JPWO2012157406A1 - Image analysis apparatus, program, and image pickup apparatus - Google Patents

Abstract

  In the numerical analysis for obtaining the load on the bone body using the three-dimensional data, the analysis conditions such as the direction and the magnitude of the force necessary for the analysis are obtained accurately and simply. The muscle data 106, which is three-dimensional data representing muscles, and the bone data 105, which is three-dimensional data representing bones, are fused, and on the fused image, from the contact points of the bones and muscles and the running direction and size of the muscles. A muscle running vector is calculated to model the arrangement of bones and muscles, and arrangement model data 110 is created. Analysis conditions are set based on the analysis model data 112 of the arrangement model, and numerical analysis is performed on the subject image. As a result, for example, a simulation of a load on the bone body can be performed, and a surgical plan can be realized accurately and easily.

Description

  The present invention relates to an image analysis apparatus, and more particularly to an image analysis technique for fusing a plurality of types of images extracted from a radiation image or the like and analyzing the fused image.

  CT (Computed Tomography) apparatus, cone beam CT apparatus, MRI (Magnetic Resonance Imaging) apparatus, PET (Positron Emission Tomography) apparatus, SPECT (Single Photon Emission Computed Tomography) apparatus, Furthermore, there is an ultrasonic device. Many studies have been conducted in which a bone model is created from three-dimensional image data of a subject obtained with these apparatuses, and load data obtained by numerical analysis is used for an operation plan. For example, in the technical document 1, a technique is described in which finite element analysis is performed on the femur data extracted from the CT image under conditions assuming a static standing position, and the obtained stress distribution is used for selecting an artificial hip joint. .

  On the other hand, a technique for supplementing information by superimposing three-dimensional data of a plurality of types of subject images obtained by different apparatuses has been reported. For example, Patent Document 1 and Patent Document 2 disclose a technique for displaying a SPECT image and an MRI image in a superimposed manner. Patent Documents 3 and 4 disclose a technique for displaying a PET image and an MRI image in a superimposed manner. Patent Documents 5 and 6 disclose a technique for displaying a CT image and an MRI image so as to overlap each other.

JP 2008-125658 A JP 2006-053102 A JP 2008-036284 JP 2002-214347 A JP 2006-341085 JP Japanese Patent Laid-Open No. 05-344964

Kagoshima Prefectural Industrial Technology Center research report No.15, pp.81-86, 2001

  In the technology that makes use of the above CT images to select an artificial hip joint, in order to perform numerical analysis, the direction and magnitude of the force applied from the outside, and the direction and magnitude of the force that the muscles on the living body respond to are analyzed. Required as a condition. However, it is difficult to identify soft tissues such as muscles using CT images. As a result, the user empirically determines the direction and size, and accurate condition setting is difficult. Also, setting analysis conditions requires complicated work by the user.

  Therefore, it is conceivable to use the three-dimensional image data obtained by the different devices as described above, but until now, numerical analysis is performed using the data obtained by superimposing the three-dimensional image data obtained by different devices. There are no reports on methods for obtaining the necessary models and analysis conditions.

  In order to solve the above-described problems, an object of the present invention is to accurately and simply obtain a model and analysis conditions necessary for numerical analysis using data obtained by superimposing a plurality of three-dimensional image data obtained by different apparatuses. It is an object of the present invention to provide an image analysis apparatus and a program that can be used.

  Another object of the present invention is to solve the above-mentioned problems, in a numerical analysis for obtaining a load on a bone body that is a hard tissue using three-dimensional image data, such as a direction and a magnitude of a force necessary for the analysis. An object of the present invention is to provide an image analysis device, a program, and an image pickup device that can accurately and simply obtain analysis conditions.

  In order to achieve the above object, in the present invention, an image analysis apparatus including a processing unit that processes three-dimensional image data, and a program thereof, the processing unit stores muscle data representing muscles and a bone body. Fusion data is created by fusing the bone data to represent, and using this fusion data, arrangement model data showing the arrangement of bones and muscles is created, and analysis conditions are set based on the arrangement model data An image analysis apparatus and a program thereof are provided.

