are executed in parallel. However, the process in which the physical model evaluation function 105c calculates an evaluation value and the process in which the morphological feature evaluation function 105d calculates an evaluation value may be executed at different times. Therefore, such a process will be described with reference to FIG. 9.

[0069] 9 is a flowchart showing another example of the flow of processing executed by the medical information processing apparatus 100 according to the first embodiment. As shown in FIG. 9, the biological component estimation function 105b sets a randomly generated (determined) value or an arbitrary value as an initial value to the biological component amount a m and the component amount ε of the unknown component is set (step S201).

[0070] The physical model evaluation function 105c then calculates an evaluation value L based on the physical model shown in equation (1). p (a, ε) is calculated (step S202).

[0071] Then, the biological component estimation function 105b calculates the evaluation value L calculated in step S202. p Based on (a, ε), the component amount a m and the component amount ε (step S203). In step S203, the biological component estimation function 105b updates the evaluation value L p The component amount a is set so that (a, ε) becomes small. m and the component amount ε is updated.

[0072] Then, the morphological feature evaluation function 105d calculates the component amount a updated in step S203. m and the component amount ε are used to calculate the evaluation value L based on the morphological features. m(a, ε) is calculated (step S204).

[0073] Then, the biological component estimation function 105b calculates the evaluation value L m Based on (a, ε), the component amount a m and the component amount ε (step S205). In step S205, the biological component estimation function 105b updates the evaluation value L using an optimization method such as gradient descent or Bayesian optimization. m The component amount a is set so that (a, ε) becomes small. m and the component amount ε is updated.

[0074] Then, the biological component estimation function 105b determines whether a predetermined condition is met (step S206). For example, the predetermined condition is that the processes of steps S202 to S205 are repeated a predetermined number of times. In this case, the biological component estimation function 105b determines whether the processes of steps S202 to S205 have been repeated a predetermined number of times, and if these processes have been repeated the predetermined number of times (step S206: Yes), the biological component estimation function 105b proceeds to step S207. On the other hand, if these processes have not been repeated the predetermined number of times (step S206: No), the physical model evaluation function 105c returns to step S202.

[0075] The predetermined condition is the evaluation value L calculated in step S202. p (a, ε) and the evaluation value L calculated in step S204 m Alternatively, the condition may be that the evaluation value L(a, ε) calculated by integrating (a, ε) is equal to or less than a threshold value. In this case, in step S206, the biological component estimation function 105b calculates the evaluation value L p (a, ε) and the evaluation value L mThe physical model evaluation function 105c calculates an evaluation value L(a, ε) by integrating (a, ε), determines whether the calculated evaluation value L(a, ε) is equal to or less than a threshold, and if the evaluation value L(a, ε) is equal to or less than the threshold (step S206: Yes), the biological component estimation function 105b proceeds to step S207. On the other hand, if the evaluation value L(a, ε) is greater than the threshold (step S206: No), the physical model evaluation function 105c proceeds to step S202.

[0076] The predetermined condition may also be that the error calculated as the evaluation value in step S204 is equal to or less than a threshold. In this case, the biological component estimation function 105b determines whether the error calculated as the evaluation value in step S204 is equal to or less than the threshold, and if the error is equal to or less than the threshold (step S206: Yes), the biological component estimation function 105b proceeds to step S207. On the other hand, if the error is greater than the threshold (step S206: No), the physical model evaluation function 105c returns to step S202.

[0077] In step S202, the physical model evaluation function 105c returns to step S205. m and the component amount ε are used to calculate an evaluation value based on the physical model shown in equation (1).

[0078] In step S207, the biological component estimation function 105b determines the final component amount. For example, the biological component estimation function 105b determines the component amount a corresponding to the smallest evaluation value L(a, ε) among the evaluation values ​​L(a, ε) calculated in step S206. m and the component amount ε is the final component amount a m and the component amount ε. To explain this by taking an example, when the processing of steps S202 to S205 is repeated a predetermined number of times, a predetermined number of errors for the predetermined number of times are calculated as evaluation values ​​in step S204. Then, the biological component estimation function 105b determines the component amount a corresponding to the smallest error among the predetermined number of errors. m and the component amount ε is the final component amount a mand the component amount ε is determined. Then, the biological component estimation function 105b ends the process shown in FIG.

