Information processing device, information processing method, and program

Abstract

To provide a technique for acquiring features by a suitable method for each area in an anatomical site of an analyte.SOLUTION: An information processing device includes extraction means for extracting a prescribed area in an anatomical site from an examination image of an analyte, first feature value acquisition means for acquiring a first feature value of the prescribed area associated with movement of the anatomical site, second feature value acquisition means for acquiring a second feature value of the prescribed area associated with the movement, dividing means for dividing the prescribed area, and integration means for integrating the first feature value and the second feature value together based on a result of division by the dividing means.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] In the medical field, patients are photographed using medical imaging devices such as X-ray CT scanners and PET scanners. Here, CT stands for Computed Tomography, and PET stands for Positron Emission Tomography. By observing the medical images in detail, the anatomical structure and functional information of various types of organs (anatomical parts) inside the patient's body can be obtained, and this information can be used for diagnosis and treatment.

[0003] Among various types of organs that constitute the human body, there are some that move relative to the surrounding organs. For example, the lungs move due to respiratory motion, and the heart moves to circulate blood in the body. It is known that even the same organ moves (direction of movement and amount of movement) differently depending on the position inside the organ or on the surface (hereinafter referred to as intra-organ position) due to the structure, the presence or absence of a lesion, etc. Here, there is a demand from users (doctors, etc.) to recognize intra-organ positions having abnormal movement and discover functional disorders and lesions by visualizing the difference in the direction of movement and amount of movement (hereinafter referred to as movement information) depending on the intra-organ position of the target organ from medical images (i.e., visualizing the distribution of the direction and amount of movement). For example, there is a demand to identify adhesion positions on the surface of the lungs from medical images by visualizing the difference in movement information due to respiratory motion of the lungs at different positions on the surface of the lungs.

[0004] Patent Document 1 discloses a technique for calculating the amount of sliding of the surface position due to respiratory movement, which is closely related to adhesions on the lung surface. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2016-67832 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, with conventional techniques, there is a problem in that it is difficult to acquire features for each region in an anatomical part (eg, lungs) of a subject (eg, a human body) using a method suitable for that region.

[0007] The present invention has been made in consideration of the above problems, and has an object to provide a technique for acquiring features using a method appropriate for each region in an anatomical part of a subject. [Means for solving the problem]

[0008] To achieve the above object, an information processing device according to the present invention comprises: An extraction means for extracting a predetermined region of an anatomical part from an examination image of a subject; a first feature amount acquiring means for acquiring a first feature amount related to a movement of the anatomical part in a first region included in the predetermined region; The first feature amount regarding the motion in a second area that is included in the predetermined area and is different from the first area is Types A second feature amount acquiring means for acquiring a different second feature amount; an integration means for integrating the first feature amount and the second feature amount so as to represent the first feature amount in the first region and the second feature amount in the second region; The present invention is characterized by comprising: Effect of the Invention

[0009] According to the present invention, it is possible to acquire features using a method suitable for each region of an anatomical part of a subject. [Brief description of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating an example of a device configuration of an information processing system according to a first embodiment. [Diagram 2] 4 is a flowchart showing an overall processing procedure in the first embodiment. [Diagram 3] FIG. 11 is a diagram for explaining in detail the process of S1020 in the first embodiment. [Figure 4] 13 is a flowchart showing the processing procedure of S1020 in the first embodiment. [Diagram 5] FIG. 11 is a diagram for explaining in detail the process of S1030 in the first embodiment. [Figure 6] 13 is a flowchart showing the processing procedure of S1030 in the first embodiment. [Figure 7] FIG. 11 is a diagram for explaining in detail the process of S1040 in the first embodiment. [Figure 8] FIG. 11 is a diagram illustrating an example of a device configuration of an information processing system according to a second embodiment. [Figure 9] 13 is a flowchart showing an overall processing procedure in a second embodiment. [Figure 10] FIG. 2 is a diagram illustrating an example of a hardware configuration of an information processing device according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, the embodiments will be described in detail with reference to the attached drawings. Note that the following embodiments do not limit the invention according to the claims. Although the embodiments describe a number of features, not all of these features are essential to the invention, and the features may be combined in any manner. Furthermore, in the attached drawings, the same reference numbers are used for the same or similar configurations, and duplicated descriptions are omitted.

[0012] [First embodiment] The information processing system according to the first embodiment of the present invention provides users, such as doctors and engineers at medical institutions, with a function of grasping the state of pleural adhesion of a subject (patient) to be examined and supporting diagnosis. More specifically, the system acquires the state of lung movement and calculates its feature amount by analyzing three-dimensional moving images of the subject's lungs. Then, the system provides a function of generating an observation image that can be easily visually recognized by the user based on the state of lung movement. The information processing system according to this embodiment calculates a feature amount related to slippage calculated from the relative movement amount between the subject's lungs and the surrounding parts, and a feature amount related to movement calculated from the absolute movement amount of the lungs. Then, an observation image is generated by integrating these feature amounts based on the parts of the lungs.

[0013] <Configuration of information processing system> 1 is a diagram showing the overall configuration of an information processing system according to a first embodiment. The information processing system includes an information processing device 10, an examination image database 30, an examination image photographing device 40, and a display device 60, and these devices are communicably connected to each other via a communication unit. In this embodiment, the communication unit is configured as a LAN (Local Area Network) 50, but may be a WAN (Wide Area Network). The communication unit may be connected by wire or wirelessly.

[0014] The examination image database 30 holds a plurality of examination images related to a plurality of patients and their associated information. The examination images are medical images taken by an imaging diagnostic device such as CT or MRI, and may be two-dimensional images, three-dimensional images, or four-dimensional images that are video of three-dimensional images. Here, CT stands for Computed Tomography, and MRI stands for Magnetic Resonance Imaging.

[0015] Moreover, each image may be in various formats such as monochrome or color. The examination image database 30 in this embodiment holds four-dimensional CT (4DCT) data of the subject. The examination image database 30 holds, as additional information of the examination image, the patient name (patient ID), examination date information (date when the examination image was taken), the name of the imaging modality of the examination image, and the like. Each examination image and its additional information is assigned a unique number (examination image ID) to enable identification from others, and the information processing device 10 can read out the information based on the number. The examination image database 30 also holds a map of the standard movement amount and standard sliding amount of the pleura of a normal case (specific examples of the movement amount and sliding amount will be described in detail later). Here, a normal case refers to a case in which the pleura does not have adhesion.

[0016] <Functional configuration of information processing device> The information processing device 10 acquires information held in the inspection image database 30 via a LAN 50. The information processing device 10 includes a contour extraction unit 110, a movement feature amount calculation unit 120, a slippage feature amount calculation unit 130, a contour division unit 140, an integrated map generation unit 150, and a display control unit 160.

[0017] The data acquisition unit 100 acquires the examination images of the subject captured by the examination image capturing device 40 and stored in the examination image database 30, and a map of the standard amount of movement and standard amount of sliding (described in detail later) of the pleura in normal cases.

