JP5264090B2 - Image processing device - Google Patents

Description

  The present invention relates to an image processing program and an image processing apparatus, and more particularly to an image processing program and an image processing apparatus for setting a three-dimensional region of interest for an image captured by a medical image capturing apparatus.

  The performance of diagnostic imaging apparatuses such as CT, MRI, and PET / SPECT is improved, and the amount of obtained image data tends to increase. In particular, in a multi-slice CT apparatus, high-definition three-dimensional data can be acquired in a short time, so that the burden on the interpreting physician is increasing.

  On the other hand, there is an increasing need for quantitative analysis as load information when interpreting the obtained image data. In the past, the goal was to depict blood vessels by CT angiography, but now it is required to provide a stenosis rate. In lung nodules, volume measurement is required, and evaluation of benign / malignant grade based on doubling time (time during which the volume of the lesion is doubled) by volume comparison between past image data and current image data. There is a request. In the brain perfusion examination, there is a demand to evaluate an analysis value such as CBF (cerebral blood flow) in an MCA (middle cerebral artery) dominant region.

  In addition, there is an increasing need for quantitative evaluation in terms of determining treatment policy. For example, in the case of obstructive arteriosclerosis (ASO), blood vessel morphology such as stenosis, occlusion, collateral circulation, etc. by angiography using an X-ray diagnostic device, spiral CT device, MRI device, etc. Diagnosis is made by observing (see Non-Patent Document 1). In addition, treatments such as exercise therapy, medical therapy, endovascular treatment, and surgical treatment (such as revascularization) are performed depending on the pathological condition. However, severe cases (Critical Limb Ischemia) in which none of the treatments are effective and there is no treatment method require amputation of the lower limbs, so early diagnosis and appropriate treatment are necessary.

  The effect of the treatment (recovery of blood flow and blood flow) can be evaluated to some extent by making full use of the diagnostic device as described above. However, it is possible to quantitatively grasp changes in muscle mass etc. to determine whether subsequent angiogenesis and collateral circulation lead to improvement of symptoms such as intermittent claudication, which is the original purpose It is difficult to do so.

  Moreover, in the treatment which needs to be performed continuously, such as drug treatment, it is desirable that the continuation of the treatment can be quantitatively determined each time. Furthermore, if the muscles that require treatment can be identified, efficient and effective treatment can be expected by administering drugs by intramuscular injection. Therefore, it is hoped that an evaluation method will be established that can establish treatment prospects and determine treatment strategies. It is rare. For this purpose, it is necessary to photograph a specific section with good reproducibility.

  In order to cope with this, in Patent Document 1, a plurality of cross sections are set apart from each other in a three-dimensional coordinate space, and the region of interest drawn on the image of each cross section is approximated by a polygon. For the cross section, a method is described in which polygons are obtained by interpolation, and a three-dimensional region of interest is set by stacking these polygons.

Patent Document 2 describes a method of extracting muscle and adipose tissue from a CT image by discriminating built-in fat and subcutaneous fat by threshold processing. Patent Document 3 describes a method for displaying the distribution of extracted adipose tissue.
JP-A-11-56841 JP 2003-339694 A JP-A-2005-192656 Imaging, Shujunsha, Vol. 23, no. 8, 2003

  However, in the method of setting the three-dimensional region of interest described in Patent Document 1, the accuracy greatly depends on the method of determining the cross section that is set apart from each other. Alternatively, there is a problem that a GUI (graphical user interface) that assists in setting an appropriate cross section is required.

  In the method of setting a three-dimensional region of interest described in Patent Document 1, if the organ is limited to a specific region such as the liver, the entire volume can be calculated and compared with past data with high accuracy. I can do it. However, when it is necessary to limit the range of tissues such as muscles for comparison, the range cannot be set with good reproducibility.

  In the method of setting a three-dimensional region of interest described in Patent Document 1, absolute quantitativeness is ensured if the same cross section can be accurately captured. However, when a tissue such as muscle is targeted as in ASO, the density of muscle fiber tissue in the cross section changes depending on the degree of muscle tension (how much force is applied). Have difficulty. Therefore, when the same tissue is extracted by the method of setting the three-dimensional region of interest described in Patent Document 1, the range cannot be set with good reproducibility even if the extraction methods described in Patent Documents 2 and 3 are used. There is an unsolved problem that absolute quantitativeness cannot be secured.

  The present invention has been made in view of the above circumstances, and provides an image processing apparatus and an image processing program capable of easily setting a three-dimensional region of interest for evaluating a desired target tissue. Objective.