[0079] The medical image processing device 100 according to the first embodiment has been described above. The medical image processing device 100 includes an acquisition function 105a for acquiring medical images depicting a living body; an in vivo component estimation function 105b for estimating the component amounts of in vivo components and the component amounts of noise components and unknown components using a physical model that takes noise components and unknown components into account; a physical model evaluation function 105c for calculating the likelihood of the component amounts relative to the physical model as a first evaluation value; and a morphological feature evaluation function 105d for calculating a second evaluation value based on the morphological features of the in vivo components, noise components, and unknown components. Furthermore, the in vivo component estimation function 105b updates the component amounts of in vivo components and the component amounts of noise components and unknown components based on the first evaluation value and the second evaluation value. As described above, the medical image processing device 100 can estimate the "unknown component contained in the living body but not expressed by the physical model" out of the "noise component" and the "unknown component contained in the living body but not expressed by the physical model."

[0080] In addition, in the example shown in FIG. 5, the morphological feature evaluation function 105d calculates a second evaluation value by comparing the morphological features of the components, noise components, and unknown components in the living body with the morphological features of the medical image data.

[0081] In the example shown in FIG. 6, the morphological characteristic evaluation function 105d calculates a second evaluation value based on the morphological characteristics of the in-vivo components, noise components, and unknown components.

[0082] The morphological feature evaluation function 105d may calculate a feature other than the HLAC feature and the CHLAC feature, as long as the feature has additive properties. For example, the morphological feature evaluation function 105d may calculate a GLAC (Gradient Local Auto-Correlation) feature or a simple histogram. Furthermore, the morphological feature evaluation function 105d may use a superpixel method to divide the CT image 60 into local regions and extract features for each local region.

[0083] Furthermore, the medical information processing apparatus 100 may switch between the process shown in Fig. 5 and the process shown in Fig. 6. For example, the medical information processing apparatus 100 executes the process shown in Fig. 5 to update the component amounts, and if the error calculated as the evaluation value is larger than a threshold value even after the number of times the component amounts are updated reaches a predetermined value or more, the medical information processing apparatus 100 executes the process shown in Fig. 6 to update the component amounts. That is, the morphological feature evaluation function 105d switches the method of calculating the second evaluation value depending on the conditions.

[0084] (Second embodiment) In the first embodiment, a case where the medical information processing device 100 executes various processes has been described. However, an X-ray CT device, which is a medical image diagnostic device, may execute the same process as the process executed by the medical information processing device 100 to estimate the "unknown component contained in the living body but not expressed by a physical model" out of the "noise component" and the "unknown component contained in the living body but not expressed by a physical model." Therefore, such an embodiment will be described as the second embodiment. In the description of the second embodiment, configurations different from the first embodiment will be mainly described, and a description of configurations similar to those of the first embodiment may be omitted.

[0085] 10 is a diagram showing an example of the configuration of an X-ray CT apparatus 2 according to the second embodiment. The X-ray CT apparatus 2 is, for example, an apparatus capable of performing photon-counting CT.

[0086] For example, as shown in Fig. 10, the X-ray CT apparatus 2 according to this embodiment includes a gantry device 10, a bed device 30, and a console device 40. For convenience of explanation, Fig. 10 shows a plurality of gantry devices 10.

[0087] In this embodiment, the rotation axis of the rotating frame 13 in the non-tilted state or the longitudinal direction of the tabletop 33 of the bed device 30 is defined as the Z-axis direction, the axis perpendicular to the Z-axis direction and horizontal to the floor surface is defined as the X-axis direction, and the axis perpendicular to the Z-axis direction and perpendicular to the floor surface is defined as the Y-axis direction.

[0088] The gantry device 10 is a device that irradiates an object P (such as a patient) with X-rays, detects the X-rays that have passed through the object P, and outputs the detected X-rays to a console device 40. The object P is an example of a living body. The gantry device 10 has an X-ray tube 11, an X-ray detector 12, a rotating frame 13, a control device 15, a wedge 16, an X-ray aperture 17, and an X-ray high-voltage device 14.