  In order to achieve the above object, according to the present invention, there is provided an image imaging device that captures and analyzes three-dimensional image data of a subject, the first three-dimensional image data and the second three-dimensional image data. Is extracted, muscle data is extracted from the first three-dimensional image data, and bone body data is extracted from the second three-dimensional image data, and the extracted muscle data and bone body data are fused to obtain the fusion data. Provided is an image pickup apparatus having a configuration including a processing unit that creates and creates arrangement model data indicating the arrangement of bones and muscles using fusion data, and sets analysis conditions based on the arrangement model data.

  The image analysis apparatus, program, or image pickup apparatus of the present invention enables accurate and automatic modeling and setting of analysis conditions according to the structure of each subject, and provides an accurate analysis result at high speed. There is an advantage that complicated manual processing is not required.

It is a flowchart figure which shows the procedure of the process of a 1st Example. It is a side view of a radiation imaging device concerning each example. It is a side view of another radiation imaging device based on each Example. It is a side view of another radiation imaging device based on each Example. It is a side view of another radiation imaging device based on each Example. It is a side view of another radiation imaging device based on each Example. It is explanatory drawing which shows the outline | summary of a process based on a 1st Example. It is explanatory drawing which shows the outline | summary of a process based on a 1st Example. It is a flowchart figure which shows the procedure of another process based on a 2nd Example. It is explanatory drawing which shows the outline | summary of another process based on a 2nd Example. It is a flowchart figure which shows the detail of the principal part of the process sequence based on 1st Example. It is a figure which shows the example of 1 structure of the image analysis apparatus based on each Example.

    Prior to describing embodiments of the present invention with reference to the drawings, a radiation imaging apparatus to which each embodiment is applied will be described below.

  FIG. 2 is a side view of a radiation imaging apparatus using X-rays to which each embodiment is applied. The X-ray source 201 and the detector 202 in the X-ray tube 200 are installed at both ends of a C-shaped column 203, and are arranged so that the detection surfaces of the X-ray source 201 and the detector 202 face each other. The subject 207 is arranged in a state lying on the subject holding device 205, and the X-ray source 201 and the detector 202 installed on the support column 203 rotate around the rotation axis 206 around the subject 207. In this apparatus, the X-ray source 201 and the rotating shaft 206 of the detector 202 are on the paper surface and are installed in parallel to the floor surface. Reference numeral 208 denotes an image analysis device that analyzes an image of X-rays transmitted through the subject 207 in accordance with the output of the detector 202.

  FIG. 3 shows a side view of another radiation imaging apparatus. In this device, the column 303 slides on the rotation mechanism 304 and rotates. Therefore, the rotation shaft 306 is set to be parallel to the floor surface and penetrate the paper surface. In the radiation imaging apparatus of FIG. 3, the length of the subject 307 that can be photographed in the body axis direction is not limited because the column 303 is inserted from the side surface of the subject 307. For example, even when a C-arm having a small diameter is used, it is possible to take an image of the center of a subject that is long in the body axis direction, such as the abdomen. It is a form suitable for IVR (Interventional Radiology).

  FIG. 4 shows a state in which the column 403 is rotated 90 degrees around the rotation mechanism 404 with respect to the radiation imaging apparatus of FIG. In this apparatus, the X-ray source 401 and the detector 402 rotate in such a manner that the support column 403 slides on the rotation mechanism 404. The rotation shaft 406 is on the paper surface and is set perpendicular to the floor surface. In the radiation imaging apparatus of FIG. 4, the subject 407 can be imaged while standing or sitting. It is a form suitable for orthopedics and dentistry. Further, the same radiation imaging apparatus can take both the imaging modes shown in FIGS.

  FIG. 5 shows the radiation imaging apparatus in a state where the column 503 is suspended. FIG. 6 shows a radiation imaging apparatus when a gantry is used as the column 603. The device of FIG. 5 can rotate in a stable horizontal plane and is suitable for dentistry. The apparatus of FIG. 6 can rotate at high speed, and is suitable for a moving object.