  In order to solve the above-described problem, an image processing apparatus according to claim 1, an acquisition unit that acquires a three-dimensional image captured by a medical image capturing apparatus, and a predetermined tissue included in the three-dimensional image are three-dimensionally acquired. A reference region extracting means for extracting as a reference region, a region of interest setting means for setting a three-dimensional region of interest based on the extracted reference region, an output means for outputting information on the set region of interest, It is characterized by having.

  According to the image processing apparatus of claim 1, a 3D image captured by a medical image capturing apparatus is acquired, a specific tissue is extracted as a 3D reference area from the acquired 3D image, and the reference area is extracted. Based on this, a three-dimensional region of interest is set. Thereby, a three-dimensional region of interest can be easily set. In particular, when a bone is used as a reference region, the region of interest has high reproducibility. Therefore, an image useful for image diagnosis can be easily created by performing post-processing on the region of interest. In addition, quantitative evaluation such as left-right comparison and temporal comparison can be performed.

  In the image processing apparatus according to the first aspect, a feature amount of the extracted reference region may be calculated, and a three-dimensional region of interest may be set using the feature amount. The feature amount is at least one of a primary feature amount that is a geometric feature of the reference region and a secondary feature amount that is calculated based on the primary feature amount. As a result, a three-dimensional region of interest can be set with high reproducibility even in a region where the CT value, the pixel value, etc. are not characterized, or in a region where contraction changes.

  The image processing program according to claim 4 includes a step of acquiring a three-dimensional image photographed by a medical image photographing device, a step of extracting a predetermined tissue included in the three-dimensional image as a three-dimensional reference region, A calculation device is configured to execute a step of setting a three-dimensional region of interest based on the extracted reference region, and a step of outputting information on the set region of interest.

  According to the present invention, a three-dimensional region of interest for evaluating a desired target tissue can be easily set.

  The best mode for carrying out the present invention will be described below with reference to the accompanying drawings.

  FIG. 1 is a block diagram showing a configuration of an image processing apparatus 10 according to the present invention.

  The image processing apparatus 10 mainly provides volume data input means 1, three-dimensional region-of-interest setting means 100 for setting a three-dimensional region of interest from volume data, and information on the region of interest to an observer, another system, or the like. It is comprised with the information provision means 300. FIG.

  The volume data input means 1 acquires tomographic image data or three-dimensional volume data from a tomographic apparatus such as CT, MRI, PET / SPECT. The acquired tomographic image data or three-dimensional volume data is input to the three-dimensional region-of-interest setting means 100. For example, when volume data of the lower limbs is acquired, tissues such as bones, fats, muscles, and blood vessels are included. When the three-dimensional volume data is acquired, the three-dimensional volume data is output to the three-dimensional region-of-interest setting unit 100. When the tomographic image data is acquired, three-dimensional volume data is created by a reconstruction unit (not shown), and the created three-dimensional volume data is output to the three-dimensional region-of-interest setting unit 100.

  The three-dimensional region-of-interest setting unit 100 mainly sets a region of interest based on a reference region extraction unit 110 that extracts a reference region used as a reference for determining a region of interest, a reference region feature amount calculation unit 120, and the feature amount. The region of interest setting means 140 and a user interface 130 for an operator to input region setting parameters are configured.

  The reference area extraction unit 110 determines which tissue is used as the reference area. Tissues such as bones, fats, muscles, and blood vessels included in the data acquired by the volume data input means 1 are extracted and extracted using a technique such as threshold processing or a generally well-known region expansion method. An optimal organization is determined as a reference area among the organizations. As a method of determining the optimum organization as the reference region, the optimum organization may be automatically selected by protocolization. In this embodiment, the operator can select the optimum organization via the user interface 130. By selecting an organization, a reference area is determined. As the reference region, it is desirable to select a tissue with little temporal change due to treatment or the like.

  As a method for extracting the tissue, a threshold process, a region expansion method, or the like can be considered. In the case of bone, extraction can be performed by simple threshold processing in many cases, but the region expansion method may be more advantageous when the contrast difference is small. The scope of rights of the present invention is not limited by which method is used.

  The reference area feature quantity calculation means 120 calculates the feature quantity of the extracted reference area. Typical features include shape-related primary feature values such as contours, centroids, core lines, surfaces, normal vectors, direction vectors, and area profiles calculated based on the primary feature values. Secondary feature quantities. Further, examples of the concentration include a concentration gradient, an average concentration, and a concentration profile. In addition, since the appropriate feature amount and the method for calculating the feature amount differ depending on the organization selected as the reference region and the application using this method, it is possible to select a method for calculating the feature amount, or a plurality of methods. May be used.

