CN101283910B - Method for obtaining the coronary artery vasomotion information - Google Patents
Method for obtaining the coronary artery vasomotion information Download PDFInfo
- Publication number
- CN101283910B CN101283910B CN2008100550364A CN200810055036A CN101283910B CN 101283910 B CN101283910 B CN 101283910B CN 2008100550364 A CN2008100550364 A CN 2008100550364A CN 200810055036 A CN200810055036 A CN 200810055036A CN 101283910 B CN101283910 B CN 101283910B
- Authority
- CN
- China
- Prior art keywords
- point
- skeleton
- motion
- constantly
- vascular
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
Abstract
A method for acquiring the motion information of coronary artery, which belongs to the field of medical detection technology, is used for solving the problem in acquiring the motion information of heart. The technical scheme is that a 3D motion estimation method based on elastic registration is adopted after achieving 3D reconstruction of branch skeleton of main blood vessels at each moment in X-ray coronary artery angiogram sequences; the motion of blood skeleton is transformed to be matched with skeleton lines at continuous time moments; the motion vector and the motion trajectory of each blood skeleton point in cardiac cycle are calculated; and the motion information including global and the local motion parameters of the coronary artery in the cardiac cycle is extracted according to the estimated motion vector of each blood skeleton in the image sequence. The method has the advantages of high accuracy in extracting the motion information of blood vessels, convenient operation and high work efficiency.
Description
Technical field
The present invention relates to a kind of from the digital X-ray coronarogram that covers one or more cardiac cycles as the method for extracting the relevant movable information of blood vessel the sequence, belong to technical field of medical detection.
Background technology
Coronary artery is the important blood vessels of supply heart blood, and it is positioned at epicardial surface, with cardiac muscle the motion of rhythm and pace of moving things ground is arranged in cardiac cycle, so the motion of arteria coronaria can reflect the motion of heart itself.The X ray coronarography is diagnosis of generally acknowledging in the world at present and the main foundation for the treatment of coronary heart disease.Adopt angiographic image not only can detect the position and the degree of ischemia injury (angiostenosis), compare with one or two width of cloth still images of Same Scene, image sequence can provide the more information of horn of plenty, is promptly implying the multidate informations such as blood vessel distortion in the cardiac cycle.
Research great majority to coronary motion situation in the contrastographic picture sequence all only limit to follow the tracks of limited characteristic point in early days, for example implant the metal marker of heart or the branch point of vascular tree etc.The former is owing to operational difficulty, and Practical significance is little; The latter can only estimate the partial motion of coronary artery, and can't obtain its whole movable information because the number of discernible arteria coronaria tree bifurcation is limited.Need the operator to select the coordinate of bifurcation or input bifurcation simultaneously when process begins, the computation time that is spent is very long.Afterwards, researcheres began to utilize the method for Digital Image Processing, and for example (displacementvector field DVF) extracts two dimensional motion information coronarius with optical flow method to motion vector field from the contrastographic picture sequence.But the greatest drawback of two dimensional image is by the perspective projection imaging space structure to be overlapped onto on the two-dimensional imaging plane, thereby has lost most of spatial information.And actual vasomotion is three-dimensional, and inhomogeneities with room and time, therefore only can not obtain accurate estimation, need to adopt the image sequence of a plurality of angles, estimate the three-dimensional motion of blood vessel the arteria coronaria motion according to the two-dimensional image sequence of an angle.Puentes (" Dynamic feature extraction of coronary artery motionusing DSA image sequences; " IEEE Transactions on Medical Imaging, vol.17, no.6, pp.857-871,1998) proposition is adopted the three-dimensional vascular skeleton difference downward left and right sides imaging plane back projection constantly that a certain moment reconstructs then optical flow method to obtain the two dimensional motion vector of vascular skeleton between adjacent moment, and then is reconstructed three-dimensional vector.This method can be introduced calibrated error and the matching error in the three-dimensional reconstruction once more in back projection's process, thereby result's precision is not high.Adopting optical flow method to carry out two dimensional motion simultaneously estimates, not only amount of calculation is big, but also exist noiseproof feature poor, to the frame sampling rate of image limitation such as have relatively high expectations, and the clinical captured image sequence of x-ray imaging imaging system is difficult to satisfy its requirement to picture quality and sample rate.Ding (" Quantification of 3-D coronary arterial motion using clinicalbiplane cineangiograms; " International Journal of Cardiac Imaging, vol.16, no.00, pp.331-346,2000) adopt by coarse nonuniform motion, by the window