[0018] The contour extraction unit 110 extracts the contour of the lungs from the examination image of the subject. The motion feature calculation unit 120 analyzes the examination image acquired by the data acquisition unit 100, and calculates a map of the amount of motion of the pleura at the contour extracted by the contour extraction unit 110. The slippage feature calculation unit 130 analyzes the examination image acquired by the data acquisition unit 100, and calculates a map of the amount of slippage of the pleura at the contour extracted by the contour extraction unit 110.

[0019] The contour dividing unit 140 divides the contour extracted by the contour extraction unit 110 into a thoracic-side contour and a longitudinal-side contour. The integrated map generating unit 150 generates an integrated map by integrating the map of the movement amount calculated by the movement feature amount calculating unit 120 and the map of the slippage amount calculated by the slippage feature amount calculating unit 130 based on the result of the contour division by the contour dividing unit 140. The display control unit 160 performs control for displaying the integrated map (integration result) generated by the integrated map generating unit 150 on the display device 60.

[0020] 1 is merely an example. For example, the information processing device 10 may have a storage unit (not shown) and may have the functions of the inspection image database 30. The information processing device 10 may also have a display device 60.

[0021] <Hardware configuration of information processing device> Next, the hardware configuration of the information processing device 10 will be described with reference to Fig. 10. The information processing device 10 includes a CPU 511, a ROM 512, a RAM 513, an auxiliary storage device 514, an operation unit 515, a communication I / F 516, and a bus 517.

[0022] The CPU 511 realizes each function of the information processing device 10 shown in FIG. 1 by controlling the entire information processing device 10 using computer programs and data stored in the ROM 512 and the RAM 513. The information processing device 10 may have one or more dedicated hardware different from the CPU 511, and at least a part of the processing by the CPU 511 may be executed by the dedicated hardware. Examples of the dedicated hardware include an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), and a DSP (Digital Signal Processor). The ROM 512 stores programs that do not require modification. The RAM 513 temporarily stores programs and data supplied from the auxiliary storage device 514, and data supplied from the outside via the communication I / F 516. The auxiliary storage device 514 is composed of, for example, a hard disk drive, and stores various data such as image data and audio data.

[0023] The operation unit 515 is composed of, for example, a keyboard, a mouse, a joystick, a touch panel, and the like, and inputs various instructions to the CPU 511 in response to operations by the user.

[0024] The communication I / F 516 is used for communication with an external device of the information processing device 10. For example, when the information processing device 10 is connected to the external device by wire, a communication cable is connected to the communication I / F 516. When the information processing device 10 has a function of wireless communication with the external device, the communication I / F 516 includes an antenna. The bus 517 connects each unit of the information processing device 100 to transmit information.

[0025] In this embodiment, the operation unit 515 is assumed to exist inside the information processing device 10, but it may exist as a separate device outside the information processing device 10. In this case, the CPU 511 may operate as an operation control unit that controls the operation unit 515.

[0026] <Processing> Next, the overall processing procedure by the information processing device 10 in this embodiment will be described in detail with reference to Fig. 2. In addition, the following will be described as an example of using CT data as the examination image, but the embodiment of the present invention is not limited to this. For example, as long as it is time-series three-dimensional volume data of lungs, it may be an MRI image or an ultrasound image.

[0027] (Step S1000: Acquiring data) In step S1000, the data acquisition unit 100 acquires 4DCT data of the lung field of the subject from the examination image database 30. The 4DCT data in this embodiment is time-series three-dimensional volume data, and is data of the dynamic state of the subject due to breathing. More specifically, the 4DCT data is acquired that is composed of 3DCT data at two points in time, that is, the inhalation position (e.g., maximum inhalation position) and the exhalation position (e.g., maximum exhalation position). In this embodiment, the 3DCT data at the inhalation position is represented as I_ins, and the 3DCT data at the exhalation position is represented as I_exp. The 4DCT data including these is represented as I. It is assumed that the 3DCT data in this embodiment includes the entire lungs of the subject.

[0028] In this embodiment, the case where 3DCT data at two time points, the inhalation position and the exhalation position, is used as an example as described above, but is not limited thereto. If the movement of the lung field due to the breathing of the subject can be captured, 3DCT data at two time points in other respiratory states may be used. For example, an image at the inhalation position and an image at a time phase adjacent thereto may be used. Alternatively, instead of using images at two time phases, images at three or more time phases may be acquired. When images at three or more time phases are acquired to evaluate the movement, for example, a plurality of pairs of two time points may be selected from each time phase, and the movement amount and the slip amount, which will be described later, may be evaluated, and the statistics (average value, median value, maximum value, etc.) of the movement amount and the slip amount obtained from each pair may be used.

[0029] The data acquisition unit 100 also acquires, from the examination image database 30, a standard motion amount map, which is a standard motion amount of the pleura of a normal case without pleural adhesion, and a standard slip amount map, which is a standard slip amount. In this embodiment, the standard motion amount map and the standard slip amount map are acquired as volume data discretized to the same extent as the 3DCT data of the subject. In this embodiment, the standard motion amount map is represented as Rm(x) and the standard slip amount map is represented as Rs(x). Rm(x) and Rs(x) are functions that take a coordinate value x representing a position in the map as an argument and return the standard motion amount and standard slip amount of the position, respectively.

[0030] (Step S1010: Extraction of lung contour) In step S1010, the contour extraction unit 110 analyzes the 3DCT data I_ins of the subject's inhalation position acquired in step S1000, and executes a process of extracting the lung contour. This process can be executed by a segmentation method known as medical image processing. As a result of this processing step, the contour extraction unit 110 acquires a lung contour mask L(x). Here, the lung contour mask L(x) is a function that takes a coordinate value x representing a position in the 3DCT data I_ins as an argument, and returns 1 if the position is the lung contour, and 0 otherwise.

[0031] (Step S1020: Analysis of lung contour movement) In step S1020, the motion feature amount calculation unit 120 processes the 4DCT data I to analyze the motion of the lungs of the subject, and calculates and acquires the motion feature amount of the lung contour due to breathing (first feature amount acquisition). In this embodiment, the case of calculating the motion feature amount of the right lung of the subject will be described as an example. However, this is not limited to this, and the left lung or both the left and right lungs may be targeted. The movement (movement) of the lungs due to breathing will be described with reference to FIG. 3. FIG. 3 is a diagram showing a coronal section of the lungs at the inspiration position and expiration position. In the same figure, 200 represents the contour shape of the lungs at the inspiration position. Also, 202 represents the contour shape of the lungs at the expiration position. In this way, the contour shapes of the lungs are different between the inspiration position and the expiration position. The arrows 210 in the figure represent the movement (movement) of the lungs from the inspiration position to the expiration position due to breathing at each position of the lung contour. In this processing step, a movement feature amount at each position of the lung contour is calculated as a feature amount representing the movement (motion) of the lungs at each position of this lung contour.

[0032] Here, the detailed processing procedure of this processing step will be described in more detail with reference to the flowchart of FIG.

