CN103054563A - Vascular wall pathological changes detection method - Google Patents

Abstract

The invention provides a method for quantifying and extracting image texture characteristics of vascular wall and is used for early recognition and diagnosis of vascular wall pathological changes. The method includes steps of acquiring vascular wall images of multiple continuous frames by means of a medical imaging system; selecting an inner membrane-intermediate membrane layer of the vascular wall as an interested area from the vascular wall images of each frame, extracting texture characteristic values of the interested area, and acquiring average value of the texture characteristics values; and computing and analyzing the texture characteristic values and establishing a texture characteristic parameter group used for early recognition and diagnosis of the vascular wall pathological changes. The method for quantifying and extracting image texture characteristics of the vascular wall is capable of extracting the texture characteristic values of the vascular wall in any interested area, is beneficial to accurate positioning of pathological changes during pathological changes detection of the vascular wall, high in applicability, free of hardware upgrading in existing clinical imaging system, low in cost, easy to accept by hospitals and doctors and convenient to clinically popularize.

Description

Blood vessel wall lesion detection method

Technical field

The present invention relates to quantification and the extracting method of blood vessel wall image texture characteristic, relate in particular to a kind of quantification and extracting method of the blood vessel wall image texture characteristic based on image texture characteristic.

Background technology

Cardiovascular and cerebrovascular disease serious harm human health has become day by day urgent great social common problem.Atherosclerosis is the arch-criminal of cardiovascular and cerebrovascular disease.Atherosclerosis is the process of a long-term and complex, if can be in early days it be realized accurate identification and diagnosis, and just can intervening timely in early days and treat Effective Raise cure rate, reduction medical expense in disease.Judge clinically that at present atherosclerotic Main Basis is blood vessel structure, function and Hemodynamic Factors, be media thickness in the blood vessel, the elasticity that has or not speckle, luminal stenosis degree, tremulous pulse and blood flow rate and flow etc., the Main Diagnosis method has digital subtraction angiography, magnetic resonance angiography, CT angiography, pulse wave velocity and color doppler ultrasonography.These methods can to the structure of blood vessel, function, and hemodynamics detect.

In the present noninvasive detection method, media thickness (IMT) measurement method is based on ultrasound image acquisition in the blood vessel, ultra sonic imaging is simple to operation, but image resolution ratio is limited, and because the image that gathers is video signal, the quantity of information that image can provide and certainty of measurement all are vulnerable to impact, even degree of accuracy also can only reach 0.1mm after amplifying, the IMT error that records is larger.Pulse wave velocity (PWV) measurement method be heart with the blood pulsation inject aorta, aorta wall produces pulse pressure wave, and conduct to peripheral blood vessel along blood vessel wall with certain speed, by measuring two pulse wave translation time and distances between the tremulous pulse record position, can calculate PWV.But the sensitivity of PWV is relatively poor, be difficult for to find that slight arterial elasticity changes, and PWV is when measuring, and the body surface range measurement has error, can obviously affect the accuracy of data.Arm index (ABI) measurement method be by manual method or by automatic checkout equipment measure respectively the two artery of upper extremity of patient and bilateral posterior tibial artery and (or) the dorsal artery of foot systolic pressure, calculate arm index (ABI) value.Yet it is narrow or inaccessible that the reduction of arm index (ABI) can only point out the above blood vessel of ankle to have, and can not accurately locate pathological changes.Therefore PWV and ABI are merely able to measure the average hardness of one section blood vessel.Flow-mediated vasodilation (FMD) detection technique be a kind of depend on have high-resolution two dimensional image, M type or A type, the method that obtaining information is recorded and analyzed.This inspection can reflect and judge arterial vascular endothelial function because of it, so can make diagnosis to early stage arteriosclerosis.But because of its operation more loaded down with trivial details, higher to operator's requirement, make its in clinical application, popularize and to be affected.Therefore, these detection methods the blood vessel structure function not yet change early stage, usually all can't effectively differentiate vascular lesion.

Summary of the invention

For the problems referred to above, the purpose of this invention is to provide a kind of quantification and extracting method of blood vessel wall image texture characteristic, be used for the EARLY RECOGNITION diagnosis of blood vessel wall pathological changes.

