JP2003514600A - Method and apparatus for reformatting a tubular volumetric object - Google Patents

Abstract

(57) Abstract: A volumetric image of a blood vessel is acquired by an imaging system, and the volumetric image is reformatted into linear data by a computer-readable medium storing the present method, apparatus, and program. . More specifically, for volumetric data acquired in an imaging modality, separating the structure, such as separating blood vessels from the rest of the tissue, and stretching the structure to obtain a linear morphology corresponding to the structure Reformatting. Reformatting the volumetric data corresponding to the structure by roughly segmenting the structure from adjacent components by segmenting the structure from adjacent components using an appropriate CT intensity threshold; Expanding the segmented structure, determining the axis of the structure along the entire length of the structure, determining a plane that is an orthogonal cross-section of the structure at a selected location along the axis of the structure, Reforming the volumetric data along with generating CT data after reformatting. The reformatted CT data corresponds to the data that causes the CT imaging system to reconstruct if the structure is straightened and the CT imaging system assumes that the image of the straightened structure has been acquired directly by the CT imaging system. .

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates generally to a reformatting of data acquired in various imaging modalities.
More specifically, the present invention relates to a
Reformatting such data to facilitate accurate evaluation and estimation
Related. In a specific embodiment, the present invention relates to the vasculature of an artery or a specific organ.
Three-dimensional volumetric to help quantify the stenosis affecting the blood vessels
・ Reformat computed tomography (CT) data into three-dimensional cross-sectional data
About things. BACKGROUND OF THE INVENTION Vascular disease is one of the main causes of infarction and ischemia. Several imaging modalities
Provide a method for assessing atherosclerosis. blood
Angiog is one of the clinically accepted modalities for assessing duct stenosis
There is luffy. In angiography, a catheter is inserted into the vessel to be diagnosed.
You. Next, a contrast agent opaque to radiation is injected into the blood vessel to be diagnosed. Proceed with the injection
While the affected organ or tissue area (and more particularly, the blood vessel to be diagnosed)
The X-ray projection image for the image intensifier and the digital detector
Collected using Ray. The acquired X-ray projection image is rear-mounted on an adjacent monitor.
Displayed in real time. Diameter of blood vessel is estimated from X-ray projection image of blood vessel to be diagnosed
This leads to a quantitative estimation of the amount of vascular stenosis. [0003] As a problem related to angiography of blood vessels, angiography is extremely
May be invasive. In addition, the vasculature image of the organ obtained from the angio image
The images are typically acquired from a two-dimensional projection of the three-dimensional vasculature.
, Bones, overlapping vessels, twisting of target vessels, etc. Image
Adjust the positioning of the phasing system prior to contrast injection to minimize these effects.
However, some overlap of adjacent structures is inevitable. [0004] Ultrasound imaging can also be used to evaluate structural features such as vessel dimensions.
Wear. Dynamic images are acquired by intravascular ultrasound (IVUS)
It has become. As with angiographic catheter operation, IVUS technology
The law is also very invasive. When using non-invasive techniques, vascular disease can be reliably
To evaluate with good repeatability, place the probe with reference to bones such as ribs.
Required, the probe approaching the target vessel, and the need for a skilled operator.
There is a serious problem due to the necessity. Recently, magnetic resonance imaging (MRI), X-ray computed tomography,
(CT) and other volumes created to quantify blood vessel stenosis
There is interest in using trick data. In MRI, the magnetic polarization of the tissue to be imaged and the polarization of the polarized atoms (eg, electromagnetic waves)
Field excitation) to produce a three-dimensional (3D) model of the structure or organ of interest.
Collecting volumetric data that will be reconfigured to create Dell
. Using MR images, the vasculature of an organ is segmented from surrounding tissue (segme
ntation), and typically, the created image is visually reviewed by a skilled reader
This can determine the amount of stenosis. MRI scan to create a medical image
Takes a long time to collect volumetric data corresponding to the region of interest
Sometimes. This is often because data collection is often associated with physiological functions such as respiration.
This is because the gate is controlled for adaptation. [0007] X-ray computed tomography (CT) has also been evaluated for stenosis in several vascular structures.
It has been considered one of the possible options to value. However, accurate CT
To keep the patient being imaged stationary during data acquisition,
A drawback arises from the need to be scientifically consistent. Stillness
Otherwise, artifacts will occur in the reconstructed image. Breathing exercise
It is usually the largest cause of gender mismatch. The effects of this breathing
During patient acquisition, this can be limited by having the patient stop breathing. [0008] Using a computed tomography (CT) imaging system, volume
Segmentation of blood vessels from adjacent tissue by metric reconstruction (segmentation)
can do. However, delineation of the structure of interest using the reconstructed image
It is often inconvenient for studying architectural features. The result is all
As a body, it is in a state where precision is low for performing an evaluation. [0009] A volumetric image of an object obtained by processing digital data and displaying the image is displayed.
Where the processed data is referred to as volumetric
Reconstructed data, or `` volumetric image data
) ". However, any imaging modality (eg, MRI, CT,
Ultrasound, etc.) and then reconstructed
Quantification of stenosis remains difficult. In general, the target vessel is a well-known reconstruction
Not at right angles to the linear grid used in the algorithm. Therefore,
Reformat volumetric reconstruction data to reduce stenosis in blood vessels and other structures
There is a need for a method to improve quantification. More generally, volumetric
Reformat data to accurately assess structural features represented in images
In order to achieve this, it is desirable to provide images that are oriented in a more appropriate direction. SUMMARY OF THE INVENTION [0010] The present invention provides a method for generating a volumetric image in one or more two-dimensional directions in a selected plane direction.
(2D) A method, apparatus, and professional for reformatting a display.
A computer-readable medium having stored thereon a program is provided. In a first aspect, the invention provides a method for imaging a volumetric structure, comprising:
Methods, apparatus, systems and software are provided. Method according to this first aspect
Is based on volumetric image data representing the imaged object.
Calculates quantitative information representing the structural features of the volumetric structure in the object
Including The method may further include volumizing within a selected plane based on the quantitative information.
Including creating cross-sectional image data that specifies
No. [0012] In a second aspect, the present invention provides a method for imaging a volumetric structure.
And other methods, devices, systems and software. In this second aspect
The method uses a volumetric representation of the volumetric structure in the imaged object.
Includes collecting trick data. The method according to this second aspect further comprises:
Selected pairs based on quantitative information representing the structural features of the volumetric structure
In the corresponding planes, each specifies a cross-sectional image of the volumetric structure.
Generating a plurality of image data sets. [0013] These and other aspects and advantages of the present invention are described in the accompanying figures.
The following detailed description of the preferred embodiments taken in connection with aspects
It will be easier and easier to understand. DETAILED DESCRIPTION OF THE INVENTION A typical radiographic system or computed tomography (CT) image
A zing system 10 is shown in FIG. An imaging system of the type disclosed in FIG.
For the stem, see Avash C. et al. Kak and Malcolm Slan
"Principles of Computerized Tom"
Graphic Imaging "(IEEE Press, 1988: 12).
Pp. 6-132). As shown in FIG.
System 10 includes an object 14 (eg, a patient) positioned on a support 16 such as a bed.
) For transmitting a main signal to the source. Main signal is target
The light passes through the object 14 and the support 16 and is detected by the detector array 18. Detector
The detection of the main signal by the array 18 is controlled by the data acquisition component 19.
Have been. The CT imaging system 10 shown in FIG. 1 typically has a so-called third
A generation CT imaging system. In such a third generation system,
Both the source 12 and the detector array 18 can maintain mutual alignment while maintaining an established focusing alignment.
It is controlled by the common control device 20 to move in a tandem state.
The detector array 18 includes a collimating plate that is focused and aligned with the X-ray source.
It may or may not include these plates. Focusing alignment (focal
alignment means that the collimating plate of the detector array 18 is the source
This means that it points in the direction of twelve. The control device 20 typically has a CT
Based on the instructions issued by the system computer 22, the source 12 and the detector array
The on / off state and the movement of the b18 are controlled. Compile for CT system
The computer 22 also controls the data collection component 19. After detecting the X-ray signal, the data collection component 19 detects the detected X-ray intensity signal.
To digital data for supply to the CT system computer 22.
Replace. The computer 22 for the CT system then follows the well-known techniques.
This digital data is processed, and the processed digital data is stored in a system memory.
The digital data stored in the memory 24 and further processed is displayed on the display device 26.
Let me show you. The system memory 24 is a local memory resident inside the computer 22.