The region-of-interest setting unit 140 determines a region of interest based on the feature amount calculated by the reference region feature amount calculation unit 120. For example, when the core line is calculated as the feature amount, the region of interest is set based on the start position on the core line, the end position, the start position and the end position, and boundary conditions along the core line. Boundary conditions, specifically, the surface of Cartesian to the core wire, a sphere, is to define in such a distance from the core wire is often preferred. However, in the case of the boundary conditions and rectangle, since the rotation of a plane Cartesian the core wire is necessary to consider the criteria for determining the rotation angle (e.g., the direction vector of the plane) must be provided is there.

  Hereinafter, the operation of the image processing apparatus 10 configured as described above will be described with reference to an example in which a CT image is acquired. The following processing is performed by the three-dimensional region-of-interest setting unit 100 after tomographic image data or three-dimensional volume data is input from the volume data input unit 1 to the three-dimensional region-of-interest setting unit 100. Note that the region of interest set by the following processing is output to the information providing unit 300.

<First Embodiment>
The first embodiment is a case where the present invention is applied to the lower limb and the lower limb is used as the region of interest.

  First, in the reference area extraction unit 110, the bone having the highest CT value and the easiest extraction from the volume data is determined as the reference area.

  When the bone is determined as the reference region, the reference region feature amount calculation unit 120 calculates the core line as the extracted reference region feature amount. In the case of a bone of the lower limb such as a tibia, the core wire is used because it is a long tissue in the body axis direction. The core wire is often used for analysis of blood vessels and large intestines, which are long tissues like the bones of the lower limbs.

  In each cross section, a center point of each cross section is defined, and a line connecting the center points is calculated as a core line. For example, the center point of each cross section can be defined by the center point of a circle obtained by approximating a contour, which is a primary feature amount, as a circle, the center of gravity of the cross section, and the like. The center of gravity is defined by calculating the extracted internal region of the bone as a primary feature amount and calculating the center of gravity of the internal region of the bone as a secondary feature amount. However, if a joint or the like is included, there is a possibility that the error becomes large, so that it can be calculated using the following method.

  FIG. 2 shows a method of calculating a core line for a bone BN including joints J1 and J2.

In FIG. 2 (a), AX is the axis defined by a method such as described above, the surface PP1, PP2 denotes a surface that Cartesian axially AX. The areas P1 and P2 of the bone regions on the surfaces PP1 and PP2 can be calculated by a known method. The calculated area profile F is shown in FIG. The horizontal axis of the area profile F is the position P on the axis AX.

  By providing a threshold value in the area profile F, the joints J1 and J2 can be excluded. In a state where the joints J1 and J2 are excluded, a center point is defined in each cross section, and a line connecting the center points in each cross section is calculated as a core line. It should be noted that feature quantities other than the area profile can be used. Convenient examples include density information such as average density and density histogram.

  However, in the method shown in FIG. 2, the first analysis cross section (for example, the planes PP1 and PP2 in FIG. 2) is an axial position image, and the position of the actual patient is the tomographic cross section and the patient body position. It depends on the relative positional relationship. For example, FIGS. 3A and 3B show bones BN including joints J1 and J2 extracted from the volume data obtained by the first imaging, and joint J1 extracted from the volume data obtained by the second imaging. Bone BN 'including', J2 'is shown in absolute space, and axes AX and AX' are defined, respectively. Such a situation can be considered when the positions of the two shots are different. If both axes AX and AX 'are obtained with high accuracy, the region of interest can be set accurately. However, depending on the accuracy of the data, the accuracy may be insufficient.

  In such a case, as shown in FIG. 3C, the deviation of the axis is determined based on the mismatch of the area profiles. If it is determined that the axis is shifted, the accuracy can be improved by redefining the axis. Further, by knowing the axis deviation, by combining rotation about the axis AX (direction (d) α) and translation in relation to the axis AX ′, the axis AX and the axis AX ′ are combined. Can be matched. In addition, when using two volume data with different imaging ranges, or when using two volume data acquired by changing the patient's body position, an area profile or the like can be used for the axis AX defined by the tissue. The amount of deviation along the line can be obtained.