[0089] The X-ray tube 11 is a vacuum tube that generates X-rays by irradiating thermoelectrons from a cathode (filament) toward an anode (target) when a high voltage is applied from the X-ray high voltage device 14. For example, the X-ray tube 11 is a rotating anode type X-ray tube that generates X-rays by irradiating a rotating anode with thermoelectrons.

[0090] The wedge 16 is a filter for adjusting the amount of X-rays irradiated from the X-ray tube 11. Specifically, the wedge 16 is a filter that transmits and attenuates the X-rays irradiated from the X-ray tube 11 so that the X-rays irradiated from the X-ray tube 11 to the subject P have a predetermined distribution. For example, the wedge 16 is a filter made by processing aluminum so as to have a predetermined target angle and a predetermined thickness. The wedge 16 is also called a wedge filter or a bow-tie filter.

[0091] The X-ray aperture 17 includes a lead plate or the like for narrowing down the irradiation range of the X-rays transmitted through the wedge 16, and a slit is formed by combining a plurality of lead plates or the like.

[0092] The X-ray detector 12 detects X-rays emitted from the X-ray tube 11 and passing through the subject P. Specifically, the X-ray detector 12 has a plurality of detection elements arranged in a channel direction along an arc centered on the focal point of the X-ray tube 11. For example, the X-ray detector 12 has a structure in which a plurality of detection element rows, each having a plurality of detection elements arranged in the channel direction, are arranged in a row direction (also called a slice direction or row direction). Here, a collimator is installed on the X-ray incident surface side of the X-ray detector 12 to reduce scattered X-rays. The collimator is sometimes called an anti-scatter grid or a post-collimator.

[0093] Each time an X-ray photon is incident on each of the multiple detection elements, the detection elements output a signal capable of measuring the energy value of the X-ray photon. Specifically, the detection elements are composed of multiple electrodes, and each time an X-ray photon is incident on each of the detection elements, the detection elements output an electrical signal corresponding to the incident X-ray. For example, the detection elements may be made of CZT (cadmium zinc telluride: CdZnTe), CdTe (cadmium telluride), Ge (germanium), Si (silicon), or the like. Note that the detection elements may also be made of other types of semiconductor crystals, such as scintillator crystals.

[0094] That is, the X-ray detector 12 is a direct conversion type detector having a semiconductor element as a detection element that converts incident X-rays into an electric signal. Note that the X-ray detector 12 may be an indirect conversion type detector that combines a phosphor that emits light when excited by X-rays and an optical sensor that converts light emitted by the phosphor into an electric signal.

[0095] The X-ray detector 12 also includes a signal processing circuit connected to the above-mentioned multiple detection elements and processing the electrical signals output from each detection element. The signal processing circuit counts the number of X-ray photons incident on the detection elements by performing pulse height discrimination on pulses having heights proportional to the individual charge amounts of the electrical signals output from the detection elements. The signal processing circuit also measures the energy of the counted X-ray photons by performing arithmetic processing based on the magnitude of each charge. The signal processing circuit also performs analog-to-digital (A / D) conversion on the signals from the detection elements, outputting the counted number of X-ray photons as a digital data signal. For example, the signal processing circuit is realized by an application specific integrated circuit (ASIC).

[0096] The X-ray detector 12 also has a DAS (Data Acquisition System) that outputs detection data based on a signal output from the signal processing circuit. The DAS generates detection data based on a signal representing the counting result of the number of X-ray photons output from the X-ray detector 12. Here, the detection data is data indicating the detection result. The detection data is, for example, a sinogram. The sinogram is data arranging the results of a counting process that counts photons incident on each detection element at each position of the X-ray tube 11. Specifically, the sinogram is data arranging the counting results of the number of X-ray photons in a two-dimensional orthogonal coordinate system with the view direction and channel direction as axes. For example, the DAS generates a sinogram for each row in the slice direction of the X-ray detector 12. The DAS then transfers the generated detection data to the console device 40.

[0097] The X-ray high voltage device 14 includes a high-voltage generator having electrical circuits such as a transformer and a rectifier, and having the function of generating a high voltage to be applied to the X-ray tube 11, and an X-ray control device that controls the output voltage according to the X-ray output emitted by the X-ray tube 11. The high-voltage generator may be of a transformer type or an inverter type. The X-ray high voltage device 14 may be provided on the rotating frame 13, which will be described later, or on the fixed frame (not shown) side of the gantry device 10. Here, the fixed frame is a support frame that rotatably supports the rotating frame 13.