-When a memory is used, or when a magnetic disk is placed inside or outside the computer 22
In some cases, a large-capacity storage medium such as a disk. FIG. 2 illustrates the merging system 34 of the present invention. Shown in FIG.
Imaging system 34 is a third generation CT imaging system,
Used as an example of an imaging system embodying the present invention.
is there. More generally, the invention relates to three-dimensional (ie, CT, MRI, ultrasound, etc.)
Any volume tomographic system or volume level
Applicable to a daring system. [0018] The imaging system of the present invention was implemented in a CT imaging system.
Is not limited to 3rd generation CT imaging systems
Special attention should be paid. The present invention provides, as an alternative, a so-called fourth generation CT image.
It can also be implemented in a streaming system. 3rd generation CT and 4th generation CT
In either case, the CT system is in a mode to collect the appropriate three-dimensional data.
With either a fan-beam system or a cone-beam system
You can also. Throughout the following description, reference will be made to the imaging system 34 shown in FIG.
In order to achieve this, “X-ray imaging system”, “X-ray photography system”,
Computerized Tomography Imaging System ”,“ CT Imaging System ”
"Is used interchangeably (synonymously). In the x-ray imaging system 34 according to the invention shown in FIG.
Similar parts corresponding to the prior art X-ray imaging system 10 shown in FIG.
For the sake of clarity, in this specification, similar parts are not repeated.
Not listed. In the description of the merging system 34 of the present invention, a computer for a CT system will be described.
22, system memory 24 (stores data reformatting program 25)
) As well as the display 26. When used in the present invention,
The computer 22 and the memory 24 cooperate with each other to include an image analysis device 36. table
The indicating device 26 may optionally be considered as part of the device 36. These
The imaging device may alternatively be separate from the imaging system 34.
For example, imaging by a computer network (not shown)
-Can be coupled with the system 34. More specifically, the description of the present invention will begin with an imaging system 34 comprising an X-ray source.
12. Object 1 using detector array 18 and data acquisition component 19
4 volumetric (three-dimensional, or “3D”) projection data collected
Start with Volumetric data using imaging systems
Collecting is well known in the art. Then, this projection data
The data is obtained by appropriate reconstruction procedures, as is well understood in the art.
D image data is reconstructed. Here, “projection data” refers to an object from a plurality of different directions.
Data that represents the relative transmission of the radiating energy wave or flux.
You. If the projection data is data corresponding to irradiation in many different directions,
The projection data can be reconstructed into an image of the object by applying an appropriate reconstruction procedure.
Wear. "Volumetric projection data"
It corresponds to the point occupying the three-dimensional area of the imaged object, and creates a three-dimensional image.
Is the two-dimensional projection data from which the data is derived. After collecting the volumetric data, the CT system computer 22
The data reformatting program 25 is read from the system memory 24.
Subsequently, as shown in FIG. 3, the volumetric data is re-formatted according to the present invention.
Format program 25, and the system computer 22
The data reformatting program 25 of the invention is executed. The data reprocessing of the present invention
The output obtained from the format program 25 is
-It becomes volumetric data after mat. Volume after reformatting
The rick data is called cross-sectional linear data. “Sectional image data” is a sectional image, ie, a cubic image
This is data that specifies a two-dimensional image of a cross section of the original object. Cross section image data
Is `` rectilinear data '' or `` linear image data (rectiline
ar image data) ". A set of data allows display or reconstruction from the data set
One image at that time is "specified". `` Image data set
Means a set of such data designating an image. image data
Can specify one two-dimensional image or one three-dimensional image.
. FIG. 4 shows that the data reformatting program 25 of the present invention uses the CT system of the present invention.
Reforming process 4 of the present invention instructing the system computer 22 to execute.
FIG. The data reformatting process 40 is described with reference to FIGS.
This will be described in further detail with reference to the drawings. Referring now to FIG. 4, first, volumetric image data is obtained by CT
Reconstructed using standard techniques suitable for a given style, such as
After the acquisition by the imaging method of
Data reformatting program 25. Volumetric data
, Must support images of complex objects, including longitudinally extending structures inside
There is. Each such structure includes one or more spatially curved sections.
Can be Such a structure extending in the longitudinal direction includes, for example, an arbitrary tubular structure.
There is structure. In particular, such structures should be fluid-carrying structures such as animal blood vessels
There is. In one specific embodiment, such a longitudinally extending structure is a patient (human)
It may be a blood vessel. In this latter case, this complex structure provides more blood supply.
Adjacent to the blood vessel, such as the tissue of the receiving organ (ie, "adjacent
ng) ") may include tissue. In the exemplary embodiment to be described, the book
The invention applies in the specific context of blood vessel imaging. The present invention
In order to make examination and evaluation of structures such as worm-like structures inside the
Extensive range where it is desirable to reformat existing volumetric image data
It is within the skill of the art to provide equivalent efficacy and advantages for various imaging problems.
You will understand. As used herein, “volumetric structure”
Is contained within a larger object, possibly as part or all of it,
A physical structure extending in a spatial three-dimensional direction. Therefore, a three-dimensional object
The body is a volumetric structure and may contain volumetric structures
However, the shadow on the two-dimensional surface extends only in two dimensions,
It is not structured. The “section” of a volumetric structure is defined as
Is a two-dimensional view of that structure. The “structural features” of a volumetric structure are quantitative dimensions and properties of the structure
Means For example, a structural feature is a three-dimensional area occupied by the structure.
The volume of the area, the length of the characteristic part of the structure, the direction in which the structure mainly extends
The major axis, the minor axis across the major axis of the structure, the surface area of the structure, etc.
However, these are presented for illustrative purposes and are not intended to be limiting.
Absent. A "vermiform" structure is a longitudinal axis (sometimes a curved axis)
Along at least several times the largest diameter of the structure perpendicular to the longitudinal axis
Volumetric structure like This part of the insect-like structure
Insect-like, although the part itself extends less than several times the maximum diameter of the structure
Sometimes treated as a structure. A "tubular" structure is substantially along a longitudinal axis.
Worm-like structure that defines an internal gap that extends. A tubular structure
The interstitial space (ie, “lumen”) in the structure is typically, but not always,
Its tubular structure is topologically equivalent to a torus (“isometric and isomorphic”).
ly isomorphic) "), it is a connected area. The reformatting process 40 is represented in the volumetric image data and
For creating a mask for a region of interest corresponding to a portion of the blood vessel extending in the longitudinal direction.
The operation proceeds to operation 402. The preferred procedure for making this mask is
It is desirable to include high value segment splitting operation and dilation operation
. Next, a process 40 creates a longitudinal axis curve for this vessel portion in operation 402.
The operation proceeds to operation 404, which is an operation determined from the volumetric mask that has been set. More specifically, each slice containing a vessel along the z axis of the reconstructed volume
When the center represented by the x and y coordinates is determined, the x and y coordinates become a function of z.
The least squares fit to the x, y coordinates is expanded. Basically, x is a blood vessel
Is a function of z with respect to the x coordinate of the center, and similarly y is with respect to the y coordinate of the vessel center.
It is a function of z. For each of these data, a polynomial is calculated using the least squares technique.
Fitting creates a smooth display for the vessel axis. As a result of the above processing, the center of the blood vessel
An indication of the line (ie the longitudinal axis) is obtained. Then, at right angles to this centerline
The surface is calculated and the data on this oblique surface is examined. Calculated for center line
The pixel value is estimated on the right-angled surface. In operation 406, image data for oblique cutting along the longitudinal axis curve is obtained.
Creating. Here, “oblique cutting” refers to volumetric image data.
Selected not parallel to the 3D linear grid for display on
Means a cross-sectional image of a structure extending in the longitudinal direction in a plane oriented in a different direction. Described in detail
In a specific embodiment, the diagonal cut is made with respect to the longitudinal axis curve.
Thus, the right angle is selected at each position along this curve.
When selecting two or more oblique cuts, different oblique cuts are
In different directions with respect to the The “direction” of a cross-section is defined as the cross-section of a set of predetermined orthogonal axes in three dimensions.
This is the angular direction of the surface. The orientation of the section is expressed in various equivalent ways.
Can be For example, this direction is determined by the coefficient of a linear equation in three dimensions.
That is, it can be expressed by Ax + By + Cz + D = 0. Otherwise
The direction can also be expressed by a unit normal vector for the surface. One surface is “selected” by specifying its position and orientation with respect to that surface
. Given a three-dimensional vector, the common unit normal vector parallel to this vector
It is a well-known geometric fact that one can specify a family of parallel planes with
. Therefore, for a certain surface, for example, (1) one point specified on the surface is specified.