  In addition to the C-shape shown in the figure, the columns 203 to 603 of the radiation imaging apparatuses described above are U-shaped, U-shaped, the gantry described above, and the like. In addition to the form of supporting columns 203 to 603 supported from the side, there are forms suspended from the ceiling and forms supported on the floor. A bed or a chair is used for the subject holding devices 205 to 605. In addition to fixing the subjects 207 to 607 and rotating the X-ray sources 201 to 601 and the detectors 202 to 602, rotating the subjects 207 to 607 and fixing the X-ray sources 201 to 601 and the detectors 202 to 602 There is a form to do. Since rotation can be performed with a simpler apparatus as compared with a system that moves a detector or the like, stable rotation is possible. Further, there is a form in which the subjects 207 to 607 and the X-ray sources 201 to 601 and the detectors 202 to 602 are rotated, and the imaging time can be shortened compared to a system in which one of them is rotated. Also, by moving both or one of the support columns 203 to 603 and the subject holding devices 205 to 605, the rotation shafts 206 to 606 are set obliquely with respect to the axes of the subjects 207 to 607, or oblique to the floor surface. It is also possible to set to.

  A one-dimensional detector or a two-dimensional detector is used for the detectors 202 to 602 of each radiation imaging apparatus. As a two-dimensional detector, a one-dimensional detector arranged in multiple rows, a planar X-ray detector, a combination of an X-ray image intensifier and a CCD (Charge Coupled Device) camera, an imaging plate, a CCD detector, There are solid state detectors. As a planar X-ray detector, there is a combination of an amorphous silicon photodiode and a TFT (Thin Film Transistor) paired on a square matrix and directly combined with a fluorescent plate.

  X-rays generated from the X-ray sources 201 to 601 pass through the subjects 207 to 607, converted into electric signals according to the X-ray intensity by the detectors 202 to 602, and input as measurement images to the image analysis devices 208 to 608. And constructed into a three-dimensional image. Alternatively, after a three-dimensional image is constructed by a processing device (not shown), it is input to the image analysis devices 208 to 608 as three-dimensional image data. In the image analysis devices 208 to 608, numerical analysis processing is performed using the obtained three-dimensional image data. By being connected to the detectors 202 to 602, it can be executed as a series of processes from photographing to analysis, and the burden on the user can be reduced.

  The image analysis devices 208 to 608 are separated from the detectors 202 to 602, and can perform numerical analysis processing alone. In that case, it can be freely carried and can be miniaturized, improving the convenience for the user. The 3D images handled by the image analysis devices 208 to 608 can be real-time, previously captured data input from various 3D image capturing devices such as the above-described MRI and PET devices. It is also possible to use data stored in a storage unit, a storage medium, or the like. Further, the type is not limited to the above-described three-dimensional image data by X-rays, and other three-dimensional image imaging apparatuses such as image data, for example, three-dimensional image data such as MR image data and ultrasonic image data are available. Used.

  FIG. 12 is a diagram illustrating a configuration example of the image analysis apparatus. As shown in the figure, it has a normal computer configuration, and a central processing unit (CPU) 1202 which is a processing unit for executing software such as an image analysis program on an internal bus 1207, various programs and various types Main memory (MM) 1201 and hard disk drive (HDD) 1204 as storage units for storing processing data such as three-dimensional image data, and liquid crystal display (LCD) as a display unit for various displays An input unit (Input) 1205 such as a keyboard and a communication interface (I / F) 1206 are connected to each other, and sequentially process the analysis target data. Various types of 3D image data are input via the input unit 1205 or the communication interface 1206 and stored in the storage unit. Note that this image analysis apparatus can be installed as a part of the three-dimensional image capturing apparatus.

  Embodiments of the present invention will be described below with reference to the drawings. In this specification, bones, teeth, and the like are referred to as hard tissues or bone bodies corresponding to soft tissues such as muscles and internal organs. Further, in this specification, the muscle running vector means a vector indicating the running direction of the muscle.

  The first embodiment relates to an image analysis apparatus including a processing unit for processing three-dimensional image data, and a program thereof. The processing unit includes muscle data representing muscles, bone data representing bones, To create fusion data, create placement model data showing the arrangement of bones and muscles using the fusion data, set analysis conditions based on the placement model data, and further based on the analysis conditions It is the Example of the image-analysis apparatus of the structure which performs a numerical analysis, and its program.