that comprises the target blood section is mated the tracking of realization to target in image to meticulous template matching method tracking arteria coronaria.Because blood vessel is long and narrow structure, no matter adopt rectangular window or circular window, blood vessel all only occupies a part very little in the window, therefore is difficult to set a window that comprises whole vessel segment.On the other hand, if select a series of wickets that vessel segment is followed the tracks of,, also be difficult to obtain satisfied result owing to lack enough constraints between the window.Grandson just (exercise question is " in the angiographic image sequences coronarius three-dimensional motion estimate ". carry biomedical engineering's magazine, vol.23, no.2, pp.428-432,2006) propose three-dimensional reconstruction and estimation are combined the method for carrying out: the skeleton that at first from original image, extracts main vessel branch, adopt matching method to estimate the two dimensional motion vector of each skeleton point between adjacent moment then, and adopt outer utmost point constrained optimization estimated result, reconstruct the three-dimensional motion vector at last.The result of this method depend on to a great extent that high-quality two dimension is extracted and two angle projections between accurate pointwise coupling.The former is because the general signal to noise ratio of coronary angiography image of clinical collection is lower, and the phenomenon that different blood vessel overlaps each other often appears in the image, therefore it is very difficult accurately extracting blood vessel structure from image, often need a large amount of of operator manually to participate in, existing full-automatic algorithm can not guarantee can both be successful to which kind of image; The latter generally adopts the foundation of the outer utmost point constraint conduct coupling in the computer vision, and the accuracy of outer polar curve coupling depends on the precision of imaging system geometric transformation, and the error of two-dimensional process introducing may cause the erroneous matching of pixel equally.The coordinate of three-dimensional point is to calculate according to two-dimentional corresponding point to try to achieve, and therefore the matching error of point will directly influence the precision of three-dimensional point coordinate.
Summary of the invention
The objective of the invention is to overcome the deficiencies in the prior art, propose a kind of precision height, the method for obtaining the coronary artery vasomotion information of easy and simple to handle, high efficiency.
The alleged problem of the present invention realizes with following technical proposals:
A kind of method of obtaining the coronary artery vasomotion information, it is on the basis of each constantly main vessel branch skeleton three-dimensional reconstruction in finishing X ray coronary angiography image sequence, adopt a kind of three-dimensional motion method of estimation based on elastic registrating, the skeleton line that the estimation of vascular skeleton was converted into the continuous moment mates, calculate motion vector and the movement locus of each vascular skeleton point in cardiac cycle, according to each the vascular skeleton point that estimates each motion vector constantly in sequence, extract the movable information of coronary artery in cardiac cycle, concrete steps are as follows:
A. gather X ray coronarography synchronous images sequence two angles, that cover one or more cardiac cycles, each vascular skeleton constantly in the sequence is carried out three-dimensional reconstruction, obtain the three-dimensional vascular skeleton sequence of representing with continuous B-spline curves;
B. adopt the method for elastic registrating, the estimation of adjacent moment skeleton in the three-dimensional vascular skeleton sequence be converted into coupling to the deeply ingrained stringing of consecutive hours, calculate the motion vector of each skeleton point:
The vessel branch skeleton that the 3D B-spline curves are represented carries out uniform sampling, obtains the different ordered set { s of skeleton point constantly in the sequence
i| s
i(m)=[x
i(m), y
i(m), z
i(m)], (m=0,1 ..., M
i-1; I=1,2 ..., T) }, and by making the matching error function
Minimum finds s
iAnd s
I-1Between Optimum Matching, thereby obtain the three-dimensional motion field.M wherein
iThe vascular skeleton that is i is constantly counted; T is the frame number of image sequence; α, β, γ are weight factors;
s
I-1 mExpression s
I-1In m point, s
i nExpression s
iIn n point; D
Shape=| τ
i(n)-τ
I-1(m) |+| κ
i(n)-κ
I-1(m) |, κ
i(x) and τ
i(x) be respectively that curve i is in some curvature at x place and a torsion; D
Index=| n-n ' |, n and n ' they are respectively s
iIn with s
I-1 mAnd s
I-1 M-1The sequence number of the point that is complementary.
The search to optimal value is finished in the employing dynamic programming, and the global optimum that assurance is separated and the stability of numerical calculation are very suitable for programming.Additional simultaneously uniqueness, monotonicity and displacement constrains, amount of calculation is reduced in the search volume of restriction feasible solution.
C. according to each the vascular skeleton point that estimates each motion vector constantly in sequence, extract and be included in each three-dimensional point the movable information in (starting point of motion vector and terminal point).