(2) By selecting a line that is perpendicular to the plane,
Can be. It is preferable that the specified line also passes through the specified point.
Optional arrangement. One skilled in the art will understand that this oblique cutting can alternatively be
It will be appreciated that other orientations can be selected based on the axis of the structure
. Further, the process 40 may include reformatting on the oblique cut (s)
Operation 410 to perform additional processing on the volumetric image data of the
desirable. For example, diagonal cutting provided by reformatted image data
Is an example of a frequently used post-processing operation. Image display
Is preferred for evaluation of imaged structures, especially for medical images.
It is well-known as a post-treatment style. Reforms to create quantitative grades
Other post-processing operations, such as a calculation process for image data after matting, are also possible. The operations 402 to 406 of FIG. 4 will now be described in detail with reference to FIGS.
I will post it. FIG. 5 is a flowchart illustrating an exemplary process for performing operation 402 of FIG.
. In FIG. 5, in operation 512, a so-called seed pixel is selected. Here,
`` Seed pixel '' refers to the area of interest in volumetric image data.
A pixel that is surely included in the area. For example, assess stenosis
In the context of blood vessel imaging to
Pixels that are known reliably can be seed pixels. The seed pixel is the first
Operator input, such as cursor instructions for images from linear grid image data
It is desirable to determine from force. Alternatively, the seed pixel is the first volume
It can also be specified in advance from preliminary analysis on the click image data. Using the seed pixel, in operation 522, a request is made to connect the pixel.
Rough thresholding procedure for volumetric image data
Perform a segment split. The region of interest is roughly divided by the operation 522.
Initial operation, but this operation should be designed to eliminate border pixels.
Is desirable. Next, from the result of the rough segment division, select in operation 532
Using morphological operations that use the defined connectivity criteria,
Perform finer classification. Next, in the dilation transformation operation 542, the region of interest (
For example, the so-called mask of the lumen of a blood vessel is converted into a rough segment morphology.
-Created by expansion transformation. The seed pixel in operation 512 is that the pixel of the first image data is “segmented
Segmented "(that is, divided into pixels of the region of interest and pixels of other regions).
Is the starting point of the processing. The region of interest (eg, vascular lumen)
In the threshold segment division operation 522, it is roughly divided. Inside the blood vessels
, The remaining tissue using a CT value threshold selected based on the pixel distribution of the first image
Are roughly specified by segmenting blood vessels from Generally, seg
The desired result from the segmentation operation 522 is to enhance the region of pixels with high intensity.
Identifying from regions of relatively lower degrees of pixels or vice versa
Will be done. The result of segment segmentation operation 522 is roughly
This portion is referred to as a “seed pixel segment”.
As part of this process, the connectivity criterion is set for pixels that meet the threshold criterion.
Apply. As is well known to those skilled in the art, this connectivity criterion allows the seed pixel seg
Pixels that are not adjacent to the segment are not included in the output of the segmentation process.
You. The value of this threshold value maintains a continuous path in the length direction of the imaged portion of the blood vessel.
Segment blood vessels from adjacent components (eg, organ tissue)
It is desirable for the user to make a selection. Further details
Specifically, the points in the blood vessel and the threshold value of the threshold are determined by the remaining tissue (eg, by the artery).
Blood vessels (eg, vasculature)
Pulse, etc.) are selected so as to be appropriately cut out. The threshold is the largest in the lumen
If a large percentage of the pixel brightness is selected, the rough segmentation
The action eliminates the area outside the lumen. On the other hand, the threshold is a rough
It is preferable to ensure that the application is continuous along the length of the imaged part.
New Threshold operation to achieve such a rough segmentation result
And connectivity criteria are well known to those skilled in the art. This seed pixel segment is used to define a finer segment for the first image data.
It is desirable to set the start point of segmentation. The rough segmentation of operation 522 is
Because it is preferable to bias towards under-inclusion,
Can be expected to form only a part of the region of interest.
it can. Therefore, additional images are added to the seed pixel segment by operation 532 of FIG.
Select criteria for connecting elements. These connectivity criteria in operation 532
May be performed prior to other operations of the present invention. Specific criteria
Can be selected taking into account the geometrical features of the region of interest.
You will understand. Roughly segmented vessels (ie, seed
The pixel segment) is dilated in operation 542. More specifically,
To ensure that the contrast-enhanced blood is contained within the segmented vessel segment.
Dilate a blood vessel using a dilation morphology operator. In addition, blood vessels
(The volume involved is larger than before),
Quantitative analysis of formatted image data (when trying to perform quantitative analysis)
Can be minimized by image artifacts. Threshold processing
The vascular surface determined by the technique is irregularly segmented in the presence of image artifacts.
It may have been split. The expansion conversion operation has a tendency to reduce these effects.
Direction, so that more accurate quantitative results can be obtained from cross-sectional linear data.
It becomes. To add to the area of the seed pixel segment during the dilation transformation process, 3
The pixel of interest observed as a rectangular box in D needs to touch an existing area
Where the contact is along the sides of the box, along the sides of the box
Direction or along the box vertex. Connectivity criteria is above
Any combination of the above three options may be used. Therefore, the figure
5 as part of operation 542, using the connectivity criteria selected in operation 532.
Apply the dilation transformation morphology operator to the result of the segmentation operation 522.
Use. The expansion transformation morphology operator allows for
More pixels are added to this area. For example, this area is dilated
The condition is that a pixel is at least one pixel already in the area and a side, surface or
Indicates that the pixel can be added only when connected at the vertex.
Think about it. By applying the morphological operator, the entire volume
And determine whether the pixel satisfies this criterion. If the conditions are met
Add that pixel to this area. These morphological operators and other
For more information on morphological operators, see Anil K. et al. Jain (Prent
Ice Hall, Englewood Cliffs, NJ)
damentals of Digital Image Processin
g "(1989, pp. 384-390). In operation 542, the dilation transformation operator is applied a selected number of times to mask the luminal region.
Can be created. Typically, the mask is a single pixel pattern,
This can be used to choose to leave or remove portions of another pattern of pixels.
Selective control can be performed. Here, the “mask (ma
sk) ”is first used as a standardized representation of the lumen section,
The center of the cross-sectional mass can be derived by calculating the moment (center of gravity). further
This mask is later used in a more typical manner, i.e., volumetric
It is also used to select the pixels of the lumen region in the scan data. In one specific implementation of the present invention, the mask created in operation 542 is
Corresponding to the pixel array of image data, but with binary values (ie,
, 0 or 1). This mask pixel
Has a value of 1 for pixels considered to be in the lumen region,
On the other hand, it has the value 0. Therefore, this mask can be segmented and expanded.
The luminal area specified by the dilation conversion is represented in binary notation. FIG. 6 is a flowchart detailing one embodiment of operation 404 of FIG. 4 in detail.
. The procedure in Fig. 6 is the total length of the blood vessel portion represented by the first volumetric data
Direction to determine the long axis of the blood vessel. The determination of the blood vessel axis is performed according to the procedure shown in FIG.
Successive slice images from the resulting mask (i.e.,
Using a masked slice image "). The first volume
Metric data has a corresponding grid defined by grid points of a linear grid
In a continuous image plane, a series of slice images (scan
That is, "axial slice" is shown. The following of FIG.
In the discussion, the two dimensional directions in each successive image plane were defined by the x and y axes.
Meanwhile, a series of axial slices extends in a third dimension defined by the z-axis.
It is assumed that In operation 614, first, a mask processing slice image is selected. Operation 62
At 4, for each of the x and y axes, the moment of the vessel lumen area is calculated.
, The center of the blood vessel. More specifically, the center of this blood vessel is
Momentometer in slice units for binary notation of vessel cross section provided by disk
It is determined by performing the calculation. Operation 634 allows this mask to calculate additional slices for which moments are to be calculated.
Is determined. If it contains the slice to be calculated,
The next mask processing slice image is selected by operation 644, and the processing procedure is changed to operation 62.
Return to the moment calculation of 4. Since this mask is in binary notation, each axial
For each slice, the moment (center of gravity) of the entire mask-processed slice image
The moment of the lumen area represented by the slice is equivalently represented. Its axial
If there are multiple vessels in the slice, the calculated moment
Note that the masked area is made smaller to correspond to the cross section of the vessel
I want to be. After calculating the moments, the processing procedure uses the calculated moments to calculate
This operation specifies the x and y coordinates of the center point of the lumen in the slice image.