A kind of quantification of blood vessel wall image texture characteristic and extracting method, it may further comprise the steps:

Utilize medical image system to gather the continuous blood vessel wall image of multiframe;

Choose blood vessel wall inner membrance-middle rete as area-of-interest at the described blood vessel wall image of each frame, and extract the textural characteristics value of described area-of-interest, then obtain its meansigma methods;

Described textural characteristics value is carried out statistical analysis, make up the textural characteristics population of parameters that is used for the diagnosis of blood vessel wall pathological changes EARLY RECOGNITION.

In the present invention's one preferred embodiments, described medical image system is ultrasonic image-forming system, optical imaging system, CT imaging system or MRI imaging system.

In the present invention's one preferred embodiments, described textural characteristics value comprises first-order statistical properties, fractal dimension texture analysis, gray level co-occurrence matrixes, grey scale difference statistics, local gray level difference matrix and statistical nature matrix.

In the present invention's one preferred embodiments, described first-order statistical properties comprises gray average and the standard deviation of vascular wall area.

In the present invention's one preferred embodiments, described fractal dimension texture analysis parameter is fractal dimension.

In the present invention's one preferred embodiments, described gray level co-occurrence matrixes feature comprises contrast, dependency, energy, unfavourable balance distance, entropy.

In the present invention's one preferred embodiments, described grey scale difference statistical nature comprises contrast, angle direction second moment, entropy, meansigma methods.

In the present invention's one preferred embodiments, described local gray level difference matrix feature comprises roughness, contrast, frequency, complexity, texture strength.

In the present invention's one preferred embodiments, described statistical nature matrix character comprises contrast, covariance, diversity.

In the present invention's one preferred embodiments, the quantification of described blood vessel wall image texture characteristic and extracting method further comprise and adopt grader that the blood-vessel image of healthy and pathological changes is classified.

Compared to prior art, quantification and the extracting method of described blood vessel wall image texture characteristic provided by the invention have the following advantages: one, the suitability are stronger, applicable to the image of the fast imaging system acquisition such as various medical image systems and optics, optoacoustic.Two, can extract the textural characteristics value of any area-of-interest blood vessel wall, when helping the blood vessel wall lesion detection pathological changes accurately be located.Three, the blood-vessel image that directly gathers based on medical image system of the quantification of described blood vessel wall image texture characteristic and extracting method, clinician and the operator with certain medical imaging devices use experience can collect required image, are difficult for being affected by other factors.Four, the quantification of described blood vessel wall image texture characteristic and extracting method can be used as a post processing of image software module and are integrated in the existing medical image system, to promote the function of medical image system, need not existing clinical imaging system is carried out HardwareUpgring, upgrade cost is low, accepted by hospital and doctor easily, make things convenient for clinical expansion.

Above-mentioned explanation only is the general introduction of technical solution of the present invention, for can clearer understanding technological means of the present invention, and can be implemented according to the content of description, and for above and other objects of the present invention, feature and advantage can be become apparent, below especially exemplified by embodiment, and the cooperation accompanying drawing, be described in detail as follows.

Description of drawings

The quantification of the blood vessel wall image texture characteristic that Fig. 1 provides for one embodiment of the invention and the flow chart of extracting method.

Fig. 2 is for adopting the high frequency ultrasound imaging system to the sketch map of the total arteriography of mouse carotid.

Fig. 3 is for selecting blood vessel wall inner membrance-middle rete as the sketch map of area-of-interest.

The B ultrasonic image of Fig. 4 for adopting the high frequency ultrasound imaging system that the common carotid artery of anesthetized mice is carried out the sketch map of imaging and obtains.

Fig. 5 is for adopting healthy mice common carotid artery and the Carotid B ultrasonic image of apoe knock out mice of high frequency ultrasound imaging system acquisition.

Fig. 6 be with 14 effective textural characteristics populations of parameters as the characteristic of division collection, the sketch map that adopts the KNN grader that healthy mice and the Carotid ultrasonoscopy of pathological changes mice are classified.