[0033] (Step S10202: Image registration within the lung field) In step S10202, the motion feature amount calculation unit 120 executes registration processing within the lung field between the inspiration image I_ins and the expiration image I_exp acquired in step S1000. This acquires a displacement vector field V1(x) from each position of the inspiration image I_ins to the corresponding position of the expiration image I_exp. More specifically, registration is performed to capture the movement of each position occurring within the lung field of the subject from inspiration to expiration. This registration processing may be performed by any known registration processing between images, but it is preferable to perform deformation registration that can capture the deformation of the target.

[0034] In addition, in order to capture the movement of the subject in the lung field more accurately, an image in which image features in the lung field are emphasized or image features outside the lung field are weakened may be generated, and the image may be used for alignment. For example, an image in which pixel values ​​outside the lung field are masked based on the lung contour mask L(x) extracted in step S1010 may be generated and used for alignment. Alternatively, an image may be generated that has been window-transformed (pixel value transformation that saturates pixel values ​​outside a predetermined pixel value range) so that the distribution of pixel values ​​in the lung field is emphasized based on the distribution of pixel values ​​in and outside the lung field, and the image may be used for alignment.

[0035] By the above-described registration process, a displacement vector field V1(x) in the lung field is obtained from each position of the inspiration image I_ins to the corresponding position of the expiration image I_exp. Here, V1(x) is a function that takes the coordinate value x on the inspiration image as an argument and returns a three-dimensional vector that represents the displacement of each position of the subject's lung from the inspiration position to the expiration position. More specifically, it is volume data that holds the displacement amount of each axis at each position discretized to the same extent as 3DCT data.

[0036] In the above description, an example has been described in which an image at an inhalation position and an image at an exhalation position are used as the object of deformation alignment, but the present invention is not limited to this, and any image at a time phase may be used as long as the movement of each position of the lung field due to breathing can be acquired. For example, an image at an inhalation position and an image at an adjacent time phase may be used. Alternatively, images at three or more time phases may be acquired in step S1000, and the images may be aligned in a daisy chain to acquire the movement of each position of the lung field due to breathing. This is also true for the alignment performed as the process of step S1030 described later.

[0037] (Step S10204: Calculate the amount of motion) In step S10204, the motion feature amount calculation unit 120 executes a process of calculating the amount of motion (scalar value) on the lung contour based on the displacement vector field V1(x) calculated in step S10202 and the lung contour L(x) calculated in step S1010. Specifically, the norm of V1(x) is calculated as the amount of motion. In this embodiment, the case where the amount of motion is calculated at a predetermined interval (for example, 1 mm) over the entire lung contour of the subject will be described as an example.

[0038] In this embodiment, the position on the contour for which the amount of motion is calculated is represented as P1_i (1≦i≦N), and the calculated amount of motion is represented as M_i (1≦i≦N). Here, i is an index for identifying multiple positions on the contour, and N is the total number of positions on the contour. In this embodiment, N amounts of motion M_i are stored as a motion amount map M(x). The motion amount map M(x) is a function that returns the amount of motion at a position x in the image coordinate system of the 3DCT data at the inspiration position as an argument. More specifically, it is stored as volume data discretized to the same extent as the 3DCT data. Note that the motion amount map M(x) returns an invalid value (for example, an identifiable value such as a negative value) when a position other than the contour is used as an argument.

[0039] (Step S10206: Calculation of movement feature amount (difference from standard movement amount)) In step S10206, the motion feature amount calculation unit 120 calculates the motion feature amount at each position of the lung contour as information representing the difference (deviation) between the motion amount at each position of the lung contour calculated in step S10204 and the standard motion amount at each position. In this embodiment, the motion feature amount is stored as a motion feature amount map M'(x). Specifically, it can be calculated as, for example, M'(x)=M(x) / Rm(d(x)). Here, d(x) is a displacement function that corrects the positional relationship between the standard motion amount map Rm(x) and the motion amount map M(x). Specifically, it is a correction function that matches the lung contour, the apical position, the basal position, and the like between the two maps. d(x) can be obtained, for example, by using the difference between the shapes of the lung contour as a cost function, and performing rigid body registration or affine registration that minimizes the cost function, or by performing any nonlinear registration using FFE (Free Form Deformation), etc. This allows the calculation of a motion feature map M'(x) that indicates the relative relationship between the amount of motion of the subject's lungs and the standard amount of motion of a normal subject without pleural adhesions.

[0040] In the above description, the motion feature map M'(x) is calculated based on the ratio between the motion amount map M(x) and the standard motion amount map Rm(x), but the present invention is not limited to this. For example, the motion feature map may be calculated by subtracting the motion amount map M(x) from the standard motion amount map Rm(x) M'(x)=M(x)-Rm(d(x)). That is, any method may be used to calculate the motion amount at a certain position as long as it can digitize the difference (deviation) between the motion amount at that position and the standard motion amount. Also, the motion amount itself may be used as the motion feature, instead of the difference (deviation) from the standard motion amount. That is, M'(x)=M(x). In this case, the process of acquiring the standard motion amount map Rm(x) is not required in step S1000.

[0041] The above process calculates the motion feature amount at each position of the lung contour in the inhalation position of the subject. In general, in the case of a subject with pleural adhesion, the motion feature amount tends to be small at the adhesion site. The calculation process of the motion feature amount by the above method can be stably performed even if image features (e.g. pixel value patterns such as edges and corners) are scarce on the outside of the lung field of the subject, as long as there are image features on the inside of the lung field. Therefore, in the vicinity of the lung field contour adjacent to the longitudinal side where image features are relatively scarce, a feature amount that more accurately captures the presence or absence of adhesion can be calculated compared to the slippage feature amount described later. The calculation of the motion feature amount at each position of the lung contour may be performed by any method other than the above method.

[0042] (Step S1030: Lung contour sliding analysis) Returning to FIG. 2, in step S1030, the slippage feature amount calculation unit 130 processes the 4DCT data I to analyze the motion of the contour of the lungs of the subject, and calculates and acquires the slippage amount due to breathing (second feature amount acquisition). In this embodiment, the case of calculating the slippage amount for the right lung of the subject will be described as an example. The slippage amount at the contour of the lungs due to breathing will be described with reference to FIG. 5.

[0043] FIG. 5 is a diagram showing a coronal section of the lung at the inhalation and exhalation positions, similar to FIG. 3. In FIG. 5, 200, 202, and 210 respectively represent the lung contour shape at the inhalation position, the lung contour shape at the exhalation position, and the movement of the lung from the inhalation position to the exhalation position due to breathing at each position of the lung contour, similar to FIG. 3. Also, arrow 212 represents the movement of the chest wall from the inhalation position to the exhalation position due to breathing at each position of the lung contour. As shown by the direction and size of arrows 210 and 212 in the figure, at each position of the lung contour, a movement accompanied by slip occurs at the position of the pleura between the lung field side and the chest wall side due to breathing. In this processing step, a slip feature amount at each position of the lung contour is calculated as a feature amount representing the magnitude of slip at each position of the lung contour.