, Proceed to operation 654. Therefore, x _c Is set equal to the x moment and y _c To y
By setting equal to the moment, the middle for each axial slice
Center point (x _c , Y _c ) Is specified. Operations 624 and 654 are, for example, as follows:
Then, it can be realized by a set of calculations. [Outside 1] These are optimized for the specific processing architecture of the computer 22
It is preferably implemented with appropriate object code. These practices are often
Any number of functionally equivalent programming languages (C, C ++, Fortran, etc.)
Routine programming tasks that can be performed in any language
Even if a test is necessary to perform
It will be understood that we are doing it. As a result, the weighted “1” pixel is
Summation in the y-direction gives a weighted "mass" in the y-direction. Similarly,
The weighted "1" pixels are summed in the x direction to obtain a weighted group in the x direction.
The remaining operations are standard calculations for calculating the center of gravity, and many programs
It is well known in the art of technology. Next, in operation 664, the two-dimensional center point (x _c , Y _c ) To a three-dimensional point
Connect. That is, the z-axis position of each axial slice is determined by operation 664.
The lumen center point of the slice (x _c , Y _c ). As a result of that
Each fits on the longitudinal axis of the vessel lumen at the corresponding z-value of the linear grid.
As a result, a set of discrete points in three dimensions is obtained. Next, in operation 674, a three-dimensional curve is applied to the three-dimensional center point.
I will. Using the moment calculated from each slice, determine the center point of each slice.
A smooth curve using the small mean squared error measurement method (ie, the curve
(A smooth curve that minimizes the square error between the center points). Then
This smooth curve represents the axis of a blood vessel in a three-dimensional space. Specific implementation
In one form, this smooth curve is a quadratic of z with x and y variables as parameters.
It is defined by a polynomial. “Curve fitting” is used herein to refer to a given type of curve
(I.e., the "family" curve), the closest to a particular set of points
It is defined as a calculation procedure for determining a specific curve. "Clan" curve
Is defined by a set of common defining equations whose coefficients are represented by parameters.
Refers to a set of curves such as One specific curve in that family is the coefficient
Specific values for the parameters can be specified and selected. Therefore,
One or more expressions with unspecified parameters can be pre-selected
Serves as a template that can be stored. With proper fitting operation
A particular value for the parameter is determined, which allows one
Is selected (fit to the data). Further, a segment definition function (pie) using a spline interpolation method, a collocation method, or the like.
It is also possible to define curves as a combination of cewise-defined functions)
. Another alternative is to use an independent basis, such as the finite element method or the wavelet method.
The curve is defined as a combination of functions (i.e., superposition). Song
Other alternative methods for defining lines are numerical analysis and computer-calculated models.
Will be apparent to those skilled in the art. The curve fitting operation may be, for example, to match a particular point (by interpolation) or to
Or an "optimal fit" curve that satisfies some specific optimization criteria
Thus, matching with the data can be performed. Typically,
A suitable fit curve is a graph in which the corresponding objective function has an optimal value (eg,
This is a curve to be estimated based on the maximum value or the minimum value. did
Thus, for that set of pixel data and the curve under consideration, the separation measure is
Corresponds to the value of the objective function. The optimization criterion is preferably minimization of the least squares value, but alternatively, in the art
One can also choose from a variety of well-known curve fitting optimization criteria. A curve
Models, or “families,” are low-order polynomials such as second-order and third-order polynomials.
It may be a family parametric curve defined by a term. Otherwise
And the fit operation 674 performs various spline interpolations (cubic splines, etc.)
Any of a variety of interpolation procedures, such as and others well known in the art.
The curve can also be determined by the interpolation procedure. FIG. 7 is a flowchart showing a procedure for realizing the oblique cutting generation operation 406 in FIG.
FIG. Cut at various locations along the longitudinal axis of the vessel using a smooth curve
Determine the plane. In one specific embodiment of the invention, the orientation of this plane is perpendicular to the vessel axis.
Choose to be a corner. Next, the volumetric data corresponding to this blood vessel
Data along the cut surface and cut off the reformatted CT data.
It is generated as surface linear data. More specifically, in this specific embodiment,
, A cut perpendicular to the vessel corresponds to an oblique cut in the volumetric data.
Data for every point on each cut plane perpendicular to the vessel axis.
Is interpolated from the data. In operation 716, z values at various locations are selected. For example, the z position to select is
The z position of a continuous linear slice image determined by the linear grid
Can be. Although not required, the origin of the axis (z = 0) is usually
It is desirable to be located at the center of the minute. Alternatively, this z-value may be
The distance may be selected to be equally spaced. In operation 726, the first oblique cutting is selected. For each oblique cut
In operation 736, the three-dimensional point where the oblique cutting intersects the longitudinal axis curve
Identify the point. The z position of the oblique cutting is the z of one of the linear grid surfaces.
If so, this intersection is simply entered in operations 624-654 of FIG.
At or near the 3D center point of the calculated lumen
Will come. In operation 746, a diagonal cut is selected for the three axes of the linear grid.
Direction is determined. In one specific embodiment, depending on the selected orientation,
An oblique cut perpendicular to the hand axis curve is placed at this intersection. In this selected orientation
Based on the operation 756, the oblique cutting grid point at the coordinates of the linear grid
(That is, the pixel position of the oblique cutting) is calculated. The direction to be selected is, for example, that the tangent is formed at the intersection point with respect to the straight grid axis.
Can be determined by calculating the angle. As an alternative, oblique cutting
The orientation can also be determined by calculating the oblique section equation. Long
The hand axis curve describes the x and y variables as functions of z (ie, where z is a parameter
Can be represented by a parametric equation defined as
The luminal center point along this curve is determined as outlined above. in this case
, At each lumen center point, the unit vector n = (
a, b, c) can be easily determined from the derivatives of x and y with respect to z. Of course, the vector n is the unit normal vector for the plane perpendicular to this curve at that point.
It is also a coutre. As is well known, the equation for this perpendicular surface is: Ax + by + cz + d = 0 In the above equation, d is determined by substituting the values of x, y and z at the center point.
Can be specified. This perpendicular surface is discretized, and
Of linear image data for oblique cutting from volumetric image data
The position of the oblique cutting can be determined at the position to be cut. The two alternatives described above, as well as others, provide for a given curve
This is a well-known method for determining a surface having a specified orientation. these
All of the alternatives are described in detail in the
Within the range that can be realized by the selected programming language without testing
It belongs well. Therefore, for the sake of brevity, these details are
The details are omitted. In operation 766, interpolation from pixel values of the first volumetric image data
As a result, the pixel value at the oblique cutting pixel position is generated. In one exemplary implementation,
The pixel value of each diagonal cut pixel is, for example, the nearest eight pixels on a linear grid.
Can be calculated by trilinear interpolation from neighboring pixel values of
Can be. The nearest eight neighbors are denoted by N _l (L = 1, ..., 8)
And the pixel value xsection (j, k, i) of the oblique cutting can be determined as follows.
You. [Mathematical formula-see original document] Here, the weight α _l Is the oblique cutting point xsection (j, k, i) and the pixel value N _l Based on the spacing between the linear grid points with It is determined so that For example, the eight nearest neighbors on a linear grid
Let us define the neighbor value as the vertex of a parallelepiped with a nominal volume of one. Then, the weight α _l Is
Of the parallelepipeds, the diagonal cutting position xsection (j, k, i) is used as a reference.
As N _l Of the part on the opposite side of the straight grid position diagonally (fra
ctional volume). About this exemplary embodiment for interpolation calculations
Will be described below with reference to FIG. In operation 776, it is determined whether or not the oblique cutting to be determined still remains.
You. When the oblique cutting to be determined remains, the processing procedure of FIG.
The operation proceeds to operation 786, which is the operation for selecting, and returns to operation 736. Oblique cutting is all
If it has already been created, the processing procedure of FIG. 7 ends. Here, the linear data indicated by the operation 410 in FIG. 4 can be displayed. blood
The reformulated volumetric data corresponding to the vessel
Assuming that it was straightened above, it was reconstructed from the projection data of that blood vessel.
It becomes CT image data that should be obtained. Depending on the selected orientation of the oblique slice,
Visualization of imaged structures from imagemetric image data and evaluation of structural features
Costs for imaged objects that would benefit evaluation tasks such as
Can provide alternative views. The above-described reformatting process 40 of the present invention comprises a data reformatting process according to the present invention.