The specific embodiment

The present invention is further detailed explanation below in conjunction with drawings and the specific embodiments.

See also Fig. 1, one embodiment of the invention provides a kind of quantification and extracting method of blood vessel wall image texture characteristic, and it may further comprise the steps:

Step S101, utilize medical image system to gather the continuous blood vessel wall image of multiframe.

In the present embodiment, adopt medical image system to gather the continuous blood vessel wall image of N frame, N is the image of containing at least one cardiac cycle, and for example, the image acquisition frame frequency FR of medical image system is that 100 frame/seconds, human body palmic rate f are 60Hz/ minute, cardiac cycle Tc=60/f=1 second; N=m * FR * Tc=100m frame (m=1,2,3 then ...).Be that N should be 100 integral multiple, as shown in Figure 2.

Be understandable that described medical image system can be ultrasonic image-forming system, optical imaging system (such as X-ray machine), CT imaging system or MRI imaging system etc.In the present embodiment, describe as an example of the ultrasonoscopy of ultrasonic image-forming system collection example.

Step S103, choose blood vessel wall inner membrance-middle rete as area-of-interest at the described blood vessel wall image of each frame, and extract the textural characteristics value of described area-of-interest, then obtain the meansigma methods of described textural characteristics value.

Be understandable that the blood vessel wall inner membrance is made of subcutaneous connective tissue in simple squamous epithelium cell and the skim, be continuously smooth equal echo light belt of a fine rule shape at ultrasonoscopy; Rete is a blanking bar in the blood vessel wall, mainly is made of smooth muscle cell and elastic [connective, and the blood vessel wall theca externa is made of blood vessel wall outermost layer loose connective tissue, is the euphotic zone more bright than inner membrance.Therefore, blood vessel wall shows as typically " two-wire bar " at ultrasonoscopy, and namely two parallel strong echoes are separated by a low echo area or no echo area.In the present embodiment, selecting " two-wire bar " at ultrasonoscopy is area-of-interest, as shown in Figure 3.

In the present embodiment, the textural characteristics value of described area-of-interest comprises first-order statistical properties, fractal dimension texture analysis, gray level co-occurrence matrixes, grey scale difference statistics, local gray level difference matrix and statistical nature matrix, below will be introduced these textural characteristics values.

1, described first-order statistical properties is called again grey level histogram, the general evaluation system characteristic of its reflection piece image intensity profile, and it mainly comprises gray average and the standard deviation of vascular wall area.

In the present embodiment, establish described region of interest area image (gray value of any point (x, y) is I (x, y) among the M * N), and then the meansigma methods of described area-of-interest gray scale is:

$\mathrm{ROIavg}=\frac{1}{M\×N}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}\underset{y=1}{\overset{N}{\mathrm{\Σ}}}I(x,y)---\left(1\right)$

Wherein, ROIavg characterizes the average gray value of area-of-interest.

The standard deviation of area-of-interest gray scale is:

$\mathrm{ROIstd}=\sqrt{\frac{1}{M\×N-1}\underset{x=1}{\overset{M}{\mathrm{\Σ}}}\underset{y=1}{\overset{N}{\mathrm{\Σ}}}{(I(x,y)-\mathrm{ROIavg})}^2}---\left(2\right)$

Wherein, ROIstd has characterized the uniformity coefficient of intensity profile in the described area-of-interest, and ROIstd is less, and then intensity profile is more even.

2, fractal dimension is as the tolerance of imaging surface degree of irregularity, and it is consistent with human vision to the perception of imaging surface coarse texture degree, and namely fractal dimension is larger, and corresponding imaging surface is more coarse; Otherwise the imaging surface of the less correspondence of fractal dimension is more smooth.In the present embodiment, described fractal dimension texture analysis parameter is fractal dimension.