  Next, a boundary condition for setting the region of interest is input to the region of interest setting means 140 by the user interface 130. Then, the region of interest setting unit 140 sets the region of interest using the core line that is the secondary feature amount calculated by the reference region feature amount calculating unit 120 and the input boundary condition. Consider the case where the start position and end position on the core line and the distance from the core line are set as boundary conditions. When the start position and the end position are the distance L from the center of gravity G and the distance from the core line is r, a cylindrical area having a radius r and a length of 2L centered on the core line is set as the region of interest.

  For example, as illustrated in FIG. 3A, a cylindrical region of interest is set with a radius r with respect to the reference region. A conical region of interest with varying radius r is also shown. Further, as illustrated in FIG. 3B, when the core wire of the reference region is curved, the region of interest is set to be curved.

Here, when limiting the range (start position, end position, etc.) for setting the threshold value and the region of interest on the user interface 130 based on the feature amount of the reference area such as a core line, a general GUI (graphical) · by using the user interface), Figure 4 intuitive operation is possible, when the area profile (see FIG. 2 (b)) the feature quantity of a plane Cartesian core wire shown in FIG. 2 An example of the GUI is shown. When the operator specifies the range in the axial direction of the core wire, the range is easily set by operating the slide bars 1 and 2 and when the range is specified by the feature amount F, the slide bar 3 is operated. can do.

  When a region of interest of a cylinder having a radius r centered on the core line is set, the boundary condition is set to a distance r from the core line only, and the start position and end position of the cylinder are based on an area profile that is a secondary feature amount. Can be set. For example, as shown in FIG. 2, it is possible to set the positions obtained as joint positions J1 and J2 in the area profile as the start position and end position of the cylinder. In this case, a region of interest is set based on the core line, the area profile, and the boundary condition, which are secondary feature amounts.

  According to the present embodiment, since the feature amount of the tissue with little change due to external factors such as a change due to treatment, a change in body position, and a muscle contraction is used as a reference, the distance between the knee and the ankle regardless of the external factor. A specific same region such as a lower limb region can be easily set as a region of interest with high reproducibility.

  In the present embodiment, the same region of interest can always be set, but some post-processing may be performed on the set region of interest. As post-processing, an analysis target is extracted based on the region of interest and an evaluation value is calculated as follows.

  FIG. 5 is a block diagram showing a configuration of the image processing apparatus 10a capable of calculating an evaluation value. In addition, the same code | symbol is attached | subjected to the part same as FIG. 1, and description is abbreviate | omitted.

  The image processing apparatus 10a mainly includes a volume data input unit 1, a three-dimensional region-of-interest setting unit 100 that sets a three-dimensional region of interest from volume data, and an analysis target tissue feature amount calculation unit that calculates an evaluation value of the analysis region. 200 and an information providing unit 310 that provides an evaluation value calculated by the analysis target tissue feature amount calculating unit 200 to an observer or another system.

  The analysis target tissue feature amount calculation unit 200 mainly includes an analysis target tissue extraction unit 210 that extracts an analysis target tissue based on the region of interest set by the region of interest setting unit 140, and an evaluation that calculates an evaluation value of the analysis target tissue. It is comprised with the value calculation means 220. FIG.

  For example, when the reference region is set as the bone of the lower limb and the region of interest is a region on a cylinder centered on the bone core line, the muscle region existing in the region of interest is the analysis target, and the muscle volume is the evaluation value. If you want to.

  Thereby, left and right comparison of limbs and quantitative comparison of changes over time are possible. In the case of performing a quantitative comparison, an objective evaluation is possible by defining the range of the muscle mass of the left and right lower limbs based on the area profile. Further, in the follow-up of treatment, etc., even if the cross sections do not exactly match, a quantitative comparison can be made within an arbitrary range. As a result, ASO, osteoporosis, etc. can be diagnosed and quantitative evaluation of therapeutic effects can be performed.

<Second Embodiment>
In the first embodiment, a core line is calculated as a feature amount of a bone of a lower limb that is a reference region, and a cylindrical region around the bone of the lower limb is set as a region of interest. It is not limited.

  In the second embodiment, the heart is set as a region of interest. In addition, the same code | symbol is attached | subjected to the part same as 1st Embodiment, and description is abbreviate | omitted.

  First, the reference area extracting means 110 extracts the heart lumen as a reference area.

  When the heart lumen is extracted as the reference region, the reference region feature amount calculation unit 120 calculates the center of gravity and the direction vector, which are secondary feature amounts, as the feature amount. The direction vector can be derived from the form and derived from the concentration. For example, when the sum of vectors connecting the center of gravity and each voxel is used as the direction vector derived from the form, the direction vector derived from the form is considered to correspond to the heart axis.