  FIG. 1 shows a processing procedure in the image analysis apparatus of the first embodiment. This processing procedure can be realized by a program executed by the CPU 1202 which is the processing unit of the image analysis apparatus described with reference to FIG. The same applies to various processing procedures in the following description.

  In this embodiment, first, a hard tissue and a bone are extracted from a three-dimensional CT image 101 obtained by CT imaging (103), and bone data 105 is obtained. In addition, muscle, which is a soft tissue, is extracted from a three-dimensional MR image 102 obtained by MR imaging (104), and muscle data 106 is obtained. Bone body data and muscle data are fused (107) to obtain bone body and muscle fusion data 108. On the fusion data 108, the contact point between the bone and the muscle and the muscle running vector which is the running direction of the muscle are calculated (109), and the arrangement model data 110 is created. Analysis conditions obtained from the arrangement model data are set on the analysis model created from the bone body data (111), and the analysis model data 112 is created. Further, numerical analysis 113 is executed on the analysis model data 112 to obtain an analysis result 114.

  By calculating the muscle running vector and contact points and creating the arrangement model data 110 according to the processing procedure of the present embodiment, the analysis conditions can be automatically set, and complicated manual processing becomes unnecessary. In addition, by creating an analysis model based on shooting data of a subject, modeling according to the structure of each subject can be performed, and an accurate analysis result can be obtained.

  The CT apparatus measures the difference in the amount of X-ray absorption, and is superior in imaging hard tissues such as bones and teeth as compared to other three-dimensional imaging apparatuses. Further, a high spatial resolution three-dimensional image can be obtained at high speed. Therefore, accurate bone data can be created by calculating bone data using a CT image. As a result, an accurate analysis result can be obtained. Further, since the subject is not restrained for a long time, the burden on the subject is reduced. In addition, since the result can be obtained in a short time, the burden on the user can be reduced.

  The MR device measures the state of hydrogen nuclei, and is superior to other soft tissue imaging compared to other three-dimensional imaging devices, so it is suitable for muscle imaging. Therefore, it is possible to create highly accurate muscle data by calculating muscle data using MR images. As a result, an accurate analysis result can be obtained.

  FIG. 7 shows an outline of processing in the case where the subject is the mandible as a subject by the processing procedure of the image analysis apparatus of the present embodiment. First, the inclination of the axis and the magnification of the image are corrected with the CT image and MR image. For example, the external shape of the face is extracted by threshold processing, the difference image between the CT image and the MR image is obtained by changing the angle and the enlargement ratio, and the value that minimizes the difference value is obtained. Alternatively, characteristic objects such as bones and blood vessels are extracted, and an angle and an enlargement ratio at which the value of the difference image between the CT image and the MR image is minimized.

  Next, a region extraction process is performed on each slice image of the CT image, a bone region is extracted, and bone body data 701 is calculated. Further, region extraction processing is performed on each slice image of the MR image, a muscle region is extracted, and muscle data 702 is calculated. Area extraction processing includes thresholding processing and region growing processing that considers continuity between adjacent pixels and slices.

  When extracting a region, the accuracy of extraction is improved by using the bone shape model and the muscle shape model. As the shape model, a standard object shape, a shape obtained by computer simulation, a shape given by a human atlas, or the like is used. For example, the external shape of the face of the shape model is compared with the external shape of the face of the photographed subject, the deformation of the subject is calculated, the bone shape model and the muscle shape model are deformed to fit the subject, and this is used as a template. . Region extraction is performed while comparing with the template. In particular, since the boundary is unclear in the region at the end of the muscle, it is possible to prevent the region from expanding without limit by using a template, and the accuracy of region extraction is improved.

  In a state where the angle and the enlargement ratio are combined, the extracted muscle data is superimposed on the extracted bone body data to be fused (703). A method based on the position of the spine is effective when the CT image and the MRI image are superimposed. Since the spine has a characteristic shape, it can be easily distinguished from other parts and is suitable for the reference. Further, since each shape is characteristic in the XYZ directions, the difference value is likely to change due to the difference in the enlargement ratio or angle, which is suitable as a target for finding the optimum value of the enlargement ratio or angle. In addition, because the structure is hard and difficult to deform, there is little deviation when there is a time difference between CT and MRI imaging, or when there are differences in imaging conditions such as standing and supine positions. Can be easily done.