The above-mentioned method of obtaining the coronary artery vasomotion information, the extracting method of described movable information is:
Calculate the length L that promptly gets vessels axis on the moment t vascular skeleton line between consecutive points apart from sum
t, adopt relative mistake to express the relative variation of cardiac cycle medium vessels branched shaft line length;
By
Try to achieve the average speed in the Δ t time, by
Try to achieve acceleration, wherein, D
t T+1It is the amplitude of moment t vascular skeleton point motion vector;
Calculate vascular skeleton point from cardiac cycle first constantly to the end the total displacement in the moment obtain the length of movement locus;
The geometric center that adopts the set of three-dimensional arteria coronaria skeleton point is approximate as the center of gravity of arteria coronaria skeletal tree, calculate center of gravity from cardiac cycle first constantly to the end constantly total displacement obtain the movement locus of center of gravity;
The direction of motion: it is the local reference frame of zero with this point that each skeleton point is all set up one, with this in next coordinate transform of corresponding point constantly in local coordinate system, if the z coordinate after the conversion is a positive number, then this direction of motion is pointed to the direction away from the ventricular systole center, promptly expands; Otherwise then for shrinking;
Direction of rotation:, judge the direction of rotation of this point by judging the relative position relation of arbitrary skeleton point between two motion vectors of continuous three moment (two time periods);
Curvature and torsion: for the vascular skeleton r (t) that represents with B-spline curves, curvature by κ=| r ' (t) * r " (t) |/| r ' (t) |
3Try to achieve, torsion by τ=(r ', r ", r " ')/| r ' * r " |
2Try to achieve.
The present invention is on the basis of finishing each moment vascular skeleton three-dimensional reconstruction, adopt the method for elastic registrating, the three-dimensional vascular skeleton that the estimation of vascular skeleton sports ground is converted into adjacent moment mates, and by making suitable object function minimum, obtains global optimum's coupling.Comprised the measurement of measurement to the sports ground slickness, shape of blood vessel similarity in the object function and to the processing of no compatible portion.Need not any relevant arteriomotor priori, with the relevant hypothesis of motion essence, can be used for relatively serious deformation taking place, and distortion is estimated motion when being unknown when blood vessel.The search to Optimum Matching is finished in the employing dynamic programming, guarantees the global optimum of understanding and the stability of numerical calculation, is fit to very much the programming realization.Introduce corresponding constraints, the search volume of restriction feasible solution has been shortened more than ten times computation time at least.Afterwards, motion vector according to the vascular skeleton point between the adjacent moment that estimates, extract integral body coronarius and local kinematic parameter, comprise the center of gravity of length, movement locus, speed, acceleration, the direction of motion, direction of rotation, curvature, torsion, vascular tree of vessels axis and track thereof etc.The present invention not only extracts the precision height of vasomotion information, and easy and simple to handle, high efficiency.
Description of drawings
Fig. 1 is an implementation step flow chart of the present invention;
Fig. 2 is the corresponding relation sketch map between vascular skeleton point before and after the distortion;
Fig. 3 is the local coordinate system sketch map of three-dimensional vascular skeleton point;
Fig. 4 is the three-dimensional motion vector of same point in continuous two moment;
Fig. 5, Fig. 6 are two-dimensional projection's sketch maps of three-dimensional motion vector;
Fig. 7 be continuous three moment contrastographic picture to the three-dimensional reconstruction result of the main vessel branch skeleton of arteria coronaria;
Fig. 8 is the sports ground of vascular skeleton between the adjacent moment that estimates;
Fig. 9, Figure 10 are respectively any and the left 3D movement locus of being preced with the skeletal tree center of gravity on the circumflex branch at 2;
Figure 11 left and right sides two parts are respectively to expand in the time period and the motion vector of the point that shrinks;
Figure 12 left and right sides two parts are respectively to take place in this time period clockwise and the motion vector of the point that is rotated counterclockwise.
Each symbol is among the figure: s
1, constantly 1 three-dimensional vascular skeleton point is gathered; s
2, constantly 2 three-dimensional vascular skeleton point is gathered; s
1(m-1), s
1(m), s
1(m+p), s
1In m-1, m and m+p point; s
2(n
p), s
2(n), s
2(u (m+p)), s
2In respectively with s
1(m-1), s
1(m) and s
1(m+p) point that is complementary;
Point s
1(m-1) and s
1(m) motion vector of locating;
With
Between difference vector; OXYZ, imaging system coordinate system; P
i, P
I+1, two consecutive points on the three-dimensional vascular skeleton; P
iXl
iYl
iZl
i, with P
iLocal coordinate system for initial point; P
i, P '
i, respectively with P
iAnd P
I+1The point that obtains behind the translation Δ d (0＜Δ d＜＜1); Δ θ, a mistake point P
iAnd P '
iStraight line and the angle of imaging system coordinate system Y-axis forward (0＜Δ θ＜＜π/10); O
i, cross P
iAnd perpendicular to straight line P '
iP '
I+1The intersection point of vertical line; (x
t i, y
t i, z
t i), (x
T+1 i, y
T+1 i, z
T+1 i), (x
T+2 i, y
T+2 i, z
T+2 i), on the three-dimensional vascular skeleton a bit respectively at the coordinate (based on the imaging system coordinate system) of moment t, t+1 and t+2;
On the three-dimensional vascular skeleton a bit two continuous time section motion vector;
The two-dimensional projection of three-dimensional motion vector on the z=0 plane;
With
Vector product.