[0044] The detailed processing procedure of this processing step will be described in more detail with reference to the flowchart of FIG.

[0045] (Step S10302: Image registration within the lung field) In step S10302, the slippage feature calculation unit 130 executes a process of aligning the positions of the inhalation image I_ins and the exhalation image I_exp acquired in step S1000 within the lung field. As a result, a displacement vector field V21(x) is acquired from each position x of the inhalation image I_ins to the corresponding position of the exhalation image I_exp. This process is the same as the process of step S10202 described above, and a detailed description is omitted here. Also, instead of executing this process step, the process result of step S10202 described above may be acquired. That is, V21(x)=V1(x) may be set. By the method described above, a displacement vector field V21(x) is acquired from each position x of the inhalation image I_ins to the corresponding position of the exhalation image I_exp.

[0046] (Step S10304: Image registration of lung field) In step S10304, the slippage feature calculation unit 130 executes a registration process for the outside of the lung field between the inspiration image I_ins and the expiration image I_exp acquired in step S1000. As a result, a displacement vector field V22(x) is acquired from each position x of the inspiration image I_ins to the corresponding position of the expiration image I_exp. More specifically, a registration process is performed to capture the movement of each position outside the lung field of the subject from inspiration to expiration. This registration process may be performed by any known registration process between images, but it is preferable to perform a deformation registration process that can capture the deformation of the subject. In addition, in order to capture the movement outside the lung field of the subject more accurately, an image in which the image features of the outside of the lung field are emphasized or, conversely, the image features of the inside of the lung field are weakened may be generated, and the registration may be performed using the image. For example, an image in which pixel values ​​in the lung field are masked based on the lung contour mask L(x) extracted in step S1010 may be generated and used for registration. Alternatively, an image may be generated that has been window-converted based on the distribution of pixel values ​​inside and outside the lung field so that the distribution of pixel values ​​outside the lung field is emphasized, and the image may be used for alignment.

[0047] By the above registration process, a displacement vector field V22(x) outside the lung field is obtained from each position on the inspiration image I_ins to the corresponding position on the expiration image I_exp. Here, V22(x) is a function that takes the coordinate value x on the inspiration image as an argument and returns a three-dimensional vector that represents the displacement of the position from inspiration to expiration. More specifically, the displacement amount for each axis is held as volume data discretized to the same extent as 3DCT data.

[0048] (Step S10306: Calculate slippage) In step S10306, the slippage feature calculation unit 130 executes a process of calculating a slippage on the lung contour based on the displacement vector fields V21(x), V22(x) acquired in steps S10302 and S10304 and the lung contour L(x) calculated in step S1010. Here, the slippage is specifically the norm of the difference vector V21(x)-V22(x). In this embodiment, a case where the slippage is calculated at a predetermined interval (for example, 1 mm) over the entire lung contour of the subject will be described as an example.

[0049] In this embodiment, the position on the contour where the slippage is calculated is represented as P2_i (1≦i≦N), and the calculated slippage is represented as S_i (1≦i≦N). Here, i is an index for identifying multiple positions on the contour, and N is the total number of positions on the contour. In this embodiment, N slippages S_i are stored as a slippage map S(x). The slippage map S(x) is a function that returns the slippage at a position x in the image coordinate system of the 3DCT data at the inspiration position as an argument. More specifically, it is stored as volume data discretized to the same extent as the 3DCT data. Note that, like the motion map M(x), the slippage map S(x) returns an invalid value (for example, an identifiable value such as a negative value) when a position other than the contour is used as an argument.

[0050] (Step S10308: Calculate slippage feature amount (difference from standard slippage amount)) In step S10308, the slippage feature calculation unit 130 calculates the slippage feature at each position of the lung contour as information representing the difference (deviation) between the slippage at each position of the lung contour calculated in step S10306 and the standard slippage at each position. In this embodiment, the slippage feature is stored as a slippage feature map S'(x). Specifically, for example, it can be calculated as S'(x)=S(x) / Rs(d(x)). Here, d(x) is the correction function described in the process of step S10206. In this way, the slippage feature map S'(x) representing the relative relationship between the slippage of the subject's lung and the standard slippage of a normal case without adhesion to the pleura is calculated.

[0051] The calculation method of the slippage feature map S'(x) is not limited to the above method. For example, similar to the calculation method of the movement feature map M'(x) described in step S10206, the slippage feature map S'(x) may be calculated by calculating the difference between the slippage map S(x) and the standard slippage map Rs(x). That is, any method may be used for calculation as long as it can quantify the difference (deviation) between the slippage at a certain position and the standard slippage at that position. Also, the slippage itself may be used as the slippage feature, instead of the difference (deviation) from the standard slippage. That is, S'(x)=S(x) may be used. In this case, the process of acquiring the standard slippage map Rs(x) in step S1000 is not necessary.

[0052] The process of step S1030 is executed by the process of steps S10302 to S10308 described above. As a result, the slip feature amount at each position of the lung contour in the inhalation position of the subject is calculated. In general, in the case of a subject with pleural adhesion, the slip feature amount tends to be small at the adhesion site. In addition, the calculation process of the slip feature amount by the above method calculates the feature amount based on the characteristics of the movement of both the inside and outside of the lung field of the subject, so that it is possible to calculate a feature amount that more accurately captures the presence or absence of adhesion than the above-mentioned movement feature amount. However, since the process cannot be stably performed in a portion where the image features are scarce on the outside of the lung field, the above-mentioned movement feature amount is a feature amount that more accurately captures the presence or absence of adhesion near the lung field contour adjacent to the nasal cavity side where the image features are relatively scarce. The calculation of the slip feature amount at each position of the lung contour may be performed by any method other than the above method. For example, the method disclosed in Patent Document 1 may be used.

[0053] (Step S1040: Segmentation of Lung Contour) 2, in step S1040, the contour dividing unit 140 executes a process of dividing the lung contour L(x) extracted in step S1010 into a longitudinal lung contour and a thoracic lung contour, based on the lung contour L(x). A specific example is shown in FIG.

[0054] FIG. 7 is a diagram showing an axial section at an arbitrary position of the inspiration image I_ins. A body surface contour 400, a right lung contour 410, and a left lung contour 420 are depicted on this axial section. In this process, a process is executed to divide the right lung contour 410 into a longitudinal contour 450 and a thoracic contour 440. A specific example of the process will be described. First, a search is performed in a direction away from the origin point in a plurality of directions as shown by an arrow 460, starting from the midline position 430 of the subject. Then, in each direction, the position that first intersects with the right lung contour 410 is set as the longitudinal contour 450. At this time, if the body surface contour 400 or the left lung contour 420 is intersected without intersecting with the right lung contour 410, the search in that direction may be terminated.