The vessel 25 is separated from adjacent tissue using the
The blood vessels are stretched so that they are straight. One simple method for processing cross-sectional linear data is on a plane perpendicular to the vessel axis.
Is to obtain the sum of all the pixels of. This process essentially involves the blood vessels
The volume of the lumen of the blood vessel at each point along can be estimated. 50% stenosis in a certain blood vessel
Given that there are areas, this method uses radiographs created by angiography.
Is more powerful than when determining the diameter of a blood vessel. That is, the diameter measurement
It will have a contrast of 50% / SQRT (2), but in the area measurement,
Ideally, the area measurement will be better for lumen
Be more sensitive to changes in range. By visual assessment or other calculator
Other evaluation schemes are possible, such as by order. Here, the reformatting process 40 of the present invention will be described with reference to FIGS.
Let us consider in more detail. These figures illustrate the operations described above with reference to FIGS.
Is exemplarily shown. FIG. 8 shows a volumetric image data to be reformatted according to the present invention.
FIG. 9 is a schematic diagram of a 3D image of a blood vessel portion 800 expressed by a graph. Operation 512 in FIG.
Referring to FIG. 8, the seed pixel 810 of FIG.
It is desirable to select from linear slices at the edges. Segment division operation 52
As a result of Step 2, a seed pixel segment 820 roughly segmented is obtained.
. As described above, the desired result from segment segmentation operation 522 is
A region composed of pixels of high intensity (such as the region 830 in FIG. 8) is composed of pixels of low intensity.
It is identified from a region (such as region 840). In FIG. 8, the selection of the threshold value ensures that the seed pixel segment 820 is
To extend continuously through the blood vessel portion 800. Therefore, in FIG.
The seed pixel 810 can be used to grow a blood vessel region. This seed is
, Basically, a point in the central region of the oblique cross section of the vessel, optionally this region
Is a point selected by the user. The threshold chosen to grow this area
Pixel intensities that exceed the value and meet special conditions are combined while traversing the blood vessel.
When the selected pixel exceeds the threshold and is connected to a pixel in the area
, The pixel is added to the area. In the application of the present invention to medical imaging,
, Typically, a finite time window (tem) for collecting volumetric projection data
poral window). In these cases, the choice of threshold (and therefore,
In addition, the choice of which pixels to add to the blood vessel image is an important consideration.
One skilled in the art will recognize that the desired criteria for selecting a threshold is to collect projection data.
Will vary depending on the situation. More specifically, if a relatively lower value is selected for the threshold,
Pixels of lower intensity will be included in the image. Pixels of lower intensity are
If the structure is tubular, it is typically located at the periphery of the imaged structure.
Will be displayed as a roll-off of that structure. taller than
Intensity pixels typically shed blood, given that the structure is a blood vessel.
It is located near the center of the structure. Typically, vascular imaging situations
Now, to enhance the detection of blood vessels to be imaged,
I am receiving a contrast agent. FIG. 9 illustrates the mask creation operation 542 of FIG. As indicated by the arrow
, The seed pixel segment 820 is expanded and converted by the expansion conversion operation to obtain a lower strength.
Include the pixel of the degree. In this expansion conversion operation, the direction of the wall of the blood vessel 800 is directed outward.
, The expansion of the segment 820 is common. Expansion operation 1
After performing one or more times (for example, a selected number of times), the seg
The comment becomes the mask 920. Mask 920 is a binary notation of the lumen region
. FIG. 10 corresponds to the processing procedure of FIG. As shown in FIG.
Essentially the first volumetric image
X, y planes 1012 to 12x stacked on the z plane based on the linear grid of data
The base of the image of the blood vessel at 1018 is formed. Next, the blood vessel space of the scanner
The oblique sections 1022 to 1028 of the Charl scan are photographed, and the oblique sections are taken.
Use and estimate the center of the vessel using moment calculation. As shown in FIG. 10, the lumen mask 920 basically has a meandering cylindrical shape.
ing. A serpentine cylinder is one example of an insect-like structure defined above. Insect-like structure
Other examples of regular cylinders, tapered cylinders, spiral structures, irregular cylinders,
and so on. The invention further generally relates to volumetric structures as a whole.
Applicable. A straight grid cross-section of a blood vessel, such as cross-section 1022, has two dimensions (x, y).
And is oval or elliptical. By the values of the x and y coordinates of the moment calculation,
The luminal center of the blood vessel in the straight grid cross section is specified. Then the straight line of the blood vessel
A cross section perpendicular to the longitudinal axis of the vessel (ie, oblique
Section). FIG. 11 illustrates the processing procedure of FIG. 6 for determining the longitudinal axis curve of lumen 920.
Output is supported. The three-dimensional curve 1100 is moved to the center point 1122 by the operation 674.
~ 1128 undergoes a curve fit while this center point 1122-1128
Is the center of gravity calculated in operations 624 to 654 of FIG. As mentioned above, the method
Depending on the specific curve fitting procedure selected to perform
If you have not fitted the curve using an interpolation method (for example, spline interpolation),
May not actually cross the center of gravity. FIG. 12 corresponds to the result of the processing procedure shown in FIG. z axis 1200
Equivalent to the reformatted longitudinal axis curve obtained in step 674 of step 6.
. The planes 1212 to 1218 correspond to the respective planes of the oblique cutting, and these planes are reproduced.
The formatted data is parallel. Lumen sections 1222-1228 are
The respective slopes along the longitudinal axis 1200 of the vessel lumen 1220 after formatting
Fig. 4 shows each cross section of a lumen in a directional cut. FIG. 13 illustrates trilinear interpolation performed in operation 766 of FIG.
You. The spacing of the linear grid in each dimension direction (x, y and z) should be normalized to 1.
There is. The pixel 1300 has a pixel value of a linear grid pixel 1302 to 131
These pixels are cut in the oblique direction as determined by interpolation between the eight pixels. Pixel 1302 is
For example, the position (xint, yint, zi) with respect to the linear position of the pixel 1300
nt). The volume element 1320 has a position (xint + 1, yint,
zint) at a position opposite to the pixel 1306 corresponding to the diagonal direction. Volume dxm
* Dyp * dzp is the pixel 1306 as described above with respect to operation 766 of FIG.
Pixel value N _l Interpolation coefficient α _l It becomes. Typical case where the present invention is applied particularly to volumetric image data representing blood vessels
Then, the reformatted data has a length of 6 to 10 millimeters (mm).
May represent a vessel segment. Of course, in general, these vessels
Longer than the imaged part. According to the above aspect of the present invention, the data is re-fed according to the features to be visualized.
By formatting, volumetric images such as 3D tomographic image data
Data visualization is improved. In particular, in the case of tomographic image data, another
Furthermore, by defining the reconstruction plane before performing the image reconstruction,
Is improved. On the other hand, in this alternative aspect of the invention, a multi-plane reconstruction (MPR)
A series of oblique slices are reconstructed as a single unit using the implementation of backprojection. Sa
Furthermore, in this alternative embodiment, information from the initial (linear grid) image data is taken.
In addition, the determination regarding the continuation of the MPR reconstruction is performed. [0102] Briefly, an embodiment of this alternative embodiment is based on the processing procedure of FIG.
May obey. However, in this embodiment, operation 406 is a straight grid image.
Instead of using interpolation on image data, create image data for oblique cutting.
Includes multiple plane reconstructions to generate In this aspect of the invention, either of the two back projection forms can be used. each
The purpose of the oblique plane is to interpolate between the linear grid image data and
This is to create image data for a surface. In the detector-driven scheme, the plane
The volume element (voxel) of each image data of the image to the square element of the detector array
You. Each of the square elements is back-projected to a point at the location of the X-ray source, thereby
Get rid of body volume. If the tetrahedron of the detector element intersects the voxel,
The signal of the instrument element is included in the intensity contribution to the pixel of interest. Each of the detector signals
For each, the partial volume of the voxel that intersects the detector element tetrahedron
Weight. In the pixel-driven method, in the presented oblique-direction image plane, from the source point to the pixel position.
Draw a line with. For each pixel of interest, extend this line to the detector plane.
Determine the points on the detector plane. The four nearest detectors adjacent to this detector plane point
The element is specified. Then, the signals of the four detector elements specified are added to the detector surface.
Performs bilinear interpolation using the relative distance to the point of interest and determines the interpolation weight.
Set. This procedure is repeated for each pixel of the presented oblique image and a set of projections is performed.