Be understandable that the fractal dimension D of the interesting image regions that will estimate in theory _fDetermined by following formula:

E(ΔI ²)=c(Δr) ^(6-2Df) （3）

In the formula, E(.) be expected value; Δ I is the variation of gray value, Δ I=I (x ₂, y ₂)-I (x ₁, y ₁);

Δ r is the variation of space length, Δ r=I (x ₂, y ₂)-I (x ₁, y ₁); C is constant.Formula (3) can be reduced to:

E(|ΔI|)=k(Δr) ^H （4）

In the formula, k=E (| Δ I|) _{Δ r=1}, D then _f=3-H takes the logarithm simultaneously to formula (4) both sides, then has:

logE(|ΔI|)=logk+Hlog(Δr) （5）

Under two-dimensional coordinate system, as vertical coordinate, log (Δ r) is abscissa, draws some discrete points, and discrete point is fitted to straight line, utilizes at last method of least square to obtain the slope of straight line, is H take logE (| Δ I|).Therefore can calculate fractal dimension D _f

3, the gray level co-occurrence matrixes feature is to describe one of modal method of textural characteristics, its main purpose be in the statistical picture pixel and gray scale in the distribution situation in space.

Its computational methods are as follows: the gray value scope of establishing image is [0, L], by calculating gray level co-occurrence matrixes, obtains a size and is (L+1) feature texture matrix (L+1).In the method, must consider two very important parameters: apart from d and angle θ, if select different θ then can obtain different feature texture matrixes from d.About angle θ, be to define like this, establish I (x, y) be in the gray value intensity of coordinate for (x, y) position, and I (x, y) be 1 with the distance of 8 neighbors on every side, under this definition, I (x, y), I (x+1, y) and I (x-1, y) angle of three pixels is 0 degree, I (x, y), the angle of I (x+1, y+1) and three pixels of I (x-1, y-1) is 45 degree, the rest may be inferred I (x, y), I (x, y+1) and the angle of I (x, y-1) be 90 the degree, I (x, y), the angle of I (x-1, y+1) and I (x+1, y-1) is 135 degree.#

(any point (x, y) and depart from its another point (x+a, y+b) among the M * N), the gray value of these 2 correspondences is (i, j) to get image.Make (x, y) point mobile at whole picture, then can obtain different (i, j) value, the progression of establishing gray value is the then square kind of the total k1 of combination of (i, j) of k1.For whole picture, count the number of times that each (i, j) value occurs, then be arranged in a square formation, the total degree of using again (i, j) to occur is normalized to the probability P (i, j) of appearance with them, and this square formation namely becomes gray level co-occurrence matrixes.

Gray level co-occurrence matrixes is carried out following normalization:

$P(i,j)=\frac{\mathrm{\Σ}(i,j)}{R},$

According to normalized gray level co-occurrence matrixes, can obtain following textural characteristics parameter: contrast (Contrast), dependency (Correlation), energy (Energy), unfavourable balance are apart from (Homogeneity) and entropy (Entropy).Wherein:

Contrast $\mathrm{Contrast}=\underset{n=0}{\overset{L}{\mathrm{\Σ}}}{n}^2\left(\underset{i=0}{\overset{L}{\mathrm{\Σ}}}\underset{j=0}{\overset{L}{\mathrm{\Σ}}}p(i,j)\right),|i-j|=n---\left(7\right)$

The definition of contrast reflection image and the degree of the texture rill depth.The texture rill is darker, and its contrast is larger, and visual effect is more clear; Otherwise contrast is little, and then rill is shallow, and effect is fuzzy.

Dependency $\mathrm{Correlation}=\frac{\underset{i=0}{\overset{L}{\mathrm{\Σ}}}\underset{j=0}{\overset{L}{\mathrm{\Σ}}}p(i,j)-u_x{u}_y}{{\mathrm{\σ}}_x{\mathrm{\σ}}_y}---\left(8\right)$

Wherein, u _xAnd σ _xP _xMeansigma methods and standard deviation, u _yAnd σ _yP _yMeansigma methods and standard deviation, and p _xWith p _yBe defined as follows:

$p_x\left(i\right)=\underset{j=0}{\overset{L}{\mathrm{\Σ}}}p(i,j)---\left(9\right)$

$p_y\left(j\right)=\underset{i=0}{\overset{L}{\mathrm{\Σ}}}p(i,j)---\left(10\right)$

Relativity measurement space gray level co-occurrence matrixes element be expert at or column direction on similarity degree, therefore, the dependency size has reflected local gray level dependency in the image.When the matrix element value evenly equated, dependency was just large; On the contrary, if matrix pixel value differs greatly then dependency is little.