[0055] Then, the position of the right lung contour 410 that is not the longitudinal lung contour 450 is set as the thoracic lung contour 440. The above process is performed on all axial sections of the inhalation image I_ins. Then, based on these results, a segmented lung contour mask L'(x) is generated. More specifically, the segmented lung contour mask L'(x) is a function that returns 1 at the position of the longitudinal lung contour 450, 2 at the position of the thoracic lung contour, and 0 at other positions. In this embodiment, L'(x) is held as volume data discretized to the same extent as the 3DCT data. In the above-described method, the midline position 430 does not necessarily have to be the position of the midline in the anatomical sense of the human body. For example, the center of gravity of the body surface contour 400, or the midpoint between the center of gravity of the right lung contour 410 and the center of gravity of the left lung contour 420, may be calculated and used as the midline position 430.

[0056] The method of calculating the lung contour mask L'(x) after division is not limited to the above method. For example, the right lung contour 410 may be divided into the longitudinal contour 450 and the thoracic contour 440 based on the relationship between the direction of the normal line of the contour at each position (direction from inside the lung field to outside the lung field) and the direction from each position to the midline position 430. Specifically, when the direction of the normal line and the direction toward the midline position 430 are within a predetermined angle, the position is determined as the longitudinal contour 450, and the other positions are determined as the thoracic contour 440. Similarly, any discrimination method based on the geometric difference such as the position and surface orientation of the longitudinal contour and the thoracic contour can be used.

[0057] Alternatively, the longitudinal contour side contour and the thoracic contour side contour may be determined and divided based on the adjacency relationship with other parts around the lung, such as the body surface contour 400 and the heart (not shown). More specifically, at each position of the right lung contour 410, the distance d1 from that position to the body surface contour 400 and the distance d2 from that position to the heart contour are calculated. If d1>d2, the position can be determined as the longitudinal contour side contour 450, and the other positions can be determined as the thoracic contour side contour 440.

[0058] In the above example, the case where the body surface contour 400 and the heart contour are used has been described as an example, but the present invention is not limited to this, and a rib contour (not shown) may be used instead of the body surface contour 400. Also, the position of the center of gravity of the heart region may be used instead of the heart contour.

[0059] (Step S1050: Generate integrated map) In step S1050, the integrated map generating unit 150 executes a process of generating an integrated map F(x) by integrating the motion feature map M'(x) calculated in step S1020 and the slippage feature map S'(x) calculated in step S1030. Here, the integrated map F(x) is a map having the values ​​of the motion feature map M'(x) at the contour position on the longitudinal cage side and the values ​​of the slippage feature map S'(x) at the contour position on the thorax cage side. More specifically, the integrated map F(x) is generated by a conditional substitution operation shown in formula (1).

[0060]

number

[0061] By the above process, a map can be generated in which the feature amount calculated by the method suitable for each position of the lung contour is integrated. In particular, a map can be generated in which the feature amount calculated by the method suitable for each position is integrated based on the adjacent relationship with anatomical parts such as other organs at each position of the lung contour, according to the image feature characteristics of the adjacent anatomical parts, etc. The method of generating the integrated map is not limited to the above method, and for example, a process may be performed to smoothly integrate the motion feature amount map and the slip feature amount map at the boundary position on the contour where the longitudinal side and the thoracic side are adjacent. Specifically, the integrated map may be generated by mixing the values ​​of both maps weighted according to the distance from the boundary position.

[0062] (Step S1060: Display / Save) In step S1060, the display control unit 160 performs control to display the integrated map F(x) generated in step S1050 on the display device 60. Specifically, the display control unit 160 generates an observation image for observing the integrated map F(x) and controls the display device 60 to display the image. The observation image can be generated as a surface rendering image obtained by performing gradation conversion of the integrated map F(x) using a grayscale or color map on the three-dimensional shape of the lung field contour of the subject. At this time, rendering may be performed so that positions where the value of the integrated map F(x) is an invalid value are rendered as transparent pixels and other positions are drawn. The rendering may be configured so that conditions such as the viewpoint position, line of sight direction, scaling, and gradation conversion can be operated by the user. For example, the input of the conditions may be accepted using the operation unit 515. In addition to this, a volume rendering image of the inspiration image I_ins may be generated, and the surface rendering image may be superimposed on the volume rendering image to generate an observation image. In addition, an arbitrary cross-sectional image may be generated from the inhalation image I_ins or 3DCT data of an arbitrary time phase in response to user operation, etc., and an observation image may be generated by superimposing pixel values ​​obtained by performing gradation conversion on the integrated map F(x) using a color map or the like at the position of the lung contour of the cross-sectional image.

[0063] Furthermore, an observation image may be generated so as to display not only the integrated map F(x) but also the motion feature map M'(x) calculated in step S1030 and the slippage feature map S'(x) calculated in step S1030. In this case, an observation image may be generated so as to switch between and display any of the integrated map F(x), the motion feature map M'(x), and the slippage feature map S'(x) based on a user operation. Alternatively, an observation image may be generated in which each map or a portion thereof is arranged side by side or superimposed. Note that the above-mentioned method is merely an example, and any method may be used to generate an observation image.

[0064] In this processing step, the information processing device 10 may further record the integrated map F(x) in the inspection image database 30. In this case, it is not necessary to generate or display the observation image. Similarly, the motion amount map M(x), the motion feature amount map M'(x), the slippage amount map S(x), and the slippage feature amount map S'(x) may be recorded in the inspection image database 30. It is preferable that these pieces of information are stored in association with the 4DCT data I acquired in step S1000. Moreover, the information of each map recorded in the inspection image database 30 may be in a form that can be observed with any viewer or the like after the completion of this processing.

[0065] The information processing device 10 of this embodiment executes processing according to the method described above. In this embodiment, a predetermined region (e.g., lung contour) of an anatomical part (e.g., lung) is extracted from an examination image of a subject, and a first feature amount (e.g., movement feature amount) and a second feature amount (slip feature amount) of the predetermined region related to the movement of the anatomical part are obtained. In addition, the predetermined region is divided, and the first feature amount and the second feature amount are integrated based on the division result.

[0066] This makes it possible to obtain features by a method suitable for each anatomical region of the subject, and therefore it is possible to provide the user with an integrated result of the analysis of features calculated by a method suitable for each region of the subject's lung contour (for example, an integrated map obtained by integrating each feature amount map).

[0067] (Variation 1-1: Method other than dividing by longitudinal / thoracic cage) In the description of step S1040 of this embodiment, the lung contour of the subject is divided into a longitudinal contour and a thoracic contour, but the present invention is not limited to this. For example, the lung contour may be divided into a contour of the lung base (a part of the lung that contacts the diaphragm) and other contours (contours other than the lung base). Or, the lung contour may be divided into a contour of the lung apex (a part of the lung close to the head of the subject) and other contours.

[0068] In addition, this embodiment can be applied not only to the case where the lung contour is divided into two, but also to the case where the lung contour is divided into three or more contours such as the longitudinal contour, the lung base contour, the lung apex contour, and other contours. In this case, in the generation process of the integrated map executed in step S1050, it is desirable to assign a map that more appropriately represents the state of adhesion of the pleura to each divided contour and integrate them. For example, a method can be adopted in which a motion feature map is assigned to the longitudinal contour, the lung base contour, and the lung apex contour, and a slip feature map is assigned to the other contours. In other words, a motion feature map that can stably capture the presence or absence of adhesion even in the relevant region is assigned to the region where image features are poor outside the lung field, such as the longitudinal contour, the lung base contour, and the lung apex contour. Then, a slip feature map that can more accurately capture the presence or absence of adhesion is assigned to the other regions by using the characteristics of movement both inside and outside the lung field.