Reconstruct the shadow data view (i.e., sinogram)
An intensity value at the pull position is created. In both detector-driven and pixel-driven cases, those skilled in the art will be aware of FIGS.
In reviewing the above description relating to aspects of the invention described above with reference to FIG.
You will understand more about its implementation. The first aspect of the present invention
An advantage over this alternative is that volumetric image reconstruction (linear grid
It only needs to be performed once (on the pad projection data). On the other hand,
An alternative method using shadows uses information extracted from linear grid image data to perform oblique
A determination is made as to the position and orientation of the surface in the direction (operations 716-756 in FIG. 7)
See discussions). The present invention is particularly useful when applied to the specific context of medical imaging.
One advantage is that the present invention can provide significant clinical benefits.
. According to the present invention, in order to properly evaluate stenosis of a blood vessel, a contrast agent is applied to a peripheral portion of the vein structure
Minimize the invasiveness of the scanning procedure, as only one injection is required
Can be. With conventional arterial angiography, on the other hand, there is no space around the affected artery.
It is necessary to administer a contrast agent. The present invention has been described using a CT imaging system and in terms of blood vessels.
However, the present invention is not limited to CT imaging systems and blood vessels.
No. On the contrary, the invention is applicable to all imaging modalities,
It can be applied to any meandering or "insect-like" structure that can be divided.
As a field of particular applicability, the present invention relates to the characteristics of blood vessels or segments containing arteries.
Provides real benefits to quantify the characteristics of any tubular structure that can be split
Wear. Many features and advantages of the invention will be apparent from the detailed specification, and thus,
It is intended that the appended claims be construed in accordance with the true spirit and scope of the invention.
It is intended to encompass all of the features and advantages of the invention. Furthermore, those skilled in the art
Since many modifications and changes can be readily made to the present invention, the present invention has been
It is not intended to be limited to dense structures and operations and, therefore, is not
All appropriate modifications and equivalents may be made so that they fall within the scope required.
. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a prior art X-ray imaging system or computed tomography image
FIG. 1 is a schematic diagram of a switching system. FIG. 2 is a schematic diagram of an X-ray system or a computed tomography system according to the present invention.
You. FIG. 3 is a schematic diagram of a data reformatting program according to the present invention. FIG. 4 is a flow chart of a method according to the present invention. FIG. 5 shows a detailed process of the method according to the present invention. FIG. 6 shows a detailed process of the method according to the present invention. FIG. 7 shows a detailed process of the method according to the present invention. FIG. 8 shows a detailed process of the method according to the present invention. FIG. 9 shows a detailed process of the method according to the present invention. FIG. 10 shows the detailed processing of the method according to the present invention. FIG. 11 shows the detailed processing of the method according to the present invention. FIG. 12 is a diagram showing the detailed processing of the method according to the present invention. FIG. 13 shows a detailed process of the method according to the present invention. DESCRIPTION OF THE SYMBOLS 10 Prior Art Imaging System 16 Support 14 Object 12 Source 18 Detector Array 19 Data Acquisition Component 20 Controller 22 Computer for CT System 24 Computer Memory for System 26 Display 34 Measuring System 25 Data Reformatting program 800 Blood vessel portion represented by volumetric image data 810 Seed pixel 820 Seed pixel segment 830 Region composed of high intensity pixels 840 Region composed of small intensity pixels 920 Mask 1012 to 1018 x, y plane 1022 to 1028 oblique Directional cross section 1100 Three-dimensional curve 1122 to 1128 Center point 1200 Z axis 1212 to 1218 Surface 1222 to 1228 Luminal cross section 1300 Pixels 1302 to 1318 Linear grid Pixel 1320 volume element N _l Pixel value α _l Interpolation coefficient

──────────────────────────────────────────────────続 き Continuation of the front page (51) Int.Cl. ⁷ Identification symbol FI Theme court 参考 (Reference) G06T 1/00 G01N 24/02 530Y (31) Priority claim number 09 / 648,956 (32) Priority date Heisei August 25, 2000 (August 25, 2000) (33) Priority Country United States (US) (81) Designated State DE, IL, JP, SG (72) Inventor Yabs, Mehmet United States of America, 75038, Texas , Irving, Apartment 3056, North MacArthur Boulevard, No. 5319 (72) Inventor Ishark, Ahmad Nadiem United States, 12065, New York, Clifton Park, Kins Road, No. 629 F-term (reference) 4C093 AA22 CA21 DA02 FD09 FF15 FF28 FF42 FG05 4C096 AA20 AB50 AC10 AD14 BA18 BA36 DB08 DB19 DC11 DC14 DC18 DC21 DC28 DC31 DC36 4C301 BB13 EE11 FF09 JC06 JC14 KK12 KK18 4C601 BB03 EE09 FE03 JC15 JC20 JC21 JC25 KK21 KK23 KK24 5B057 AA09 BA03 BA05 BA07 CA02 CA08 DC14 DB16 CB02 The continuation of the summary corresponds to data that causes the CT imaging system to reconstruct if the structure is straightened and the CT imaging system directly obtains an image of the straightened structure.

Claims (1)

Claims: 1. An object corresponding to an object and collected by an imaging system.
A method of displaying an object based on volumetric image data, wherein the volumetric image data is converted to linear image data corresponding to the object.
Reformatting; and displaying the linear image data. 2. The step of reformatting comprises the step of separating volumetric image data corresponding to an object from remaining data.
Dip transforming the segmented volumetric image data; and a longitudinal axis of the object of the segmented volumetric image data.
Determining a surface such that it is an orthogonal cross section of the object at a selected position along the axis
Re-forming the volumetric image data obtained by the dilation conversion and division along the plane.
The method of claim 1, comprising: matting. 3. The method of claim 1, wherein the object comprises a blood vessel. 4. The method of claim 1, wherein said object comprises an artery. 5. The method according to claim 1, wherein said volumetric image data is
The method of claim 1, wherein the method has been collected by a maging system. 6. The method according to claim 1, wherein said volumetric image data is magnetic resonance image data.
The method of claim 1, wherein the method has been collected by a system. 7. An ultrasonic imaging system wherein the volumetric image data is
The method of claim 1, wherein the method has been collected by a stem. 8. An X-ray imaging system comprising:
The method of claim 1, wherein the method has been collected by a system. 9. A volumetric image data acquired by an imaging system.
A method of reformatting data by segmenting blood vessels from remaining components using intensity thresholds and connectivity criteria.
Dividing the blood vessel roughly from the remaining components by splitting; expanding and transforming the segmented blood vessel; determining the longitudinal blood vessel axis of the blood vessel; and At the selected location along
Reformatting the volumetric image data along the surface,
Creating linear image data corresponding to the tube. 10. The method further includes the step of displaying the linear image data.
The method according to claim 9. 11. The blood vessel comprises an artery and the remaining component is an organ.
10. The method of claim 9, comprising a tissue. 12. The volumetric image data is computed tomography.
The method of claim 9, wherein the method is image data collected by a system. 13. The method according to claim 1, wherein the volumetric image data is magnetic resonance imaging.
The method of claim 9, wherein the method is image data collected by a system. 14. A volumetric image data acquired in an imaging mode.
Computer readable containing a program to reformat the data
Medium, the program, when executed by a computer, cuts a blood vessel from an adjacent component and cuts the cut blood vessel into the cut blood.
The computer performs the process of elongating to a linear shape corresponding to the tube
A computer-readable medium. 15. A three-dimensional volumetric image data of a blood vessel,
Compiler for reformatting to correspond to two-dimensional linear image data of a tube
A computer that executes a computer program, and a display device that displays the two-dimensional linear image data. 16. An imaging system according to a particular imaging modality.
The volumetric image data acquired and reconstructed by
A storage device for storing volumetric image data of a tubular structure, and an image of the straightened blood vessel which is straightened by stretching the blood vessel.
Assuming that the system has acquired it directly,
Therefore, the volumetric image data is set to correspond to the reconstructed data.
Executing a computer program for reformatting
An apparatus comprising: 17. The method according to claim 17, wherein the volumetric image data is computed tomography.
17. The image data of claim 16, which is image data collected by an imaging system.
Equipment. 18. The method according to claim 18, wherein the volumetric image data is an X-ray imaging system.
17. The apparatus of claim 16, wherein the apparatus is image data collected by a stem. 19. The method according to claim 19, wherein the volumetric image data is magnetic resonance imaging.
The apparatus of claim 16, wherein the apparatus has been collected by a system. 20. A volumetric image acquired by an imaging system
Read a computer containing a program to reformat the data.