Used symbol: s in the literary composition
i, the set of i vascular skeleton point constantly; M
i, constantly the vascular skeleton of i is counted; The frame number of T, image sequence; s
I-1 m, s
I-1In m point; s
i n, s
iIn n point; C (m-1, n ', m, n), the matching error function; α, β, γ, weight factor; D
Disp, motion slickness constraint;
s
I-1In motion vector poor of consecutive points;
Motion vector between the skeleton point that matches each other; D
Shape, vascular skeleton change of shape; κ
i((x), curve i are in a curvature at x place; τ
i(x), curve i is in a torsion at x place; D
Index, jump bound term; N, s
iIn with s
I-1 mThe sequence number of the point that is complementary; N ', s
iIn with s
I-1 M-1The sequence number of the point that is complementary; L
t, the length of t vessels axis constantly; L
Set, the vessel branch set that each axial length is constantly formed in cardiac cycle; D
t T+1, the amplitude of t vascular skeleton point motion vector constantly; v
t T+1, the speed of t vascular skeleton point constantly; a
T+1, the acceleration of t vascular skeleton point constantly; R (t), vascular skeleton; κ, curvature; τ, torsion.
The specific embodiment
The invention will be further described below in conjunction with drawings and Examples.
As shown in Figure 1, treatment step of the present invention comprises:
(1) image acquisition and pretreatment:
Adopt C type arm single face X ray angioradiographic system to obtain coronarogram two angles, that cover a cardiac cycle at least, require two angles between the angle between 60 °～120 ° as sequence.Be recorded as picture systematic parameter (radiography angle, x-ray source is to the distance of imaging plane).It is right that the electrocardiosignal of employing synchronous recording is chosen the image of two angles of same phase in the cardiac cycle.
Original image is carried out necessary pretreatment, mainly comprise distortion correction, balanced contrast, remove noise and figure image intensifying etc.
(2) the three-dimensional sequence of vascular skeleton is rebuild:
(grandson just to adopt document, Yu Daoyin, Jiang Hao. the cardiovascular three-dimensional motion based on distorted pattern is followed the tracks of. photoelectric project, vol.33, no.6, pp.24-27+32,2006) described in method, the skeleton of each constantly main vessel branch in the reconstruction sequence obtains the three-dimensional vascular skeleton sequence of representing with B-spline curves respectively.
(3) three-dimensional motion of vascular skeleton is estimated:
The present invention adopts a kind of vascular skeleton three-dimensional motion method of estimation based on elastic registrating, and the skeleton line that estimation was converted into the continuous moment mates, and calculates each vascular skeleton point each motion vector constantly in sequence.Concrete grammar is as follows:
The vessel branch skeleton that the 3D B-spline curves are represented carries out uniform sampling, obtains the different ordered sets of skeleton point constantly in the sequence:
{s
i|s
i(m)＝[x
i(m)，y
i(m)，z
i(m)]，(m＝0，1，...，M
i-1；i＝1，2，...，T)}(1)
M wherein
iThe vascular skeleton that is i is constantly counted, and T is the frame number of image sequence.Because heart can expand or shrink, therefore different skeletons are constantly counted may be unequal, for any two set s
iAnd s
I-1, suppose M
i〉=M
I-1For the sake of simplicity, use s
I-1 mExpression s
I-1In m point, s
i nExpression s
iIn n point.
Adopt following matching error function:
By making (2) formula minimum, find s
iAnd s
I-1Between Optimum Matching, thereby obtain the three-dimensional motion field.Wherein α, β, γ are weight factors.This function comprises three parts:
First D
DispIt is the constraint of motion slickness.Consider that 2 the motion that physics links to each other on the same object is similar,, should make s therefore in order to guarantee the slickness of sports ground
I-1The motion vector of middle consecutive points changes minimum:
Wherein
Be s
I-1In motion vector poor of consecutive points:
With
Be the motion vector between the skeleton point that matches each other:
(as shown in Figure 2).
Second D
ShapeBe the shape constraining of vascular skeleton:
D
shape＝|τ
i(n)-τ
i-1(m)|+|κ
i(n)-κ
i-1(m)| (4)
κ wherein
i(x) and τ
i(x) be respectively curve i in some curvature at x place and a torsion, the shape that they can unique definite curve.This item constraint is the measurement of shape of blood vessel similarity, guarantees that the shape of corresponding skeleton point place, motion front and back blood vessel is similar.
The 3rd D
IndexBe to guarantee not have between the adjacent match point big jump:
D
index＝|n-n′| (5)
Wherein n and n ' are respectively s
iIn with s
I-1 mAnd s
I-1 M-1The sequence number of the point that is complementary.If establish M
i＞M
I-1, s so
iIn do not have coupling with regard to some point.Bound term D
IndexEffect be exactly to make the part of not having coupling as far as possible little, thereby obtain uniform matching result.