[0069] This makes it possible to generate a map that integrates feature amounts calculated using a more appropriate method depending on the characteristics of the image features of each part and its surroundings.

[0070] The method of dividing the lung contour is not limited to the method based on the type of anatomical part or the adjacent relationship with other adjacent parts as described above. For example, the lung contour may be divided based on the reliability of the registration of the lung field processed in step S10304. More specifically, L'(x)=1 can be set at positions on the lung contour where the reliability of the registration of the lung field at each position is smaller than a predetermined threshold, and L'(x)=2 can be set at other positions on the contour.

[0071] Here, the reliability of the registration can be calculated based on, for example, the local similarity between the images of the registration result. Specifically, L'(x)=1 can be set at the position of the contour where the similarity between the images is lower (not similar) than a predetermined value, and L'(x)=2 can be set at the other positions of the contour. In addition, the reliability can be calculated based on the presence or absence or the degree of image features (e.g., image edges, etc.) outside the lung field of the image to be registered. In addition, when performing a registration process involving repeated optimization, the reliability can be calculated based on the stability and convergence of the provisional solution during the repeated calculation. This allows the integrated map to be generated without using a registration result with low reliability, so that a more appropriate integrated map can be generated.

[0072] In addition, when using this method, depending on the reliability of the registration, the entire integrated map may become either the motion feature map or the slippage feature map. In addition, the threshold for the reliability may be adaptively changed so that the size of the area where L'(x) = 1 and L'(x) = 2 is a predetermined ratio.

[0073] The method of dividing the lung contour is not limited to the above, and may be configured to obtain user input via operation unit 515 such as a mouse or keyboard (not shown), and the above division may be performed based on the obtained user input. Specifically, the shape of the lung contour extracted in step S1010 may be displayed on display device 60, and the contour with L'(x)=1 may be obtained by the user's mouse operation (for example, tracing the contour position, selecting an area, etc.).

[0074] Furthermore, the method of dividing the lung contour is not limited to the above method. For example, a standard lung contour model in which the lung contour on the longitudinal side and the lung contour on the thoracic side are distinguished may be obtained, and the contour may be divided by corresponding (mapping) this standard lung contour model to the lung contour of the subject.

[0075] In this case, the standard lung contour model can be acquired from the examination image database 30 as the process of step S1000, similar to the standard motion amount map Rm(x) and the standard slip amount map Rs(x). Also, similar to the process of step S10206 or step S10308, the lung contour of the standard lung contour model can be associated with the lung contour of the subject by a correction function that matches the lung contour, apical position, and basal position of the lung in the standard lung contour model. According to the above method, it is possible to more stably divide the lung contour on the longitudinal side and the thoracic side.

[0076] (Variation 1-2: Other than the lungs (heart)) In the description of this embodiment, the case where the subject's lungs are the processing target is described as an example, but is not limited thereto. For example, another anatomical part such as the subject's heart may be the processing target. In this case, step S1000 acquires an image of the subject's pulsating heart. Specifically, a plurality of images with different phases of the heart's pulsation are acquired. Then, the contour of the heart and the movement of the myocardium between different phases can be analyzed by a plurality of methods. Then, a map can be generated by integrating the results of the analysis by these multiple methods.

[0077] For example, multiple maps can be switched and integrated depending on the adjacency relationship between the lungs and the diaphragm and the positional relationship with the coronary artery. Here, the method of analyzing the contour of the heart and the motion of the myocardium may include, for example, a method of generating a feature map capturing local expansion and contraction and a feature map of the speed of motion. More specifically, for the contours of the coronary artery and its surrounding positions, a feature map capturing local expansion and contraction can be calculated, and for the other contours, a motion feature map can be generated. This makes it possible to generate a map that allows the functional analysis of the heart and the presence or absence of adhesion between the heart and surrounding areas to be observed at the same time.

[0078] Furthermore, the present invention is not limited to this, and may be directed to other organs or parts of the human body, or may be directed to organs or parts of organisms other than the human body, or parts of non-living things.

[0079] (Variation 1-3: Variation of feature types) In the description of this embodiment, the case where a motion feature map and a slip feature map are calculated is described as an example, but the present invention is not limited to this. For example, this embodiment can be applied to the case where a map other than a motion feature and a slip feature, such as a local volume change (expansion / contraction) on a lung contour or a local area change (expansion / contraction) on a contour surface is calculated. For example, the difference between whether or not a lesion near the pleura has invaded outside the pleura may appear in the above features. In this case, different feature maps can be integrated on a contour close to the lesion and on other contours according to the position of the lesion on the lung contour. For example, a feature map can be calculated from a local volume change on a contour close to the lesion position, and a map can be calculated from a motion feature or a slip feature on other contours. This makes it possible to generate a map that allows the presence or absence of extrapleural infiltration of the lesion and adhesion of other contours to be observed at the same time.

[0080] In addition to the above, a plurality of features calculated with or without a comparison with a standard feature executed in step S10206 or step S10308 may be integrated to generate an integrated map. For example, when dividing the lung contour into the apical contour and the other contours, the integrated map may be generated so that the apical contour has a value of the slip feature map M'(x) and the other contours have a value of the slip map M(x). This allows the comparison with the standard slip map to be selectively performed depending on the region when generating the slip feature map.

[0081] The apical lung region, as exemplified above, is a region with a relatively small amount of sliding compared to other regions, and therefore has the following effects: For the apical lung region, a feature map that emphasizes the difference from a normal case can be generated by performing a comparison operation with a normal case, and for regions other than the apical lung region, a feature map that is close to the physical measurement value analyzed from the input image can be generated.

[0082] In addition to the above, for example, the process of step S1030 may be executed multiple times with different execution conditions, and each result may be assigned to the longitudinal contour and the thoracic contour, respectively, to generate an integrated map. As a specific example, in the registration process of steps S10302 and S10304, registration may be executed under different regularization conditions (conditions for the degree to which the registration result is smoothed).

[0083] For example, registration can be performed under weak regularization conditions (capturing finer differences in slippage) for lung contours close to the ventral side or the apex of the lung, where the movement and slippage of the lung contour due to breathing is relatively small, and for contours on the longitudinal side, where the movement and slippage characteristics are relatively complex. This makes it possible to generate a map that integrates the results of analysis using an appropriate method based on the movement characteristics of the area being analyzed.

[0084] (Variation 1-4: Variation of display form) In the description of this embodiment, an example has been described in which an integrated map is generated by integrating a motion feature map and a slippage feature map in the process of step S1050, and the integrated map is displayed in step S1060, but this is not limiting. For example, the generation of an integrated map may not be executed, and an observation image in which the motion feature map and the slippage feature map are arranged may be generated and displayed in step S1060. Alternatively, a mechanism may be provided in which a user specifies either the motion feature map or the slippage feature map, and an observation image may be generated and displayed based on the specification.