A removable medium, wherein the program is executed by a computer
Reformat the volumetric image data into linear image data corresponding to the object.
Computer-readable instructions that instruct the computer to perform a process comprising: matting; displaying the linear image data.
Media. 21. The reformatting process, wherein volumetric image data corresponding to an object is classified from remaining data.
A process of dilating and transforming the divided volumetric image data; and a longitudinal axis of the object of the dilated and divided volumetric image data.
And determining a surface that is an orthogonal cross-section of the object at the selected position along the axis
And re-transforming the volumetric image data obtained by the dilation conversion and division along the plane.
21. The computer-readable medium of claim 20, comprising: formatting. 22. A volumetric image acquired by the imaging system
Read a computer containing a program to reformat the data.
A removable medium, wherein the program is executed by a computer
Using the intensity threshold and connectivity criteria to select the vessel
The blood vessel portion is roughly divided from the remaining components by segmentation.
Dilating and transforming the segmented blood vessel portion; determining the lengthwise blood vessel axis of the blood vessel portion; Decision
Reformatting the volumetric image data along the surface,
Creating linear image data corresponding to the tube portion; a computer-readable instruction directing the computer to perform a process comprising:
Media. 23. A source transmitting X-rays in the direction of the object, and a source in the object.
And detection of detecting X-rays passing therethrough and converting the detected X-rays into a signal.
A source array and a source and detector array coupled to the source and detector array.
A controller controlling the rays and a data acquisition component receiving the signal.
Or an imaging system that includes
An apparatus included in an imaging system, the apparatus coupled to the source, the controller, and the data acquisition component.
Receiving a signal from the data collection component and transforming the signal into an object volume.
Is converted into volumetric image data, and the volumetric image data
A system computer that converts the data into linear image data of the object; and a display device that displays the linear image data of the object. 24. The signal obtained by an X-ray imaging system
24. The apparatus of claim 23, wherein 25. The system as claimed in claim 25, wherein the signal is a computed tomography imaging system.
24. The apparatus of claim 23, obtained by: 26. A method for imaging a volumetric structure, comprising: quantitative information representing structural features of the volumetric structure in the imaged object.
Information based on volumetric image data representing the imaged object
Step; fingering a cross-sectional image of the volumetric structure in a plane selected based on the quantitative information.
Creating defined cross-sectional image data. 27. The step of creating cross-sectional image data further includes the step of
27. The method of claim 26, which is based on rhometric image data. 28. The volumetric image data created based on volumetric projection data.
And said step of creating cross-sectional image data further comprises volumetric projection data
27. The method of claim 26, wherein the method is based on data. 29. The object of claim 26, wherein the object is part of a human patient.
the method of. 30. The method of claim 29, wherein said portion is an organ. 31. The blood vessel wherein the volumetric structure is contained within the portion
30. The method of claim 29, wherein there is. 32. The method of claim 31, wherein said portion is an organ. 33. The volumetric structure of claim 26, wherein the volumetric structure is an insect-like structure.
The method described in. 34. The method of claim 33, wherein said insect-like structure is a tubular structure.
. 35. A tubular structure wherein said volumetric structure has a longitudinal axis.
And the quantitative information is representative of the longitudinal axis.
27. The method of claim 26, comprising a parametric value defining a curve. 36. The method of claim 35, wherein the longitudinal axis is a curved axis.
the method of. 37. The method according to claim 37, wherein the structural feature is that the volumetric structure is longitudinal.
27. The method of claim 26, wherein the axis is a curvilinear axis that extends in the direction of the arrow. 38. A parametric curve wherein said quantitative information represents said curved axis
38. The method of claim 37, wherein 39. The surface contacts the curved axis at a corresponding point on the axis.
38. The method of claim 37, wherein the method is selected to be at right angles to the line. 40. A step for displaying a sectional image from the sectional image data.
27. The method of claim 26, comprising a step. 41. In an additional plane selected based on said quantitative information,
Additional cross-sectional image data specifying additional cross-sectional images for volumetric structures
27. The method of claim 26, comprising creating data. 42. Volumetric image data representing an imaged object
At least one storage device for storing data, and quantitative information representing structural features of the volumetric structure in the imaged object.
Information and calculate the volumetric structure in the selected plane based on the quantitative information
A computer for creating cross-sectional image data specifying a cross-sectional image relating to
An imaging device comprising: 43. The computer according to claim 43, further comprising:
43. The apparatus of claim 42, wherein the apparatus is created based on metric image data. 44. The volumetric image data comprising:
And the computer generates the cross-sectional image data.
43. The data is further created based on volumetric projection data.
An apparatus according to claim 1. 45. The object of claim 42, wherein the object is part of a human patient.
Equipment. 46. The device according to claim 45, wherein said part is an organ. 47. A method according to claim 47, wherein the volumetric structure is a blood vessel contained in the organ.
47. The device of claim 46, wherein there is. 48. The volumetric structure of claim 42, wherein the volumetric structure is an insect-like structure.
An apparatus according to claim 1. 49. The device of claim 48, wherein said insect-like structure is a tubular structure.
. 50. A tubular structure in which the volumetric structure has a longitudinal axis.
Parametric curve, wherein the quantitative information represents a longitudinal axis
43. The apparatus of claim 42, comprising a parametric value defining 51. The method according to claim 51, wherein the structural feature is that the volumetric structure is longitudinal.
43. The device of claim 42, wherein the device is a curvilinear axis that extends in the direction of the arrow. 52. The computer according to claim 45, wherein the computer corresponds to the curved axis and a corresponding axis on the axis.
51. The surface is selected to be at right angles to a line tangent at points.
An apparatus according to claim 1. 53. A cross-sectional image is displayed according to the cross-sectional image data.
43. The device of claim 42, further comprising a display device for: 54. Create volumetric image data representing an object to be imaged.
Coupled to, or associated with, the imaging system
An image analysis system included in the stem, wherein the volumetric measurement in the object imaged based on the volumetric image data is performed.
Calculating quantitative information representing the structural characteristics of the network structure, and based on the quantitative information
Cross-sectional image specifying a cross-sectional image of the volumetric structure in the selected plane
Image analysis system having a computer system for creating data
. 55. The computer system stores the cross-sectional image data.
55. The method according to claim 54, wherein the data is created based on volumetric image data.
Image analysis system. 56. The imaging system according to claim 56, wherein the volumetric image data is
Is created by the stem based on the volumetric projection data, and
The computer system further stores the cross-sectional image data and the volume
55. The image analysis system according to claim 54, wherein the image analysis system is created based on lick projection data.
M 57. The object of claim 54, wherein the object is part of a human patient.
Image analysis system. 58. The image analysis system according to claim 57, wherein said part is an organ.
Tem. 59. The volumetric structure is an insect-like structure.
The image analysis system according to 1. 60. The image of claim 59, wherein said insect-like structure is a tubular structure.
Analysis system. 61. The volumetric structure is a tubular structure having a longitudinal axis, and
A parameter defining the parametric curve whose quantitative information represents the longitudinal axis.
55. The image analysis system according to claim 54, comprising a metric value. 62. The structural feature may be such that the volumetric structure is longitudinal.
55. The image analysis system according to claim 54, wherein the image analysis system is a curved axis that extends in a direction extending in the direction of the arrow. 63. The surface touches the curved axis at a corresponding point on the axis.
63. The image analysis of claim 62, wherein the image analysis is selected to be perpendicular to the line.
system. A cross-sectional image is displayed according to the cross-sectional image data.
55. The image analysis system according to claim 54, further comprising a display system. 65. A computer program for imaging a volumetric structure.
A computer-readable medium that has been converted into an image based on volumetric image data representing an imaged object.
Compute quantitative information describing structural features of volumetric structures in objects
A cross section of the volumetric structure in a plane selected based on the quantitative information
Creating a cross-sectional image data specifying an image, a computer-readable medium containing instructions for performing the following. 66. The program instruction for creating cross-sectional image data, wherein:
Instruction to create cross-sectional image data further based on volumetric image data
66. The computer-readable medium of claim 65, comprising: 67. A method according to claim 67, wherein the volumetric image data is a volumetric projection.
The above-mentioned program for creating cross-sectional image data, which is created based on the data.
A gram command further converts the cross-sectional image data based on volumetric projection data.
66. The computer-readable medium of claim 65, comprising instructions for creating
body. 68. The object of claim 65, wherein the object is part of a human patient.
Computer readable medium. 69. The computer according to claim 68, wherein said part is an organ.