By making cost function (2) formula minimum, from s
iIn find out M
I-1Individual, they are s
I-1In M
I-1Individual point
Optimum Matching.Adopt dynamic programming (Geiger D, Gupta A, Vlontzos JA, Vlontzos J. " Dynamic programming for detecting; tracking and matching deformable contours ", IEEETransactions on Pattern Analysis and Machine Intelligence, vol.17, no.3, pp.294-302,1995) finish search to Optimum Matching, not only can guarantee global optimum's property of separating, and have the stable advantage of numerical calculation, be very suitable for programming.
For fear of comparing one by one, reduce search volume and amount of calculation, the inventive method has been added constraints, and the search volume of restriction feasible solution comprises uniqueness, monotonicity and displacement constraint:
Unique constraints: guarantee only to consider s
iIn the optimum matching of each element:
Monotonicity constraint: limited the search volume of feasible solution, promptly n 〉=m has avoided s
iIn a plurality of elements and s
I-1In an element be complementary and n
i＜n
I+1Avoided cross-matched.
Displacement constraint: search is restricted to n-m≤M
i-M
I-1
By anatomical structure coronarius as can be known, should match each other between each bifurcation of vascular tree and the end points between the adjacent moment, therefore for each vessel branch, can these ramose two end points as the starting point and the terminating point of registration.
(4) extraction of kinematic parameter:
The inventive method extracts and is included in each three-dimensional point to the movable information in (starting point of motion vector and terminal point) according to each the vascular skeleton point that estimates each motion vector constantly in image sequence, comprises whole and local kinematic parameter.
(4.1) mass motion parameter:
Mass motion parameter coronarius comprises the length of vessels axis, amplitude, speed and acceleration, movement locus and the length thereof of displacement, the center of gravity and the movement locus thereof of arteria coronaria skeletal tree.By analyzing the understanding of these parameters acquisitions to the arteria coronaria mass motion.
The length L of moment t vessels axis
t:
Calculate on the vascular skeleton line and promptly get L apart from sum between consecutive points
tBecause that puts on each vessel branch axis is total unequal,, adopt normalized expression for the ease of comparing: | L
t|=L
t/ max L
Set, L wherein
SetIt is the set that this branch forms in each axial length constantly.Adopt relative mistake Δ L
SetExpress the relative variation of arterial branch axial length:
ΔL
set＝[(max?L
set-min?L
set)/max?L
set]×100％ (6)
The amplitude D of displacement
t T+1, speed v
t T+1With acceleration a
T+1:
D
t T+1It is the amplitude of moment t vascular skeleton point motion vector.Average speed in the Δ t time is
Acceleration is
The length of movement locus: i.e. vascular skeleton point first total displacement in the moment to the end constantly from cardiac cycle.
The center of gravity of arteria coronaria skeletal tree and movement locus thereof: geometric center being similar to of adopting three-dimensional arteria coronaria skeleton point set as its center of gravity.Center of gravity from cardiac cycle first constantly to the end constantly total displacement be its movement locus.
(4.2) local motion parameter:
The direction of motion:
Vascular skeleton that three-dimensional reconstruction goes out and the motion vector that estimates all are based on the imaging system coordinate system.For the ease of analyzing the local motion direction, it is the local reference frame (accompanying drawing 3) of zero with this point that the present invention sets up one to each skeleton point.With this in next coordinate transform of corresponding point constantly in local coordinate system, by judging the positive and negative of z coordinate after the conversion, the motion that can judge this point is to shrink or expansion: if z is a positive number, then this direction of motion is pointed to the direction away from the ventricular systole center, promptly expands; Otherwise then for shrinking.
Direction of rotation:
Referring to accompanying drawing 4, accompanying drawing 5, accompanying drawing 6,, judge the direction of rotation of this point by judging the relative position relation of arbitrary skeleton point between two motion vectors (Fig. 4) of continuous three moment (two time periods).These two motion vectors are projected on the z=0 plane, obtain two bivectors
With
Calculate
If z
k＞0, so
With respect to
Be to be rotated counterclockwise (Fig. 5); Otherwise, then be turn clockwise (Fig. 6).
Curvature and torsion:
For the vascular skeleton r (t) that represents with B-spline curves, adopt curvature and torsion formula in the Differential Geometry:
κ＝|r′(t)×r″(t)|/|r′(t)|
3 (7)
τ＝(r′，r″，r″′)/|r′×r″|
2 (8)
Can directly calculate the curvature and the torsion at each skeleton point place, describe the bending and the degreeof tortuosity of blood vessel respectively.