[0085] In addition, when the integrated map is displayed, the result of the division of the contour executed in step S1040 may be displayed so that it can be visually recognized on the integrated map. For example, a boundary line or the like may be displayed between the regions of the vertical contour side contour and the thorax side contour. Alternatively, the motion feature map on the vertical contour and the slip feature map on the thorax side contour may be displayed in different display forms to generate an observation image. For example, the motion feature map on the vertical contour may be a color map, and the slip feature map on the thorax side contour may be displayed in grayscale. Alternatively, the motion feature map on the vertical contour and the slip feature map on the thorax side contour may be displayed in a color map with different color schemes to generate an observation image. This allows the integrated map to be displayed in a display form suitable for each analysis method, as well as in a form in which the user can easily distinguish and recognize the regions of different analysis methods.

[0086] There are various other methods that can be used to allow the user to easily distinguish between regions with different analysis methods, such as adding annotations to each region, displaying the regions in different flashing or lit states, displaying the regions with different opacity, etc. This embodiment can also be applied to methods other than those mentioned above.

[0087] [Second embodiment] In the first embodiment, a case where the slippage feature and the motion feature are calculated for the entire lung of the subject to be examined and integrated according to the part of the lung is described as an example, but this is not limited to this. In the second embodiment, a case where the lung contour is divided before the motion feature and the slippage feature are calculated for each divided region is described as an example.

[0088] <Configuration of Information Processing System> Fig. 8 is a diagram showing the overall configuration of an information processing system according to this embodiment. The same elements as those described in the overall configuration of the first embodiment are assigned the same numbers, and detailed descriptions are omitted. The information processing system includes an information processing device 70, an examination image database 30, an examination image photographing device 40, and a display device 60, and these devices are connected to each other so as to be able to communicate with each other via a communication unit.

[0089] <Functional configuration of information processing device> The information processing device 70 acquires information held in the inspection image database 30 via the LAN 50. The information processing device 70 includes a data acquisition unit 100, a contour extraction unit 110, a contour division unit 140, a movement feature amount calculation unit 320, a slippage feature amount calculation unit 330, an integrated map generation unit 350, and a display control unit 160.

[0090] The motion feature amount calculation unit 320 calculates a map of motion feature amounts in the pleura on the longitudinal side based on the division result of the region division unit 140. The slippage feature amount calculation unit 330 calculates a map of slippage feature amounts in the pleura on the thoracic side based on the division result of the region division unit 140. The integrated map generation unit 350 generates an integrated map by integrating the maps of the features calculated by the motion feature amount calculation unit 320 and the slippage feature amount calculation unit 330.

[0091] The hardware configuration of the information processing device 70 is similar to the configuration in the first embodiment shown in FIG. 10, and therefore will not be described.

[0092] <Processing> Next, the entire processing procedure by the information processing device 70 in this embodiment will be described in detail with reference to Fig. 9. In addition, the following will be described as an example in which CT data is used as the examination image, but the present invention is not limited to this. For example, as long as the time-series three-dimensional volume data of the lungs is taken, it may be an MRI image or an ultrasound image.

[0093] (Step S2000: Acquiring data) In step S2000, the data acquisition unit 100 of the information processing device 70 executes the same process as in step S1000 in the first embodiment, and a detailed description thereof will be omitted.

[0094] (Step S2010: Extraction of lung contour) In step S2010, the contour extraction unit 110 of the information processing device 70 executes the same process as in step S1010 in the first embodiment, and a detailed description thereof will be omitted.

[0095] (Step S2020: Segmentation of lung contour) In step S2020, the contour division unit 140 of the information processing device 70 executes the same process as in step S1040 in the first embodiment, and a detailed description thereof will be omitted.

[0096] (Step S2030: Analysis of lung contour movement) In step S2030, the motion feature amount calculation unit 320 processes the 4DCT data I to analyze the motion of the contour of the lung on the longitudinal side of the subject, and calculates a motion feature amount map M''(x) due to breathing. Specifically, the motion of the lung contour on the longitudinal side calculated in step S2020, i.e., the lung contour at the position where L'(x) = 1, is analyzed to calculate the amount of motion. This can be calculated by performing the same processes as steps S10202 to S10204 in the first embodiment on the lung contour on the longitudinal side. For example, the inspiration image I_ins and the expiration image I_exp can be cropped (cut out) so as to include the lung contour on the longitudinal side, and processing can be performed on the cropped image.

[0097] Then, a process similar to step S10206 in the first embodiment is performed on the lung contour on the longitudinal side to calculate a motion feature map M''(x) of the lung contour on the longitudinal side. Here, M''(x) is a function that returns a motion feature at position x of the lung contour on the longitudinal side, and returns an invalid value for lung contours other than the longitudinal side and positions x other than the lung contour. This process step is similar to step S1020 in the first embodiment, but differs in that processing is limited to the position of the lung contour on the longitudinal side. However, the calculation method of the motion feature map performed in this process step may be the same as that in step S1020 in the first embodiment.

[0098] (Step S2040: Lung contour sliding analysis) In step S2040, the slip feature calculation unit 330 processes the 4DCT data I to analyze the movement of the lung contour on the thoracic side of the subject, and executes a process to calculate a slip feature map S''(x) due to breathing. Specifically, the slip amount is calculated by analyzing the movement of the lung contour on the thoracic side calculated in step S2020, i.e., the lung contour at the position where L'(x) = 2. This can be calculated by executing the same process as steps S10302 to S10306 in the first embodiment, but for the lung contour on the longitudinal side. For example, the calculation can be performed by cropping the inspiration image I_ins and the expiration image I_exp to include the lung contour on the longitudinal side, and executing a process on the cropped image.

[0099] Then, a process similar to step S10308 in the first embodiment is performed on the lung contour on the thoracic cage side to calculate a slippage feature map S''(x) of the lung contour on the thoracic cage side. Here, S''(x) is a function that returns a slippage feature at position x of the lung contour on the thoracic cage side, and returns an invalid value for lung contours other than on the thoracic cage side and positions x other than the lung contour. This processing step is similar to step S1030 in the first embodiment, but differs in that processing is limited to the position of the lung contour on the thoracic cage side. However, the method of calculating the slippage feature map performed in this processing step may be the same as that of step S1030 in the first embodiment.

[0100] (Step S2050: Generate integrated map) In step S2050, the integrated map generator 350 executes a process of generating an integrated map F'(x) by integrating the motion feature map M''(x) calculated in step S2030 and the slippage feature map S''(x) calculated in step S2040. Here, the integrated map F'(x) is a map that has the values ​​of the motion feature map M''(x) at the contour position on the longitudinal cage side and the values ​​of the slippage feature map S''(x) at the contour position on the thorax cage side. More specifically, the integrated map F'(x) is generated by the calculation shown in equation (2).