  When an elliptical sphere centered on the center of gravity and whose major axis coincides with the central axis is set as the boundary condition, the region-of-interest setting unit 140 uses the center of gravity as the secondary feature amount and the boundary condition. Based on the above, the heart is set as the region of interest. In addition to the elliptical sphere, a region in which the shape and the center of gravity are defined in advance may be used as the boundary condition.

  As a result, when creating a three-dimensional image of the heart, it is possible to easily delete excess peripheral regions.

  According to the present embodiment, an image useful for image diagnosis can be easily created. Further, by performing the same processing on a plurality of images, it becomes easy to compare desired tissues such as the heart.

  Further, according to the present embodiment, a desired tissue such as the heart and brain can be extracted under the same conditions by extracting a three-dimensional region of interest and removing other regions.

  In this embodiment, an elliptic sphere is set as the boundary condition. However, even when a sphere, a cube, or the like is set as the boundary condition, it can be handled in the same manner as in the case of the elliptic sphere. That is, the boundary condition may be set so that the center of the sphere and the center of gravity of the cube coincide with the center of gravity that is the secondary feature amount.

<Third Embodiment>
In the first embodiment, a cylindrical region around the bone of the lower limb is set as the region of interest, and in the second embodiment, the heart is set as the region of interest. It is not something that can be done.

  In the third embodiment, the large intestine is set as a region of interest. In addition, the same code | symbol is attached | subjected to the part same as 1st Embodiment, and description is abbreviate | omitted.

  First, the reference area extraction unit 110 extracts the large intestine, which is a luminal organ, as a reference area.

  When the large intestine is extracted as a reference region, the reference region feature quantity calculation means 120 first calculates a contour that is a primary feature quantity (indicating the surface of the intestinal gas), and calculates a core line based on the contour. Note that the method described in the first embodiment can be used as a method of calculating the core wire.

  As shown in FIG. 6, in the region-of-interest setting means 130, a boundary is set on a straight line H connecting a point (for example, the center of gravity G) where a core line is defined and a point Q indicating an outline. The boundary condition is a distance k from the contour, and if the distance from the contour is positive, the boundary is directed from the contour to the colon wall (or may be directed to the straight line H). k regions of interest are set. In this case, the region of interest is set based on the contour that is the primary feature quantity, the distance in the straight line H direction that is the secondary feature quantity, the boundary condition based on the density profile in the H direction, and the like.

  Note that substantially similar regions of interest can be set based on the contours and boundary conditions that are primary feature amounts. If the distance k from the contour is set as a boundary condition, a region of interest having a width of 2k centering on the contour is set. If the distance from the contour is positive, the region of interest having a width k is set outside the contour when the boundary is from the contour toward the colon wall.

  According to the present embodiment, it is possible to set only the vicinity of the large intestine wall as a region of interest, and it is easy to create an image useful for tumor wall infiltration or polyp evaluation. In addition, a useful image can be created by image diagnosis by performing post-processing to superimpose the set region of interest with an image taken with PET.

  According to the above embodiment, the feature amount is calculated from the extracted tissue, and the three-dimensional region of interest is set on the basis of the feature amount. It can be set highly.

  Further, according to the above embodiment, since the reproducibility of the region of interest is high, quantitative evaluation such as left-right comparison and temporal comparison, image comparison, and the like can be performed. Therefore, therapeutic effects such as drug administration are clarified, and diagnostic effects such as easy treatment planning and treatment strategy can be obtained.

  In the above embodiment, a three-dimensional region of interest with high reproducibility is set, but various application examples of the present invention based on the set region of interest can be considered.

  For example, there may be a case where the lung field is used as a reference region and the feature value to be obtained is used as a contour. When the contour is calculated from a plurality of slice images, it indicates a three-dimensional surface of the lung field. The region of interest setting means can set the region of interest using the distance from the contour. For example, when the distance in the normal direction is used as the distance from the contour, the rib region can be set as the region of interest. By using an image of a rib region, which is a region of interest set based on an image photographed by a CT apparatus, on a PET image aligned with an image photographed by a CT apparatus, the rib area on the PET image It is thought that it is useful for searching metastatic tumors only for cancer.