[0098] The rotating frame 13 is an annular frame that supports the X-ray tube 11 and the X-ray detector 12 so that they face each other, and rotates the X-ray tube 11 and the X-ray detector 12 using a control device 15, which will be described later. In addition to the X-ray tube 11 and the X-ray detector 12, the rotating frame 13 also supports an X-ray high-voltage device 14.

[0099] Here, the rotating frame 13 is rotatably supported by a non-rotating portion (e.g., a fixed frame, not shown) of the gantry 10. The rotation mechanism includes, for example, a motor that generates a rotational driving force and a bearing that transmits the rotational driving force to the rotating frame 13 to rotate it. The motor is provided, for example, in the non-rotating portion, and the bearing is physically connected to the rotating frame 13 and the motor, and the rotating frame 13 rotates in response to the rotational force of the motor.

[0100] Furthermore, the rotating frame 13 and the non-rotating part are each provided with a non-contact or contact communication circuit, which allows communication between the unit supported by the rotating frame 13 and the non-rotating part or an external device of the gantry 10. For example, when optical communication is used as the non-contact communication method, the detection data generated by the DAS is transmitted by optical communication from a transmitter having a light-emitting diode (LED) provided on the rotating frame 13 to a receiver having a photodiode provided on the non-rotating part of the gantry 10, and is then transferred from the non-rotating part to the console device 40 by the transmitter. Note that, as the communication method, in addition to non-contact data transmission methods such as capacitive coupling and radio wave methods, contact data transmission methods using a slip ring and electrode brushes may also be used.

[0101] The control device 15 includes a processing circuit having a CPU (Central Processing Unit) and the like, and a driving mechanism such as a motor and an actuator. The control device 15 has a function of receiving an input signal from the console device 40 or an input interface 43 attached to the gantry device 10 and controlling the operation of the gantry device 10 and the bed device 30. For example, the control device 15 receives the input signal and controls the rotation of the rotating frame 13, the tilt of the gantry device 10, and the operation of the bed device 30 and the tabletop 33. The control of tilting the gantry device 10 is realized by the control device 15 rotating the rotating frame 13 around an axis parallel to the X-axis direction based on inclination angle (tilt angle) information input via the input interface 43 attached to the gantry device 10. The control device 15 may be provided in the gantry device 10 or the console device 40.

[0102] The bed device 30 is a device on which the subject P, who is the subject of the scan, is placed and moved, and includes a base 31, a bed driving device 32, a top plate 33, and a support frame 34. The base 31 is a housing that supports the support frame 34 so that it can move in the vertical direction. The bed driving device 32 is a motor or actuator that moves the top plate 33, on which the subject P is placed, in the longitudinal direction of the top plate 33. The top plate 33, which is provided on the upper surface of the support frame 34, is a plate on which the subject P is placed. Note that the bed driving device 32 may move the support frame 34 in addition to the top plate 33 in the longitudinal direction of the top plate 33.

[0103] The console device 40 is a device that accepts operations of the X-ray CT apparatus 2 by an operator and reconstructs CT image data using detection data collected by the gantry device 10. The console device 40 has a memory 41, a display 42, an input interface 43, and a processing circuit 44. Note that, although an example in which the console device 40 and the gantry device 10 are separate entities will be described here, the gantry device 10 may include the console device 40 or some of the components of the console device 40.

[0104] The memory 41 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory), a flash memory, a hard disk, an optical disk, etc. The memory 41 stores, for example, projection data and CT image data.

[0105] The display 42 displays various types of information. For example, the display 42 outputs medical images (CT images) generated by the processing circuitry 44, a GUI (Graphical User Interface) for receiving various operations from the operator, and the like. For example, the display 42 is a liquid crystal display or a CRT (Cathode Ray Tube) display. Note that the display 42 may be provided on the gantry device 10. Also, for example, the display 42 may be a desktop type, or may be configured as a tablet terminal or the like capable of wireless communication with the console device 40 main body. The display 42 is an example of a display unit.