Energy $\mathrm{Energy}=\underset{i=0}{\overset{L}{\mathrm{\Σ}}}\underset{j=0}{\overset{L}{\mathrm{\Σ}}}{\left(p(i,j)\right)}^2---\left(11\right)$

Energy is a be evenly distributed tolerance of degree and texture thickness of gradation of image, and is careful when the image texture strand, when intensity profile is even, and energy value is larger, otherwise, less.

The unfavourable balance distance $\mathrm{Homogeneity}=\underset{i=0}{\overset{L}{\mathrm{\Σ}}}\underset{j=0}{\overset{L}{\mathrm{\Σ}}}\frac{p(i,j)}{1+{(i-j)}^2}---\left(12\right)$

Unfavourable balance is apart from readability and the regular degree of reflection texture, clean mark, regular strong, be easy to describe, contrary gap value is larger; Rambling, be difficult to describe, contrary gap value is less.

Entropy $\mathrm{Entropy}=\underset{i=0}{\overset{L}{\mathrm{\Σ}}}\underset{j=0}{\overset{L}{\mathrm{\Σ}}}p(i,j)\log \left(p(i,j)\right)---\left(13\right)$

Entropy is the tolerance of the quantity of information that has of Description Image, shows the complicated process of image, and when complicated process was high, entropy was larger, otherwise then less.#

4, the ultimate principle of grey scale difference statistic law is the grey scale change situation of describing between each pixel of texture image and the adjacent image point thereof.

Image f's (x, y) is (x, y) a bit, and it is grey scale difference that there is the gray scale difference value Δ I of the point (x+ Δ x, y+ Δ y) of slight distance σ=(Δ x, Δ y) this point and it.If might a value total m level of the institute of gray scale difference score value then forms the vector of a m dimension, calculate the rectangular histogram of Δ I, by rectangular histogram as can be known gray scale difference value be the probability density p of i _σ(i).When σ gets smaller value, and gray scale difference value is the probability density p of i _σWhen (i) large, illustrate that image texture is more coarse, otherwise the explanation texture is thinner.Described grey scale difference statistics comprises contrast (Contrast), angle direction second moment (Angular Second Moment), entropy (Entropy), meansigma methods (Mean).Wherein:

Contrast $\mathrm{con}=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{i}^2{p}_{\mathrm{\δ}}\left(i\right)---\left(14\right)$

Contrast is grey scale difference probability density p _σSecond moment, for example, p _σThe moment of inertia about initial point.#

The angle direction second moment $\mathrm{ASM}=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}[p_{\mathrm{\δ}}\left(i\right){]}^2---\left(15\right)$

Work as p _σThe value homogeneous phase simultaneously, ASM is minimum; And work as p _σValue when differing greatly, ASM is larger.

Entropy $\mathrm{ENT}=-\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{p}_{\mathrm{\δ}}\left(i\right)\log p_{\mathrm{\δ}}\left(i\right)---\left(16\right)$

Work as p _σThe value homogeneous phase simultaneously, ENT is maximum; And work as p _σValue when differing greatly, ENT is less.

Meansigma methods $\mathrm{MEAN}=\frac{1}{m}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{p}_{\mathrm{\δ}}\left(i\right)---\left(17\right)$

Work as p _σValue close to zero the time, MEAN is less; Otherwise MEAN is larger.