  In some cases, the skull is used as a reference region, and the feature amount to be obtained is used as a surface inside the skull. The brain surface can be set as the region of interest by setting the range of the predetermined distance inside the surface inside the skull to be the region of interest. Useful for imaging of the brain surface by using the image of the brain surface, which is the region of interest set based on the image captured by the CT device, for the MRI image aligned with the image captured by the CT device it is conceivable that.

  In addition, when applying to a dental region, teeth are extracted from a CT image, a core line of teeth is obtained, and a rectangular region including the core line is set as a region of interest, so that a panoramic image with high reproducibility can be easily obtained. Can be created.

  Further, in the present embodiment, the present invention is performed on an image photographed by a CT apparatus, but is not limited to an image photographed by a CT apparatus, but various types of images photographed by a tomography apparatus such as an MRI apparatus. Images can be used.

1 is a block diagram showing a configuration of an image processing apparatus 10 according to the present invention. It is explanatory drawing about the case where the bone of the leg of the said image processing apparatus 10 is made into the reference area. It is explanatory drawing about the case where the bone of the leg of the said image processing apparatus 10 is made into the reference area. It is explanatory drawing explaining the method to set the region of interest of the said image processing apparatus. It is a figure which shows an example of GUI which sets a threshold value etc. when the bone of the leg of the said image processing apparatus 10 is made into the reference area. It is a block diagram which shows the structure of the image processing apparatus 10a which can calculate an evaluation value. It is explanatory drawing about the case where the large intestine of the said image processing apparatus 10 is made into the reference area.

Explanation of symbols

1: volume data input means 10, 10a: image processing apparatus, 100: three-dimensional region of interest setting means, 110: reference region extraction means, 120: reference region feature amount calculation means, 130: user interface, 140: region of interest Setting means 200: Analysis target tissue feature amount calculation means 210: Analysis target tissue extraction means 220: Evaluation value calculation means 300, 310: Information providing means

Claims (5)

Acquisition means for acquiring a three-dimensional image obtained by imaging a subject with a medical imaging apparatus;
A reference region extracting means for extracting a bone region, which is a region where the bone of the subject included in the three-dimensional image is imaged, as a three-dimensional reference region;
Feature amount calculating means for calculating a core line of the bone region;
Input means for receiving an input of a distance from the core wire;
Distance from the input core, and, based on the core of the bone region, and the region of interest setting means for setting the three-dimensional region of interest,
An analysis target tissue extracting means for extracting a muscle region that is an area in which the muscle of the subject included in the region of interest is imaged;
An evaluation value calculation means for calculating the evaluation value of the muscle region, with the muscle region as an analysis target,
The feature amount calculating means calculates an area profile of the bone region in each cross section orthogonal to the axis on the bone region, and sets a threshold value for the area profile, thereby excluding the bone joint of the subject Then, a line connecting the center points in each cross section is calculated as a core line.
An image processing apparatus. Acquisition means for acquiring a plurality of three-dimensional images obtained by imaging a subject with a medical imaging apparatus;
A reference area extracting means for extracting, as a three-dimensional reference area, a bone area that is included in each of the plurality of three-dimensional images and is an area where the bone of the subject is imaged;
When it is determined that the axis set on the bone region included in each of the three-dimensional images is shifted, the rotation around one axis and the parallel movement are combined with respect to one axis. The feature amount calculation means for calculating the feature amount of each of the bone regions by matching the one axis with the other axis ,
An input means for receiving an input of the boundary condition defined based on the feature amount, which is a boundary condition used for setting a three-dimensional region of interest including each of the bone regions;
The inputted boundary conditions, and on the basis of the feature quantity of the bone region, and the region of interest setting means for setting the three-dimensional region of interest,
An analysis target tissue extracting means for extracting a muscle region that is an area in which the muscle of the subject included in the region of interest is imaged;
Evaluation value calculation means for calculating the evaluation value of the muscle region, with the muscle region as an analysis target;
An image processing apparatus comprising: The feature amount calculating means calculates a core line of the bone region as the feature amount,
The input means receives an input of a distance from the core wire as the boundary condition;
The image processing apparatus according to claim 2. The feature amount calculating means calculates an area profile of the bone region in each cross section orthogonal to the axis on the bone region, and sets a threshold value for the area profile, thereby excluding the bone joint of the subject Then, a line connecting the center points in each cross section is calculated as a core line.
The image processing apparatus according to claim 3. The evaluation value calculating means calculates the volume of the muscle region included in the region of interest as the evaluation value;
The image processing apparatus according to claim 1 , wherein the image processing apparatus is an image processing apparatus.

Images