  Here, a point where the muscle data 702 contacts or intersects the bone body data 701 is defined as a contact point between the muscle and the bone body, and a contact area is obtained. If there is no contact, the muscle data region extraction is continued until the contact is found. Alternatively, the bone body closest to the end of the muscle is used as the contact point. In that case, the size and shape of the contact area are obtained from the muscle shape model, and the muscle area is extracted around the contact.

  The contact point between the bone and the muscle may be fixed, rotated, or limited movement. If fixed, numerical analysis is easy, and an analysis result can be obtained in a short time. If rotation and limited movement are used, the analysis becomes complicated but detailed analysis is possible, and an accurate analysis result can be obtained. As described above, fused bone body data and muscle data are obtained.

  At each slice, the center of the muscle region is calculated, and a muscle running vector connecting them is obtained. The center may be the center of gravity, the center when assumed to be a circle, or the position half the length in two orthogonal directions. Most simply, in each muscle region, a straight line connecting from the center of one contact region to the center of the other contact region is obtained, and a vector is set along the straight line. Alternatively, an approximate straight line is obtained for the center of the muscle region of each slice, and a vector is set along the straight line. The direction of the vector is given by the muscle shape model. An approximate curve, arc, or ellipse approximation for the center of the muscle region of each slice may be obtained and used as a muscle running vector. In the linear approximation, the muscle running vector is simple, the vector can be easily calculated, the numerical analysis is also simple, and the analysis can be completed in a short time (705). In curve approximation, since vectors are complex, it takes time to calculate and analyze vectors, but detailed analysis is possible and accurate analysis results can be obtained.

  The magnitude of the muscle running vector is calculated by estimating the quality and strength of the muscle from the pixel value of the muscle area in the captured image and the size of the muscle area. For example, the higher the pixel value, the better the quality, and the larger the region, the stronger. A relational expression between the pixel value and the strength is obtained in advance, and the magnitude is converted from the pixel value using the relational expression to set the magnitude of the vector. As a result, the magnitude of the muscle running vector 707 can be automatically calculated (705).

  Further, a relational expression between the pixel value of the bone body region in the photographed image and the hardness of the bone body is obtained, and the strength of the bone body is set in the analysis model. Thereby, the strength of the analysis model can be automatically calculated, and the arrangement model data 706 can be obtained in the three-dimensional space.

  FIG. 8 shows an outline of calculation of the muscle running vector in the present embodiment. As shown in FIG. 8A, a muscle region 802 is extracted on the slice image 801. Here, four types of muscle regions 802 represented by different textures are extracted. As shown in (b), a muscle region is extracted on each slice image. A region center 803-1 is calculated for the muscle region 802-1 extracted on the slice image 801-1. Similarly, a region center 803-2 is calculated for the muscle region 802-2 extracted on the slice image 801-2. A straight line connecting the region center 803-1 and the region center 803-2 is obtained, and a muscle running vector 804 is set along the straight line.

  Depending on the type of muscle, whether it extends in the Coronal, Sagittal, or Axial direction is different. The general running direction of the muscle is given by the muscle shape model. When calculating the muscle running vector 804, a slice image is cut out and processed along the running direction given by the muscle shape model. Since CT images and MR images have high resolution in the Axial image, the region can be accurately extracted by cutting out the slice image as the Axial image. When a slice image is cut out as a Coronal image or a Sagittal image, a muscle running vector extending in a direction along the body axis can be easily calculated. Further, the slice image is not limited to the Coronal direction, the Sagittal direction, and the Axial direction, and may be cut obliquely with respect to the axis. As a result, the center of the region can be accurately calculated for the oblique muscle, and the calculation accuracy of the muscle running vector is improved.

  The image analysis apparatus according to the first embodiment has been described in detail above. The processing procedure is collectively illustrated in FIG.