5, local gray level difference matrix feature is established window (generally the getting 3 or 5) W=2d+1 of gray level image f (k, l) and W * W, the matrix of averaging

$\stackrel{\‾}{A}(k,l)=\frac{1}{W^2-1}\left[\underset{m=-d}{\overset{d}{\mathrm{\Σ}}}\underset{n=-d}{\overset{d}{\mathrm{\Σ}}}f(k+m,l+n)\right],(m,n)\≠\left(\mathrm{0,0}\right)---\left(18\right)$

Ask S (i), S (i) expression gray scale is all pixels and corresponding Mean Matrix of i again The absolute value sum of difference:

Described local gray level difference matrix feature comprises roughness (Coarseness), contrast (Contrast), frequency (Busyness), complexity (Complexity), texture strength (Texture Strength).Wherein:

Roughness $f_{\cos }=[\mathrm{\ϵ}+\underset{i=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}P\left(i\right)S\left(i\right){]}^{-1}---\left(20\right)$

G in the formula _hBe gray value maximum in the gray level image; ε is less constant, stops f _CosBe infinitary value; P (i) expression gray scale is the probability that the pixel of i occurs.f _CosIt is less to be worth larger interval scale gray difference, and it is more coarse to be image ratio.

Contrast $f_{\mathrm{con}}=\left[\frac{1}{N_g(N_g-1)}\underset{i=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}\underset{j=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}{P}_i{P}_j{(i-j)}^2\right]\left[\frac{1}{n^2}\underset{i=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}S\left(i\right)\right]---\left(21\right)$

N in the formula _gSum for gray difference is expressed as:

$N_g=\underset{i=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}{Q}_i,$ $Q_i=\left\{\begin{array}{cc}1& P\left(i\right)\≠0\\ 0& P\left(i\right)=0\end{array}\right.---\left(22\right)$

f _ConThe larger representative adjacent area of value gray-scale intensity differs greatly.

Frequency $f_{\mathrm{busy}}=\left[\underset{i=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}P\left(i\right)S\left(i\right)\right]/\left[\underset{i=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}\underset{j=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}\right|\mathrm{ip}\left(i\right)-\mathrm{jp}\left(j\right)\left|\right]---\left(23\right)$

Frequency reflects the difference degree of the grey scale pixel value around a certain pixel and its, when the spatial variations of gray scale is larger, and f _BusyBe worth higher.

Complexity $f_{\mathrm{com}}=\underset{i=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}\underset{j=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}(\left(\right|i-j\left|\right)/\left(n^2(p\left(i\right)+p\left(j\right))\right))(P\left(i\right)S\left(i\right)+P\left(j\right)S\left(j\right))---\left(24\right)$

The visual information amount of assorted degree reflection image texture, when quantity of information is larger, f _ComBe worth larger.

Texture strength $f_{\mathrm{str}}=\left[\underset{i=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}\underset{j=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}(P\left(i\right)+P\left(j\right)){(i-j)}^2\right]/[\mathrm{\ϵ}+\underset{i=0}{\overset{\mathrm{Gh}}{\mathrm{\Σ}}}S\left(i\right)]---\left(25\right)$

The pixel that comprises when image is easily definition, apparent, and the fstr value is larger.

6, statistical nature matrix character, statistical nature matrix can be used in the computed image pixel to the statistical property of some distances.Make (x, y) in the image a bit, I (x, y) is the gray value of this point, δ=(Δ x, Δ y) is the space length vector.Described statistical nature matrix character comprises contrast (Contrast), covariance (Covariance), diversity (Dissimilarity).Wherein:

Contrast C ON (δ)=E{[I (x, y)-I (x+ Δ x, y+ Δ y)] ²(26)

In the formula, E{} is expected value.

Covariance COV (δ)=E{[[I (x, y) η] [I (x+ Δ x, y+ Δ y) η] } (27)

In the formula, η is the gradation of image meansigma methods.

Diversity factor DSS (δ)=E{|I (x, y)-I (x+ Δ x, y+ Δ y) | } (28)

Step S105, described textural characteristics value is carried out statistical analysis, make up the textural characteristics population of parameters that is used for the diagnosis of blood vessel wall pathological changes EARLY RECOGNITION.

In the present embodiment, adopt the Minitab statistical software that described textural characteristics value is carried out statistical analysis, draw the textural characteristics population of parameters that there were significant differences, and create described textural characteristics population of parameters, be used for the EARLY RECOGNITION diagnosis of blood vessel wall pathological changes.