[0101]

number

[0102] Note that M''(x) is a function that returns an invalid value at positions other than L'(x)=1, and S''(x) is a function that returns an invalid value at positions other than L'(x)=2. Therefore, a result similar to that of equation (2) can be obtained by substituting the non-invalid value of either of the two functions into F'(x), and substituting an invalid value into F'(x) when both functions are invalid. In addition, the method of generating the integrated map is not limited to the above method, and a method of smoothly integrating two types of feature maps as described in step S1050 of the first embodiment may also be used.

[0103] (Step S2060: Display / Save) In step S2060, the display control unit 160 of the information processing device 70 executes the same process as in step S1060 in the first embodiment. A detailed description will be omitted. The process of the information processing device 70 in this embodiment is executed by the method described above.

[0104] In this embodiment, a specific region (e.g., lung contour) of an anatomical part (e.g., lung) is extracted from an examination image of a subject, and the specific region is divided into a plurality of regions including a first region and a second region. Then, a first feature amount (e.g., movement feature amount) in the first region related to the movement of the anatomical part and a second feature amount (slip feature amount) in the second region are obtained. After that, the first feature amount and the second feature amount are integrated.

[0105] This makes it possible to obtain features by a method suitable for each region of the subject's anatomical region. Therefore, for each region of the subject's lung contour, a result of integrating the analysis results of features calculated by a method suitable for that region (for example, an integrated map integrating each feature amount map) can be provided to the user. Furthermore, compared to the first embodiment, each analysis process can be performed limited to the region required to generate the integrated map, making it possible to obtain similar results with higher computational efficiency.

[0106] (Variation 2-1) The above description is merely an example. For example, the maps calculated in steps S2030 and S2040 may be displayed and saved in step S2060 without executing the integrated map generation process in step S2050.

[0107] For example, each map may be displayed by three-dimensional surface rendering, similarly to step S1060 in the first embodiment. At this time, the display may be switched between the motion feature map and the slippage feature map based on display conditions such as the viewpoint position and line of sight controlled by user operation, etc. More specifically, the motion feature map may be displayed under conditions in which the contour position on the vertical side from the viewpoint position and line of sight direction is displayed on the near side, and the slippage feature map may be displayed under other conditions.

[0108] This provides the user with the ability to switch the display of each map as well as manipulate the viewpoint position and line of sight, and also provides a mechanism for observing each feature map from a viewpoint position and line of sight that is suitable for observation.

[0109] (Variation 2-2) The above description is merely an example. For example, an information processing device having the functions of both the first embodiment and the second embodiment can also be one embodiment. In this case, the user can select which form of processing to perform, or the device may be configured to automatically select based on the characteristics of the image to be analyzed, information on the target area, etc.

[0110] (Other embodiments) The present invention can also be realized by a process in which a program for implementing one or more of the functions of the above-described embodiments is supplied to a system or device via a network or a storage medium, and one or more processors in a computer of the system or device read and execute the program. The present invention can also be realized by a circuit (e.g., ASIC) that implements one or more of the functions.

[0111] The invention is not limited to the above-described embodiments, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, the following claims are appended to apprise the public of the scope of the invention. [Explanation of symbols]

[0112] 10, 70: information processing device, 30: inspection image database, 40: inspection image capturing device, 50: LAN, 60: display device, 100: data acquisition unit, 110: contour extraction unit, 120, 320: movement feature amount calculation unit, 130, 330: slippage feature amount calculation unit, 140: contour division unit, 150, 350: integrated map generation unit, 160: display control unit

Claims

1. An extraction means for extracting a predetermined region of an anatomical part from an examination image of a subject; a first feature amount acquiring means for acquiring a first feature amount related to a movement of the anatomical part in a first region included in the predetermined region; a second feature amount acquiring means for acquiring a second feature amount, which is different in type from the first feature amount, related to the motion in a second area included in the predetermined area and different from the first area; an integration means for integrating the first feature amount and the second feature amount so as to represent the first feature amount in the first region and the second feature amount in the second region; An information processing device comprising:

2. 2. The information processing apparatus according to claim 1, further comprising display control means for displaying the integration result by said integration means on a display means.

3. 3. The information processing apparatus according to claim 2, wherein the display control means renders the integration result on the three-dimensional shape of the predetermined region of the anatomical site and causes the display means to display the result.

4. 4. The information processing apparatus according to claim 2, wherein the display control means further displays a boundary line between the first area and the second area.

5. 5. The information processing apparatus according to claim 1, further comprising a control unit configured to store the first feature amount, the second feature amount, and a result of the integration performed by the integration unit in a storage unit.

6. the anatomical site is a lung, and the predetermined area is a contour of the lung; the first feature amount is a movement feature amount indicating a movement of the contour due to breathing, The information processing apparatus according to claim 1 , wherein the second feature amount is a slippage feature amount indicating a magnitude of slippage of the contour due to breathing.

7. the movement feature is a feature based on an absolute movement amount of the lungs, The information processing apparatus according to claim 6 , wherein the slippage feature amount is a feature amount based on a relative movement amount between the lung and a surrounding area of ​​the lung.

8. 8. The information processing apparatus according to claim 6, wherein the first region is a longitudinal lung contour, and the second region is a thoracic lung contour.

9. 8. The information processing apparatus according to claim 6, wherein the first region is a contour of a lung base, and the second region is a contour of a part other than the lung base.

10. 9. The information processing apparatus according to claim 6, wherein the first region is a contour of the apical part of the lung, and the second region is a contour of a part other than the apical part of the lung.

11. the anatomical site is a heart, and the predetermined region is a contour of the heart; the first feature amount is a feature amount indicating a movement of the contour due to pulsation of the heart, 6. The information processing apparatus according to claim 1, wherein the second feature amount is a feature amount indicating local expansion and contraction of the contour due to pulsation of the heart.

12. 12. The information processing apparatus according to claim 11, wherein the second region is a contour of a coronary artery and a peripheral position of the coronary artery, and the first region is a contour other than the coronary artery.

13. the anatomical site is a lung, and the predetermined area is a contour of the lung; the first feature amount is a feature amount indicating a local volume change on the contour due to respiration, The information processing apparatus according to claim 1 , wherein the second feature amount is a slippage feature amount indicating a magnitude of slippage of the contour due to breathing.

14. 14. The information processing apparatus according to claim 13, wherein the first region is a contour of the lung in the vicinity of a lesion location, and the second region is a contour other than that.

15. 1. An information processing method, comprising: An extraction step of extracting a predetermined region of an anatomical part from an examination image of a subject; a first feature amount acquiring step of acquiring a first feature amount related to a movement of the anatomical part in a first region included in the predetermined region; a second feature amount acquiring step of acquiring a second feature amount, which is different in type from the first feature amount, related to the motion in a second area included in the predetermined area and different from the first area; an integration step of integrating the first feature amount and the second feature amount so that the first feature amount is represented in the first region and the second feature amount is represented in the second region; 13. An information processing method comprising:

16. A program for causing a computer to execute the information processing method according to claim 15.

Images