  In FIG. 11, the facial contours are extracted from the CT image 1101 and the MR image 1102 by threshold processing 1107 and 1108, respectively, and facial contour data 1109 and 1110 are obtained. For these data, a difference image 1112 between the CT image and the MR image is obtained by a difference operation (1111), and a value that minimizes the difference value is obtained (1113). Alternatively, characteristic objects such as bones and blood vessels are extracted, and an angle and an enlargement ratio at which the value of the difference image between the CT image and the MR image is minimized. The angle and the enlargement ratio are adjusted repeatedly until the difference value is minimized (1114), and the angle and the enlargement ratio are obtained (1117). Then, using the finally determined angle / magnification rate, the angle / magnification rate is adjusted (1115, 1116) for each of the bone data 1105 and muscle data 1106 described above, and the angle and magnification rate are adjusted. Are combined (1118) to obtain bone + muscle fusion data 1119.

  Here, the value of the difference image is, for example, a representative value such as an average value, median value, mode value, maximum value, or minimum value of the pixel values of the difference image. If the region for calculating the representative value is the entire image, the shape of the entire head can be reflected. When the partial region is used, for example, the shape of the bone can be strongly reflected. The minimum judgment can be made with high accuracy by plotting the number of iterations on the horizontal axis and the value of the difference image on the vertical axis, and taking the value of the difference image value as the minimum value or the minimum value. It is. Alternatively, if the value of the difference image is a value at the number of times that is smaller than a preset value, the determination becomes easy.

  Subsequently, a muscle running vector is calculated based on the fusion data 1119. First, a point where the muscle data 1106 contacts or intersects with the bone body data 1105 is defined as a contact point between the muscle and the bone body (1120), and a contact area 1121 is obtained. If there is no contact, the muscle data region extraction is continued until the contact is found. Alternatively, the bone body closest to the end of the muscle is used as the contact point. In that case, the size and shape of the contact area are obtained from the muscle shape model 1123, and the muscle area is extracted around the contact (1122). This process is performed on all slices of one muscle (1124). Such a process is executed for all types of muscles (1125).

  Then, the center of each muscle region is calculated (1126), and a muscle running vector calculation process connecting them is performed (1127) to obtain the muscle running vector 1128. The subsequent processing is as described with reference to FIG. 1, and processing such as obtaining arrangement model data in a three-dimensional space is continued based on this muscle running vector, but description thereof is omitted here.

  Moreover, although the description of the above embodiment has been described as an image analysis apparatus, the configuration of the above embodiment can also be applied to an image capturing apparatus. That is, an input unit for inputting first three-dimensional image data such as CT image data and second three-dimensional image data such as MR image data, first three-dimensional image data, and second three-dimensional Extract muscle data and bone body data from image data, fuse the extracted muscle data and bone body data to create fusion data, and use the fusion data to create placement model data indicating the bone and muscle placement In addition, it is possible to configure an image capturing apparatus including a processing unit that sets numerical conditions based on the arrangement model data and performs numerical analysis, and further includes a display unit.

  The second embodiment is another embodiment of an image analysis apparatus having a processing unit for processing three-dimensional data and a program thereof, and the processing unit uses an existing muscle running vector as muscle data. , Create fusion data by fusing with bone body data, create placement model data showing bone and muscle placement using the fusion data, set analysis conditions based on the placement model data, It is the Example of the image analysis apparatus of the structure which analyzes, and its program.

  FIG. 9 shows a processing procedure in the processing unit in the image analysis apparatus of the second embodiment using an existing muscle running vector. FIG. 10 shows an outline of the processing of the second embodiment when the mandible is the subject. First, a bone body is extracted from a newly photographed CT image 901 (902), and bone body data 903 is obtained. The existing muscle running vector 904 and the bone body data 903 are fused (905), and the arrangement model data 906 is created in a three-dimensional space. The analysis conditions obtained from the arrangement model data 906 are set on the analysis model created from the bone body data 903 (907), and the analysis model data 908 is created. A numerical analysis 909 is executed on the analysis model data 908 to obtain an analysis result 910.