Further, the quantification of described blood vessel wall image texture characteristic and extracting method comprise and adopt grader that the blood-vessel image of healthy body and pathological changes body is classified.Namely from described textural characteristics population of parameters, choose best characteristic parameter according to clinical knowledge and experience, and the employing grader, such as KNN(k-Nearest Neighbor) grader classifies to blood vessel (such as the common carotid artery) ultrasonoscopy of healthy body and pathological changes body.Thus, when utilizing the quantification of described blood vessel wall image texture characteristic and extracting method to carry out the blood vessel wall lesion detection, classification results can be used in reference to the differentiation of emissary vein pathological changes, the foundation that provides for the early diagnosis and therapy of cardiovascular disease.

Be the quantification of verifying described blood vessel wall image texture characteristic provided by the invention and feasibility and the effectiveness of extracting method, below the quantification of the described blood vessel wall image texture characteristic of employing and extracting method are to VisualSonics vevo2100 (Visualsonics Inc., Toronto, Canada) the mice common carotid artery image (comprising normal mouse and apoe knock out mice) of high frequency ultrasound imaging system acquisition carried out quantification and the extraction of textural characteristics.Wherein, Vevo2100 is equipped with the MS-550D linear array probe, and mid frequency is 40MHz.At first use isoflurane gas (flow rate 1L/min) anesthetized mice of 1% pure oxygen, then remove the chaeta at imaging position with depilatory cream, be fixed on the examining table, make the body temperature of mice remain on 36 ~ 38 ° by hot plate, adopt the high frequency ultrasound imaging Vevo2100 of system to the common carotid artery imaging of anesthetized mice, as shown in Figure 4.14 normal mouses and 16 apoe knock out mice (high lipid food fed for 36 weeks) are carried out experimentation, and two kinds of Carotid ultrasonoscopys of mice as shown in Figure 5.

Adopt the quantification of blood vessel wall image texture characteristic of the present invention and the textural characteristics value of extracting method quantification and normal mouse and apoe knock out mice, then with Minitab statistics software it is analyzed and draw 14 textural characteristics populations of parameters that there were significant differences, as shown in table 1, thus normal mice and clpp gene deratization effectively distinguished.

Table 1. normal mouse and apoe knock out mice different texture eigenvalue statistic analysis result

The textural characteristics value	Normal mouse (n=14)	Apoe knock out mice (n=16)	pvalue
				Gray average *	201.49±8.84	141.4±23.4	0.0001
Standard deviation	23.45±2.62	23.76±4.69	0.825
				Fractal dimension *	2.046±0.043	2.014±0.037	0.041
Contrast *	9.14±3.19	6.36±2.38	0.013
				Dependency	0.889±0.033	0.909±0.039	0.129
Energy *	0.034±0.007	0.022±0.005	0.0001
				The unfavourable balance distance	0.518±0.03	0.517±0.038	0.94
Entropy *	1.317±0.048	1.408±0.057	0.0001
				Contrast-1*	17.39±7.05	26.3±11.4	0.016
Angle direction second moment *	0.184±0.026	0.135±0.025	0.0001
				Entropy-1*	0.838±0.075	0.952±0.09	0.001
Meansigma methods-1*	0.0128±0.002	0.0162±0.003	0.001
				Roughness *	6.1±1.63	11.53±3.49	0.0001
Contrast-2	0.55±0.186	0.443±0.151	0.098
				Frequency *	2.73±1.42E-5	1.55±0.97E-5	0.016
Complexity *	4600±1837	7076±3446	0.02
				Texture strength *	9397±4332	31187±17728	0.0001
Contrast-3	1.27±0.225	1.22±0.107	0.436
				Covariance	0.797±0.123	0.87±0.083	0.073
Diversity *	1.266±0.159	0.892±0.22	0.01

Choose wherein 14 effective textural characteristics values (the textural characteristics value of " * " sign in the table) and make up population of parameters as feature set, and adopt the KNN grader that the Carotid ultrasonoscopy of healthy mice and pathological changes mice is classified.Ultrasonoscopy in conjunction with 30 known classification results is estimated 29 samples as training, and nicety of grading can reach 93.1%, as shown in Figure 6.

From experimental result, the quantification of described blood vessel wall image texture characteristic and extracting method can be used for instructing the differentiation of vascular lesion based on image texture characteristic, and treating for the early diagnosis of cardiovascular disease provides foundation.