Accompanying drawing 7～accompanying drawing 12 is to be that 15 frame/seconds, image size are the experimental result of the Clinical X ray left side hat contrastographic picture sequence that 512 * 512 (pixels), GTG are 256, Pixel Dimensions is 0.3mm with acquisition rate.The shooting angle of image sequence is respectively RAO30 ° CAUD24 ° and LAO46 ° CRAN21 °.Selected successive 10 frames, wherein the 3rd to 10 frame is a cardiac cycle, and the 1st and 2 frames belong to previous cardiac cycle.Because narrow sexually transmitted disease (STD) accommodation often betides major blood vessel branch, little branch having little significance in clinical diagnosis, therefore all diameters are all ignored less than the vessel branch that preestablishes threshold value in the experiment.Image sequence by two angles reconstructs the respectively three-dimensional framework of constantly main vessel branch, and optional three moment as shown in Figure 7.Adopt elastic registrating that the vascular skeleton between adjacent moment is carried out estimation, obtain the motion vector of each skeleton point, the result who chooses three time periods as shown in Figure 8.Accompanying drawing 9 and accompanying drawing 10 are respectively any and the left 3D movement locus (in the numeral moment on the trajectory, the unit of track is a millimeter) of being preced with the skeletal tree center of gravity on the LC (CX) at 2.Because the from the 3rd to the 10th moment is a cardiac cycle in the experiment sequence, therefore the position of the position of t=10 time point and t=3 is very approaching, that is to say that when one-period finished, the point on the blood vessel and the center of gravity of vascular tree can be got back to its position when the cycle begins.And the time and the spatial heterogeneity of arteria coronaria motion have also obtained checking.Accompanying drawing 11 left and right sides two parts are respectively to expand in the time period and the motion vector of the point that shrinks; Accompanying drawing 12 left and right sides two parts are respectively to take place in this time period clockwise and be rotated counterclockwise a little motion vector.
Claims (2)
1. method of obtaining the coronary artery vasomotion information, it is characterized in that, it is on the basis of each constantly main vessel branch skeleton three-dimensional reconstruction in finishing X ray coronary angiography image sequence, adopt a kind of vascular skeleton three-dimensional motion method of estimation based on elastic registrating, the skeleton line that the estimation of vascular skeleton was converted into the continuous moment mates, calculate motion vector and the movement locus of each vascular skeleton point in cardiac cycle, according to the three-dimensional motion vector of each the vascular skeleton point that estimates in each moment, extract integral body and the local movable information of coronary artery in cardiac cycle, concrete steps are as follows:
A. gather X ray coronary angiography image sequence two angles, that cover one or more cardiac cycles, each vascular skeleton constantly in the sequence is carried out three-dimensional reconstruction, obtain the three-dimensional vascular skeleton sequence of representing with continuous B-spline curves;
B. adopt the method for elastic registrating, the estimation of adjacent moment skeleton in the three-dimensional vascular skeleton sequence be converted into coupling to the deeply ingrained stringing of consecutive hours, calculate the motion vector of each skeleton point:
The vessel branch skeleton that three-dimensional B-spline curves are represented carries out uniform sampling, obtains the different ordered set { s of skeleton point constantly in the sequence
i| s
i(m)=[x
i(m), y
i(m), z
i(m)], (m=0,1 ..., M
i-1; I=1,2 ..., T) }, and by making the matching error function
Minimum finds s
iAnd s
I-1Between Optimum Matching, thereby obtain the three-dimensional motion field;
Described { s
i| s
i(m)=[x
i(m), y
i(m), z
i(m)], (m=0,1 ..., M
i-1; I=1,2 ..., T) } be the different ordered sets of vascular skeleton point constantly in the expression sequence, wherein, m represents the sequence number of vascular skeleton point; The sequence number of each frame in the i presentation video sequence; M
iIt is the sum of the vascular skeleton point of moment i; T is the totalframes in the image sequence; s
iExpression is the vascular skeleton point set of i constantly; s
i(m) m point on the vascular skeleton of expression moment i; [x
i(m), y
i(m), z
i(m)] three-dimensional coordinate of m point on the vascular skeleton of expression moment i;
(m n) is the matching error function to described C for m-1, n ', and wherein, m-1 and m represent the vascular skeleton point set s of i constantly respectively
iIn m-1 the some s
i(m-1) and m the some s
i(m) sequence number; N ' and n represent the vascular skeleton point set s of i+1 constantly respectively
I+1In and s
i(m-1) and s
iThe sequence number of the point that (m) is complementary;
Described α, β, γ are weight factors,
s
I-1 mExpression s
I-1In m point, s
i nExpression s
iIn n point, D
Shape=| τ
i(n)-τ
I-1(m) |+| κ
i(n)-κ
I-1(m) |, κ
i(x) and τ
i(x) be respectively that curve i is at some curvature at x place and a torsion, D
Index=| n-n ' |, n and n ' they are respectively s
iIn with s
I-1 mAnd s
I-1 M-1The sequence number of the point that is complementary;
C. according to each the vascular skeleton point that estimates each motion vector constantly in image sequence, extract the movable information that is included in each three-dimensional point centering.