[0106] The input interface 43 accepts various input operations from the operator, converts the accepted input operations into electrical signals, and outputs the electrical signals to the processing circuitry 44. For example, the input interface 43 accepts from the operator scan conditions for collecting projection data, reconstruction conditions for reconstructing CT image data, image processing conditions for generating post-processed images from CT images, and the like. For example, the input interface 43 is realized by a mouse, keyboard, trackball, switch, button, joystick, or the like. Note that the input interface 43 may be provided in the gantry device 10, for example. Furthermore, the input interface 43 may be configured by a tablet terminal or the like capable of wireless communication with the console device 40 main body, for example.

[0107] The processing circuitry 44 controls the overall operation of the X-ray CT apparatus 2. When executing various processes, the processing circuitry 44 specifies an address in the memory space of the memory 41 and writes various data, such as data used in the various processes and data generated in the various processes, to the specified address (an area of ​​the memory 41 corresponding to the specified address). Furthermore, when executing various processes, the processing circuitry 44 specifies an address in the memory space of the memory 41 to which the data used in the various processes has been written, reads the data written to the specified address from the memory 41, and executes various processes using the read data.

[0108] For example, the processing circuitry 44 executes a system control function 441, a preprocessing function 442, a reconstruction processing function 443, an image processing function 444, a biological component estimation function 445, a physical model evaluation function 446, and a morphological feature evaluation function 447.

[0109] The system control function 441 controls various functions of the processing circuitry 44 based on input operations received from an operator via the input interface 43. For example, the system control function 441 controls a CT scan executed in the X-ray CT device 2. The system control function 441 also controls the generation and display of CT image data in the console device 40 by controlling a preprocessing function 442, a reconstruction processing function 443, an image processing function 444, an in vivo component estimation function 445, a physical model evaluation function 446, and a morphological feature evaluation function 447. The system control function 441 also has the same functions as the display control function 105f. However, the system control function 441 displays various images and various information on the display 42 instead of the display 103. The system control function 441 is an example of a display control unit.

[0110] The preprocessing function 442 generates projection data by performing preprocessing such as logarithmic conversion processing, offset correction processing, inter-channel sensitivity correction processing, and beam hardening correction on the detection data output from the DAS of the X-ray detector 12. Note that the data before preprocessing (detection data) and the data after preprocessing may be collectively referred to as projection data.

[0111] The reconstruction processing function 443 performs reconstruction processing using a filtered back projection method, an iterative reconstruction method, or the like on the projection data generated by the preprocessing function 442 to generate a CT image (reconstructed image). In this way, the reconstruction processing function 443 collects CT images depicting the region of the subject to be examined. The reconstruction processing function 443 is an example of an acquisition unit.

[0112] The image processing function 444 converts the CT image data generated by the reconstruction processing function 443 into tomographic image data of an arbitrary cross section or three-dimensional image data using a known method based on an input operation received from the operator via the input interface 43. Note that the generation of the three-dimensional image data may be performed directly by the reconstruction processing function 443. The image processing function 444 has the same functions as the display image generation function 105e. The image processing function 444 is an example of a generation unit.

[0113] The biological component estimation function 445 has the same function as the biological component estimation function 105b, the physical model evaluation function 446 has the same function as the physical model evaluation function 105c, and the morphological feature evaluation function 447 has the same function as the morphological feature evaluation function 105d.

[0114] Here, for example, the processing circuitry 44 is realized by a processor. In this case, each processing function possessed by the processing circuitry 44 is stored in the memory 41 in the form of a program executable by a computer. The processing circuitry 44 then reads each program from the memory 41 and executes it to realize the function corresponding to each program. In other words, the processing circuitry 44 in a state in which each program has been read has each processing function shown in the processing circuitry 44 in FIG. 10.

[0115] Although the above-described processing functions are realized by a single processing circuit 44, the processing circuit 44 may be configured by combining multiple independent processors, with each processor executing a program to realize each processing function. Furthermore, the processing functions of the processing circuit 44 may be realized by distributing or integrating them as appropriate across a single or multiple processing circuits. Furthermore, the processing functions of the processing circuit 44 may be realized by a combination of hardware and software, such as circuits. Although the above description illustrates an example in which a single memory 41 stores programs corresponding to each processing function, the present invention is not limited to this. For example, multiple storage circuits may be distributed, and the processing circuit 44 may read and execute corresponding programs from the individual storage circuits.