  By using the existing muscle running vector 904 as in the present embodiment, it is possible to automatically set analysis conditions even when only a CT image is obtained as photographing data of a subject.

  In the above embodiment, bone body data is extracted from a new CT image, and muscle data is calculated from an MR image. However, bone body data and muscle data are extracted using a three-dimensional image obtained by another imaging apparatus. May be. For example, a CT apparatus, an MR apparatus, a PET apparatus, a SPECT apparatus, an ultrasonic apparatus, etc. can be considered. For example, muscle data is extracted from the ultrasound image. Since the ultrasonic device is small, light, and inexpensive, it can easily shoot compared to other photographic devices, and photographic data can be obtained easily. Moreover, since it is small in size, it can be used in combination with a CT apparatus, and imaging data of bones and muscles can be obtained simultaneously. Since there is no time lag between the bone body data and the muscle data, there is no inclination or position deviation, and the data can be easily merged, the processing speed can be increased, and an accurate analysis result can be obtained.

  Alternatively, bone body data and muscle data may be extracted using data created by computer simulation or data obtained by human body atlas. As a result, even when bone body data and muscle data cannot be photographed, numerical analysis can be executed and an analysis result can be obtained. Specifically, these data are stored as a shape model. Compared with actual imaging data, the deformation of the subject is calculated, and the bone shape model and the muscle shape model are transformed to fit the subject and used as bone data and muscle data.

  Alternatively, bone body data and muscle data may be extracted from images captured by various existing devices. Thereby, numerical analysis can be executed and the analysis result can be obtained without newly photographing the subject.

  In the middle of the above processing, the progress may be displayed on the display unit so that the user can confirm it. Further, the user may be able to correct the intermediate result on the confirmation screen. As a result, analysis conditions can be easily set semi-automatically, and occurrence of errors can be avoided. In addition, since the progress can be confirmed, the user can use the analysis results with peace of mind.

  For example, as the progress in the middle, bone body data, muscle data, muscle running data, fusion data, arrangement model data, analysis model data, and the like in FIGS. 1 and 9 can be considered. Further, as correction of the intermediate result, the CT image and the bone body data are superimposed and displayed on the display unit, and the bone body region is corrected by adding and deleting the bone body region with the mouse. The MR image and muscle data are displayed in a superimposed manner, and the muscle data is corrected by adding and deleting muscle regions with the mouse. In the fusion data, the inclination, rotation, and enlargement ratio of bone data or muscle data are corrected. In the fusion data and the arrangement model data, the direction and the size of the muscle running vector are corrected. Add and delete contacts, and change the contact area. Add, delete, or modify analysis conditions in analysis model data.

  In addition, a message for alerting the user may be displayed during the process. For example, a general muscle running vector has a symmetrical shape. Therefore, the calculated muscle running vectors are compared on the left and right, and a message is displayed if the left and right are largely separated. Thereby, the occurrence of an error can be avoided.

  In the above embodiment, the method for calculating the load on the bone body has been described, but the present invention is not limited to this. For example, it can be applied to other hard tissues such as teeth. Further, the present invention can be applied to calculating the load on an organ or blood vessel by obtaining the arrangement of the organ or blood vessel and the surrounding membrane instead of a hard tissue such as a bone body.

  Further, although several embodiments of the present invention have been described above, these embodiments have been described for easy understanding of the present invention, and are only specific examples for realizing the present invention. The present invention is not necessarily limited to the one having all the configurations of the above-described embodiments.

  The present invention is extremely useful as an image analysis technique for fusing a plurality of types of images extracted from radiation images and analyzing the fused images.

101 CT image
102 MR image
105 Bone body data
106 Muscle data
108 Bone + muscle fusion data
110 Placement model data
112 Analysis model data
114 analysis results
200,300,400,500,600 X-ray tube
201,301,401,501,601 X-ray source
202,302,402,502,602 Detector
203,303,403,503,603 Prop
204,304,404,504,604 Rotating device
205,305,405,505,605 Subject holding device
206,306,406,506,606 Rotation axis
207,307,407,507,607 Subject
208,308,408,508,608 Image analyzer
701,903 bone body data
702 muscle data
704 Bone + muscle fusion data
706,906 Placement model data
707,804,904 muscle running vector
801 slice statue
802 Muscle region
803 Area center
1201 Main memory
1202 CPU
1203 LCD
1204 HDD
1205 Input device
1206 Interface (I / F)
1207 Internal bus.