2. the method for obtaining the coronary artery vasomotion information according to claim 1 is characterized in that, the extracting method of described movable information is:
Calculate the length L that obtains vessels axis on the moment t vascular skeleton line between consecutive points apart from sum
t
By
Try to achieve the average speed in the Δ t time, by
Try to achieve acceleration, wherein, D
t T+1It is the amplitude of moment t vascular skeleton point motion vector;
Calculate vascular skeleton point from cardiac cycle first constantly to the end the total displacement in the moment obtain the length of movement locus;
The geometric center that adopts the set of three-dimensional arteria coronaria skeleton point is approximate as the center of gravity of arteria coronaria skeletal tree, calculate center of gravity from cardiac cycle first constantly to the end constantly total displacement obtain the movement locus of center of gravity;
The direction of motion: it is the local reference frame of zero with this point that each skeleton point is all set up one, with this in next coordinate transform of corresponding point constantly in local coordinate system, if the z coordinate after the conversion is a positive number, then this direction of motion is pointed to the direction away from the ventricular systole center, promptly expands; Otherwise then for shrinking;
Direction of rotation:, judge the direction of rotation of this point by judging the relative position relation of arbitrary skeleton point between two motion vectors in continuous three moment;
Curvature and torsion: for the vascular skeleton r (t) that represents with B-spline curves, curvature by κ=| r ' (t) * r " (t) |/| r ' (t) |
3Try to achieve, torsion by τ=(r ', r ", r ' ")/| r ' * r " |
2Try to achieve.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100550364A CN101283910B (en) | 2008-06-05 | 2008-06-05 | Method for obtaining the coronary artery vasomotion information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100550364A CN101283910B (en) | 2008-06-05 | 2008-06-05 | Method for obtaining the coronary artery vasomotion information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101283910A CN101283910A (en) | 2008-10-15 |
| CN101283910B true CN101283910B (en) | 2010-06-09 |
Family
ID=40056480
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008100550364A Expired - Fee Related CN101283910B (en) | 2008-06-05 | 2008-06-05 | Method for obtaining the coronary artery vasomotion information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101283910B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101625757B (en) * | 2009-08-13 | 2012-02-22 | 华北电力大学（保定） | Method for inhibiting rigid motion artifact in ultrasonic image sequence in coronary artery |
| CN102346803A (en) * | 2010-07-28 | 2012-02-08 | 北京集翔多维信息技术有限公司 | Cardioangiographic image analysis system |
| CN105105775B (en) * | 2011-07-19 | 2018-11-09 | 东芝医疗系统株式会社 | Cardiac motion resolver |
| CN102622775B (en) * | 2012-01-12 | 2016-03-09 | 北京理工大学 | A kind of real-time dynamic reconstruction technology of heart compensated based on model interpolation |
| JP6073562B2 (en) * | 2012-02-24 | 2017-02-01 | 東芝メディカルシステムズ株式会社 | Medical image processing device |
| CN102722882B (en) * | 2012-04-12 | 2015-06-03 | 华北电力大学（保定） | Elastic registration method of CAG image sequence |
| JP6381875B2 (en) * | 2012-08-16 | 2018-08-29 | キヤノンメディカルシステムズ株式会社 | Image processing apparatus, medical image diagnostic apparatus, and blood pressure monitor |
| CN103886615B (en) * | 2013-12-31 | 2016-04-20 | 华中科技大学 | The separating estimation method of parameter of doing more physical exercises in a kind of X ray angiographic image |
| CN107233106A (en) * | 2017-06-01 | 2017-10-10 | 上海交通大学 | Blood vessel correspondence position relation search method and system under multi-angle radiography |
| CN108564562B (en) * | 2018-01-05 | 2022-06-17 | 苏州润迈德医疗科技有限公司 | Cardiac cycle acquisition method based on X-ray coronary angiography image |
| US11191503B2 (en) * | 2018-07-17 | 2021-12-07 | International Business Machines Corporation | Fluid-injector for a simultaneous anatomical and fluid dynamic analysis in coronary angiography |
| TWI770235B (en) * | 2018-07-20 | 2022-07-11 | 巫湘沂 | Method for judging blood flow change and vascular obstruction area by dynamic images |
| CN109146872B (en) * | 2018-09-03 | 2021-12-17 | 北京邮电大学 | Heart coronary artery image segmentation and identification method based on deep learning and optical flow method |
| CN112912005A (en) * | 2018-10-24 | 2021-06-04 | 西门子医疗有限公司 | Generating medical result images |
| CN111435532B (en) * | 2019-01-14 | 2021-06-22 | 湖南大学 | Method for detecting tree-like structure end point in digital image |