A readable medium. 70. The volumetric structure of claim 65, wherein the volumetric structure is an insect-like structure.
A computer-readable medium according to claim 1. 71. The condenser of claim 70, wherein said insect-like structure is a tubular structure.
A computer readable medium. 72. A tubular structure wherein said volumetric structure has a longitudinal axis.
And the quantitative information is representative of the longitudinal axis.
66. The computer of claim 65, comprising a parametric value defining a curve.
Data readable media. 73. The structural feature may be such that the volumetric structure is longitudinal.
66. The computer-readable medium of claim 65, wherein the axis is a curved axis that extends in
Media. 74. The program further comprises a correspondence between the curved axis and the axis.
To select the surface so that it is perpendicular to the line tangent at the point
74. The computer readable medium of claim 73, comprising computer gram instructions.
. 75. The program according to claim 75, wherein the program further comprises:
66. The computer of claim 65, comprising program instructions for displaying
Data readable medium. 80. A volumetric object contained within the imaged object
A method of imaging a volumetric structure represented by data, wherein the volumetric image data is analyzed to obtain a volumetric structure;
Determining quantitative information representing structural features; and, in a corresponding plane selected based on the quantitative information, each of the volumes
Creating a plurality of image data sets specifying a cross-sectional image of the trick structure
And a method comprising: 81. The method according to claim 81, wherein the volumetric data is volumetric image data.
Data and the creation of a plurality of image data sets is further volumetric
81. The method of claim 80, wherein the method is based on image data. 82. The method according to claim 82, wherein the volumetric data is volumetric projection data.
Data, and the generation of the plurality of image data sets further comprises a volumetric
81. The method of claim 80, wherein the method is based on projection data. 83. The imaged object is part of a human patient.
80. The method according to 80. 84. The method of claim 83, wherein said portion is an organ. 85. The blood vessel wherein the volumetric structure is contained within the portion
84. The method of claim 83, wherein: 86. The method of claim 85, wherein said portion is an organ. 87. The volumetric structure according to claim 80, wherein said volumetric structure is an insect-like structure.
The method described in. 88. The method of claim 87, wherein said insect-like structure is a tubular structure.
. 89. A tubular structure wherein said volumetric structure has a longitudinal axis.
And the quantitative information is representative of the longitudinal axis.
81. The method of claim 80, comprising a parametric value defining a curve. 90. The method of claim 89, wherein the longitudinal axis is a curved axis.
the method of. 91. The method according to claim 91, wherein the structural feature is that the volumetric structure is longitudinal.
81. The method of claim 80, wherein the axis is a curvilinear axis that extends in a direction that extends. 92. A parametric curve wherein said quantitative information represents said curved axis
92. The method of claim 91, wherein 93. The method according to claim 93, wherein the plane is at each point along the curved axis.
Claims: selected to be perpendicular to respective lines tangent to said curved axis
Item 90. The method according to Item 91. 94. The method according to claim 94, further comprising: generating the cross-sectional image according to the plurality of image data sets.
81. The method of claim 80, comprising displaying. 95. Represents a volumetric structure in an object to be imaged
An image analysis device for analyzing volumetric data, comprising: a volumetric data and a volumetric image determined based on the volumetric image data.
And quantitative information representing structural features of the volumetric structure
And at least one storage device for each of the volumes within a corresponding plane selected based on said quantitative information.
A command to create multiple image data sets specifying the cross-sectional images of the trick structure
An image analysis device comprising: a computer; The volumetric data may be a volumetric image data.
And the computer further stores the plurality of image data sets in a volume.
The image analysis according to claim 95, wherein the image analysis is performed based on the image data
apparatus. 97. A method for generating volumetric projection data, comprising:
And the computer further stores the plurality of image data sets in a volume.
95. The image analysis according to claim 95, wherein the image analysis is made based on the projection data.
apparatus. 98. The imaged object is part of a human patient.
95. The image analysis apparatus according to 95. 99. The image analysis device according to claim 98, wherein said part is an organ.
. 100. The volumetric structure of claim 9, wherein the volumetric structure is an insect-like structure.
6. The image analysis device according to 5. 101. The method of claim 100, wherein said insect-like structure is a tubular structure.
Image analysis device. 102. The tubular wherein the volumetric structure has a longitudinal axis
Parameter, wherein the quantitative information is representative of the longitudinal axis.
97. The image analysis of claim 95, comprising a parametric value defining a curve.
apparatus. 103. The structural feature may be such that the volumetric structure is longitudinal.
97. The image analysis device according to claim 95, wherein the image analysis device is a curved axis that extends in the direction. 104. The computer, wherein each of said computers along said curved axis
The surface so as to be perpendicular to the respective line tangent to the curved axis at the point
104. The image analysis device according to claim 103, wherein the device is selected. 105. The cross-sectional image according to the plurality of image data sets.
The image analysis device according to claim 95, further comprising a display device for displaying the image. 106. A volume representing a volumetric structure in an object to be imaged.
Coupled to an imaging system that collects
Or an image analysis system included in the imaging system, the storage system for storing the volumetric data, and a selection based on quantitative information representing structural features of the volumetric structure. did
In the corresponding plane, each of them is based on volumetric data and
To create multiple image datasets that specify volumetric cross-sectional images
And a computer system for analyzing the image. 107. The volumetric data is a volumetric image
107. The image analysis system according to claim 106, which is data. 108. The volumetric data is volumetrically projected.
And the computer system stores the plurality of image data sets.
, And further based on volumetric projection data.
An imaging device according to any of the preceding claims. 109. The imaged object is part of a human patient.
107. The image analysis system according to item 106. 110. The image analysis of claim 109, wherein said part is an organ.
system. 111. The volumetric structure is an insect-like structure.
06. The image analysis system according to 06. 112. The method of claim 111, wherein said insect-like structure is a tubular structure.
Image analysis system. 113. The tubular structure wherein said volumetric structure has a longitudinal axis
Parameter, wherein the quantitative information is representative of the longitudinal axis.
107. The image solution of claim 106, comprising a parametric value defining a curve.
Analysis system. 114. The structural feature may be such that the volumetric structure is longitudinal.
107. The image analysis system of claim 106, wherein the image analysis system is a curved axis that extends in the
M 115. The method according to claim 115, wherein the plane is at each point along the curved axis.
Selected to be at right angles to each line tangent to said curved axis.
The image analysis system according to claim 114. 116. The cross-sectional image according to the plurality of image data sets.
107. The image analysis system according to claim 106, further comprising a display system for displaying an image.
Tem. 117. An object collected and imaged by an imaging system.
To analyze volumetric data representing the volumetric structure in
A computer-readable medium encoded with a program for
The program analyzes the volumetric image data to create a volumetric structure.
Determining quantitative information representative of the structural features; and in a corresponding plane selected based on said quantitative information, each of which is a volume.
Creating a plurality of image data sets specifying a cross-sectional image of the trick structure, a computer-readable medium containing instructions for performing the following. 118. The volumetric data is a volumetric image
Said program instructions including data and for creating a plurality of image data sets
Creates the plurality of image data sets further based on volumetric image data.
118. The computer-readable medium of claim 117, comprising instructions for performing. 119. The volumetric data is volumetrically projected.
Said program instructions comprising data and creating a plurality of image data sets,
The plurality of image data sets are further created based on volumetric projection data.
118. The computer-readable medium of claim 117, comprising instructions for causing the computer-readable medium. 120. The imaged object is part of a human patient.
128. A computer-readable medium according to item 117. 121. The computer according to claim 120, wherein said part is an organ.
Data readable medium. 122. The volumetric structure is an insect-like structure.
18. The computer-readable medium according to 17, 123. The method of claim 122, wherein said insect-like structure is a tubular structure.
Computer readable medium. 124. The tubular structure wherein the volumetric structure has a longitudinal axis
Parameter, wherein the quantitative information is representative of the longitudinal axis.
118. The computer of claim 117, comprising a parametric value that defines a loop curve.
A computer readable medium. 125. The structural feature may be such that the volumetric structure is longitudinal.
118. The computer-readable medium of claim 117, wherein the computer-readable axis is a curved axis.
A removable medium. 126. The program according to claim 126, further comprising:
At each point perpendicular to the respective line tangent to the curved axis
126. The control of claim 125, comprising program instructions for causing a face to be selected.
A computer readable medium. 127. The program further according to the plurality of image data sets.
And a program instruction for displaying the cross-sectional image.
A computer-readable medium according to claim 7.