Compared to prior art, quantification and the extracting method of described blood vessel wall image texture characteristic provided by the invention have the following advantages: one, the suitability are stronger, the image that produces applicable to fast imaging technology such as various medical image systems and optics, optoacoustics.Two, can extract the textural characteristics value of any area-of-interest blood vessel wall, when helping the blood vessel wall lesion detection pathological changes accurately be located.Three, the blood-vessel image that directly gathers based on various medical image systems of the quantification of described blood vessel wall image texture characteristic and extracting method, clinician and the operator with certain medical imaging devices use experience can collect required image, are difficult for being affected by other factors.Four, the quantification of described blood vessel wall image texture characteristic and extracting method can be used as a post processing of image software module and are integrated in the existing medical image system, to promote the function of medical image system.Therefore need not existing clinical imaging system is carried out HardwareUpgring, upgrade cost is low, is accepted by hospital and doctor easily, makes things convenient for clinical expansion.

The above, only be embodiments of the invention, be not that the present invention is done any pro forma restriction, although the present invention discloses as above with embodiment, yet be not to limit the present invention, any those skilled in the art, within not breaking away from the technical solution of the present invention scope, when the technology contents that can utilize above-mentioned announcement is made a little change or is modified to the equivalent embodiment of equivalent variations, in every case be not break away from the technical solution of the present invention content, any simple modification that foundation technical spirit of the present invention is done above embodiment, equivalent variations and modification all still belong in the scope of technical solution of the present invention.

Claims (10)

1. the quantification of a blood vessel wall image texture characteristic and extracting method is characterized in that, quantification and the extracting method of described blood vessel wall image texture characteristic may further comprise the steps:

Utilize medical image system to gather the continuous blood vessel wall image of multiframe;

Choose blood vessel wall inner membrance-middle rete as area-of-interest at the described blood vessel wall image of each frame, and extract the textural characteristics value of described area-of-interest, then obtain its meansigma methods;

Described textural characteristics value is carried out statistical analysis, make up the textural characteristics population of parameters that is used for the diagnosis of blood vessel wall pathological changes EARLY RECOGNITION.

2. the quantification of blood vessel wall image texture characteristic as claimed in claim 1 and extracting method is characterized in that, described medical image system is ultrasonic image-forming system, optical imaging system, CT imaging system or MRI imaging system.

3. the quantification of blood vessel wall image texture characteristic as claimed in claim 1 and extracting method, it is characterized in that described textural characteristics value comprises first-order statistical properties, fractal dimension texture analysis, gray level co-occurrence matrixes, grey scale difference statistics, local gray level difference matrix and statistical nature matrix.

4. the quantification of blood vessel wall image texture characteristic as claimed in claim 3 and extracting method is characterized in that, described first-order statistical properties comprises gray average and the standard deviation of vascular wall area.

5. the quantification of blood vessel wall image texture characteristic as claimed in claim 3 and extracting method is characterized in that, described fractal dimension texture analysis parameter is fractal dimension.

6. the quantification of blood vessel wall image texture characteristic as claimed in claim 3 and extracting method is characterized in that, described gray level co-occurrence matrixes feature comprises contrast, dependency, energy, unfavourable balance distance, entropy.

7. the quantification of blood vessel wall image texture characteristic as claimed in claim 3 and extracting method is characterized in that, described grey scale difference statistics comprises contrast, angle direction second moment, entropy, meansigma methods.

8. the quantification of blood vessel wall image texture characteristic as claimed in claim 3 and extracting method is characterized in that, described local gray level difference matrix feature comprises roughness, contrast, frequency, complexity, texture strength.

9. the quantification of blood vessel wall image texture characteristic as claimed in claim 3 and extracting method is characterized in that, described statistical nature matrix character comprises contrast, covariance, diversity.

10. the quantification of blood vessel wall image texture characteristic as claimed in claim 1 and extracting method is characterized in that, the quantification of described blood vessel wall image texture characteristic and extracting method further comprise and adopt grader that healthy and lesion vessels image are classified.

Images