[0116] The above has described the second embodiment. According to the X-ray CT apparatus 2 according to the second embodiment, similarly to the medical information processing apparatus 100 according to the first embodiment, in component estimation using a physical model that takes noise components and unknown components into consideration, it is possible to estimate "unknown components that are contained in a living body but cannot be expressed by a physical model" out of "noise components" and "unknown components that are contained in a living body but cannot be expressed by a physical model."

[0117] The program executed by the processor may be provided by being pre-installed in a ROM (Read Only Memory), a storage unit, etc. The program may be provided by being stored in a computer-readable storage medium such as a CD (Compact Disk)-ROM, a FD (Flexible Disk), a CD-R (Recordable), or a DVD (Digital Versatile Disk) in a format that can be installed or executed by these devices. The program may also be provided or distributed by being stored on a computer connected to a network such as the Internet and downloaded via the network.

[0118] Furthermore, the components of each device illustrated in the above-described embodiments are merely functional concepts and do not necessarily have to be physically configured as illustrated. In other words, the specific form of distribution and integration of each device is not limited to that illustrated, and all or part of the devices can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc. Furthermore, all or any part of the processing functions performed by each device can be realized by a CPU and a program analyzed and executed by the CPU, or can be realized as hardware using wired logic.

[0119] According to at least one of the embodiments described above, in component estimation using a physical model that takes noise components and unknown components into consideration, it is possible to estimate the "unknown component that is contained in a living body but cannot be expressed by a physical model" out of the "noise component" and the "unknown component that is contained in a living body but cannot be expressed by a physical model."

[0120] Although several embodiments of the present invention 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, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention described in the claims and their equivalents. [Explanation of symbols]

[0121] 100 Medical information processing device 105b Biocomponent Estimation Function 105c Physical model evaluation function 105d Morphological feature evaluation function

Claims

1. an acquisition unit that acquires a medical image depicting a living body; an estimation unit that estimates the component amounts of the components in the living body depicted in the medical image, as well as the component amounts of the noise components and unknown components, using a physical model that takes into account the noise components and unknown components; a first evaluation unit that calculates a likelihood of a component amount for the physical model as a first evaluation value; a second evaluation unit that calculates a second evaluation value based on morphological characteristics of the in-vivo component, the noise component, and the unknown component; an updating unit that updates the component amounts of the in-vivo components and the component amounts of the noise components and the unknown components based on the first evaluation value and the second evaluation value; A medical information processing device comprising:

2. The medical image processing apparatus according to claim 1 , wherein the second evaluation unit switches a method for calculating the second evaluation value depending on a condition.

3. The medical image processing device according to claim 1 , wherein the second evaluation unit calculates the second evaluation value by comparing morphological features of the in-vivo components, the noise components, and the unknown components with morphological features of the medical image.

4. The medical image processing apparatus according to claim 1 , wherein the second evaluation unit calculates the second evaluation value based on morphological characteristics of the in-vivo component, the noise component, and the unknown component.

5. an acquisition unit that acquires medical images depicting a living body; an estimation unit that estimates the component amounts of the components in the living body depicted in the medical image, as well as the component amounts of the noise components and unknown components, using a physical model that takes into account the noise components and unknown components; a first evaluation unit that calculates a likelihood of a component amount for the physical model as a first evaluation value; a second evaluation unit that calculates a second evaluation value based on morphological characteristics of the in-vivo component, the noise component, and the unknown component; an updating unit that updates the component amounts of the in-vivo components and the component amounts of the noise components and the unknown components based on the first evaluation value and the second evaluation value; A medical image diagnostic device comprising:

6. Acquire medical images depicting the living body, using a physical model that takes noise components and unknown components into consideration, estimating the component amounts of the components in the living body depicted in the medical image, as well as the component amounts of the noise components and unknown components; calculating a likelihood of a component amount for the physical model as a first evaluation value; calculating a second evaluation value based on morphological characteristics of the in-vivo component, the noise component, and the unknown component; updating the component amounts of the in-vivo components and the component amounts of the noise components and the unknown components based on the first evaluation value and the second evaluation value; A medical image processing method comprising:

Images