Claims (17)

An image analysis apparatus including a processing unit for processing image data,
The processor is
Create fusion data by fusing muscle data representing muscles and bone body data representing bones,
Using the fusion data, create arrangement model data indicating the arrangement of bones and muscles,
Set analysis conditions based on the arrangement model data,
An image analysis apparatus characterized by that. The image analysis apparatus according to claim 1,
The processing unit uses the fusion data to calculate the contact point between muscles and bones and a muscle running vector, thereby obtaining the arrangement model data.
An image analysis apparatus characterized by that. The image analysis apparatus according to claim 1,
The processor is
As the muscle data, an existing muscle running vector is used to create the fusion data by fusing with the bone body data.
An image analysis apparatus characterized by that. The image analysis apparatus according to claim 2,
The processor is
Extracting a muscle region using a muscle shape model, extracting the center of the muscle region on each slice image based on the muscle data, and calculating the muscle running vector by connecting the centers of the extracted regions;
An image analysis apparatus characterized by that. The image analysis apparatus according to claim 4,
The processor is
As the muscle shape model, a muscle shape model set in advance is used.
An image analysis apparatus characterized by that. The image analysis apparatus according to claim 1,
The processor is
Extracting the bone body data from a CT (Computed Tomography) image, extracting the muscle data from an MR (Magnetic Resonance Imaging) image,
An image analysis apparatus characterized by that. The image analysis apparatus according to claim 1,
The processor is
Extracting the bone body data from a CT (Computed Tomography) image, extracting the muscle data from an ultrasound image,
An image analysis apparatus characterized by that. The image analysis apparatus according to claim 1,
A display unit; and displaying the arrangement model data on the display unit.
An image analysis apparatus characterized by that. An image analysis program executed by an image analysis apparatus including a processing unit for processing image data,
The processing unit is
Create fusion data by fusing muscle data representing muscles and bone body data representing bones,
Using the fusion data, create arrangement model data indicating the arrangement of bones and muscles,
Operate to set analysis conditions based on the placement model data;
An image analysis program characterized by that. An image analysis program according to claim 9,
The processing unit is
By using the fusion data, the muscle and bone contact point and the muscle running vector are calculated to operate to create the arrangement model data.
An image analysis program characterized by that. An image analysis program according to claim 9,
The processing unit is
As the muscle data, an existing muscle running vector is used, and the bone body data is fused to create the fusion data.
An image analysis program characterized by that. An image analysis program according to claim 10,
The processing unit is
An operation for extracting a muscle region using a muscle shape model, extracting a center of the muscle region on each slice image based on the muscle data, and calculating the muscle running vector by connecting the centers of the extracted regions Let
An image analysis program characterized by that. An image analysis program according to claim 9,
The processing unit is
Extracting the bone body data from a CT (Computed Tomography) image, and operating to extract the muscle data from an ultrasound image,
An image analysis program characterized by that. An image capturing apparatus that captures and analyzes three-dimensional image data of a subject,
First three-dimensional image data, an input unit for inputting the second three-dimensional image data, muscle data from the first three-dimensional image data, and bone data from the second three-dimensional image data Extracting, creating the fusion data by fusing the extracted muscle data and the bone body data, using the fusion data to create the arrangement model data indicating the arrangement of the bone body and muscles, to the arrangement model data A processing unit for setting analysis conditions based on
An image pickup apparatus characterized by that. The image capturing apparatus according to claim 14,
The processor is
By using the fusion data, by calculating a muscle and bone contact point and a muscle running vector, the arrangement model data is obtained.
An image pickup apparatus characterized by that. The image capturing apparatus according to claim 14,
The first three-dimensional image data is CT image data, and the second three-dimensional image data is MR image data.
An image pickup apparatus characterized by that. The image capturing apparatus according to claim 14,
A display unit; and displaying the arrangement model data on the display unit.
An image pickup apparatus characterized by that.

Images