| CN111009032B (en) * | 2019-12-04 | 2023-09-19 | 浙江理工大学 | Vascular three-dimensional reconstruction method based on improved epipolar line constraint matching |
| CN111870224B (en) * | 2020-07-16 | 2022-05-20 | 武汉大学 | Tumor blood vessel normalization detection system and detection method |
| CN112233158B (en) * | 2020-10-14 | 2022-02-15 | 俐玛精密测量技术(苏州)有限公司 | Secondary projection registration method of micro-nano CT projection image |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005008597A2 (en) * | 2003-07-16 | 2005-01-27 | Philips Intellectual Property & Standards Gmbh | Reconstruction of an image of a moving object from volumetric data |
| WO2005081168A2 (en) * | 2004-02-03 | 2005-09-01 | Siemens Medical Solutions Usa, Inc. | Systems and methods for automated diagnosis and decision support for heart related diseases and conditions |
| WO2006076698A1 (en) * | 2005-01-12 | 2006-07-20 | King Pharmaceuticals Reserch & Development, Inc. | Method of detecting myocardial dysfunction in patients having a history of asthma or bronchospasm |
| CN101181160A (en) * | 2006-11-17 | 2008-05-21 | 株式会社东芝 | Image display method and image display apparatus |
-
2008
- 2008-06-05 CN CN2008100550364A patent/CN101283910B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005008597A2 (en) * | 2003-07-16 | 2005-01-27 | Philips Intellectual Property & Standards Gmbh | Reconstruction of an image of a moving object from volumetric data |
| WO2005081168A2 (en) * | 2004-02-03 | 2005-09-01 | Siemens Medical Solutions Usa, Inc. | Systems and methods for automated diagnosis and decision support for heart related diseases and conditions |
| WO2006076698A1 (en) * | 2005-01-12 | 2006-07-20 | King Pharmaceuticals Reserch & Development, Inc. | Method of detecting myocardial dysfunction in patients having a history of asthma or bronchospasm |
| CN101181160A (en) * | 2006-11-17 | 2008-05-21 | 株式会社东芝 | Image display method and image display apparatus |
Non-Patent Citations (2)
| Title |
|---|
| 孙正，郁道银.基于活动轮廓模型的冠状动脉血管段运动跟踪.生物医学工程学杂志24 1.2007,24(1),9-14. |
| 孙正,郁道银.基于活动轮廓模型的冠状动脉血管段运动跟踪.生物医学工程学杂志24 1.2007,24(1),9-14. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101283910A (en) | 2008-10-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101283910B (en) | Method for obtaining the coronary artery vasomotion information | |
| US11615894B2 (en) | Diagnostically useful results in real time | |
| CN101283911B (en) | Four dimensional rebuilding method of coronary artery vessels axis | |
| CN105105775B (en) | Cardiac motion resolver | |
| CN102722882B (en) | Elastic registration method of CAG image sequence | |
| JP3803404B2 (en) | Computer-aided 3D image identification verification method | |
| CN111009032B (en) | Vascular three-dimensional reconstruction method based on improved epipolar line constraint matching | |
| Zheng et al. | Sequential reconstruction of vessel skeletons from X-ray coronary angiographic sequences | |
| EP2946319A1 (en) | Calculating a fractional flow reserve | |
| CN102999902A (en) | Optical navigation positioning system based on CT (computed tomography) registration results and navigation method thereby | |
| AU2017358321B2 (en) | Method and system for modelling a human heart and atria | |
| Ding et al. | Quantification of 3-D coronary arterial motion using clinical biplane cineangiograms | |
| CN105741299A (en) | Coronary artery CT angiography image segmentation method | |
| CN107072531A (en) | Method and system for the dynamic (dynamical) analysis of myocardial wall | |
| JP2002306483A (en) | Medical diagnosis image processing equipment and method thereof | |
| Ruan et al. | Three-dimensional motion and reconstruction of coronary arteries from biplane cineangiography | |
| Blondel et al. | Automatic trinocular 3D reconstruction of coronary artery centerlines from rotational X-ray angiography | |
| CN103340602A (en) | Method for optimizing optimal viewing angle of multiple branch interesting blood vessel section | |
| Zheng et al. | Motion estimation of 3D coronary vessel skeletons from X-ray angiographic sequences | |
| CN104361606B (en) | Cardiac cycle restoration methods in x-ray imaging image sequence | |
| CN100438831C (en) | Time registration method of multi-view point CAG sequential images | |
| Wang et al. | Study of lumbar spine activity regularity based on Kanade-Lucas-Tomasi algorithm | |
| CN107812325A (en) | The method being tracked using motion of the magnetic resonance imaging to internal target | |
| Sun et al. | Reconstruction of three-dimensional coronary arterial dynamics from multi-view angiogram sequences | |
| Haddad et al. | A realistic anthropomorphic dynamic heart phantom |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100609 Termination date: 20140605 |