CN103720489A - Lesion tissue growth monitoring method and system - Google Patents

Abstract

The invention provides a lesion tissue growth monitoring method and system. The method includes the steps that inductive shear waves for deformation of a lesion tissue area are acquired, and an elastic image of lesion tissue is acquired according to the inductive shear waves; point-by-point matching is performed for an ROI area of the elastic image, pixel points in the area are tracked, and an elastic information image of the ROI area is constructed; the hardness attribute of the lesion tissue in the elastic information image is determined; the infiltration degree of the elastic information image is tracked, and the range of the infiltration tissue area is determined; the infiltration growth characteristic parameters of the lesion tissue are acquired according to the elastic information image, the hardness attribute and the infiltration tissue area range; the growth of the lesion tissue is monitored according to the infiltration growth characteristic parameters of the lesion tissue. Through the technical scheme, elastic information of the lesion tissue can be represented completely, the benign/malignance mass classification precision is high, the tumor growth monitoring accuracy is high, and an important basis is provided for pathological research on tumors and early diagnosis and treatment of tumors.

Description

Pathological tissues growth monitoring method and system

Technical field

The present invention relates to technical field of medical image processing, particularly relate to a kind of pathological tissues growth monitoring method and system.

Background technology

Medicine clinical research shows, during cell generation pathological changes, the biomechanics characteristic generation significant change of himself and normal surrounding tissue, utilize the image information of display organization elastic parameter to carry out quantitative sign, assessment and monitoring to the growth characteristics of breast tumor, preoperative prediction, the defining with benign tumors, pernicious differentiation etc. of target volume of protecting breast conserving surgery are all significant.

For example, in the breast tumor growth characteristics monitoring based on elastic parameter, current research work mainly concentrates on and how to build on elastic image, the perspective study of shortage to elastic image information and clinical diagnosis result, diagnostic result is often subject to the subjective impact of diagnosis person, the elastic parameter of tumor growth is extracted and is not formed unified standard, and these have all affected the assessment of breast tumor growth characteristics and monitoring accuracy.

Tumor is a lot of tissue disease the most basic the most common sign, normal according to some features of tumor clinically, as size, hardness and surface flatness thereof etc. are assessed neoplastic lesion situation, realize the growth monitoring of tumor, stiffness characteristics is as the common feature of breast tumor, owing to lacking advanced instrument, all can only measure in the mode of palpation all the time.

In recent years, elastogram technology has obtained development widely, the information such as its skew by display organization, strain are described tissue elasticity degree, make the extraction of tissue elasticity growth characteristics become possibility, are also at present clinically for measuring the main mode of tissue elasticity growth information.

Research for tumor-infiltrated growth characteristics parameter on two-dimension elastic image is scarcely out of swaddling-clothes, prior art mainly contains following several: 1. < < Biomedical Supersonics > > (ten thousand bright habit chief editors) volume two 480-483 page, Chinese Patent Application No. 201030119332.4 < < liver cirrhosis detecting instrument > >, the medical journal > > of < < PLA the 36th volume o. 11th 1131-1133 page in 2011 and < < Chinese medicine image technology > > the 28th volume the 3rd phase 529-533 page in 2012, in ultrasonic Transient elastography image, elasticity average with area-of-interest characterizes elastic characteristic, this characteristic parameter is mainly used in diagnosing liver fibrosis, the liver dispersivity pathological changes such as liver cirrhosis, 2. the clinical value research > > (author: Mei Chenling) of Guangxi Medical University's master thesis < < real-time ultrasound elastogram Diagnosis of Focal Liver Lesions in 2011, < < Chinese medicine image technology > > the 26th volume the 9th phase 1682-1684 page in 2010, < < modern biomedical progress > > the 26th volume the 9th phase 492-494 page in 2010 and < < West China medical science > > the 25th volume the 2nd phase 294-297 page in 2010, according to the COLOR COMPOSITION THROUGH DISTRIBUTION feature of focus elastic image, mark, thereby instruct the attribute of pathological changes to differentiate, five point-scores of commonly using clinically, 3. < < Cancer Control > > 17 volumes the 3rd phase 156-161 page in 2010, < < clinical medicine > > the 31st 10 phase of volume 93-94 page in 2011, < < contemporary medical science > > the 18th volume the 1st phase 7-9 page in 2012, the research > > (author: Ji Jianfeng) of The 2nd Army Medical College master thesis < < ultrasonic elastograph imaging in 2011 to Hepatic Tumors differential diagnosis value, < < Chinese medicine image technology > > the 18th volume the 7th phase 589-591 page in 2009 and < < world Chinese digest magazine > > the 18th volume the 30th phase 3254-3258 page in 2010, by tumor region strain rate ratio on calculating elastic image, infer the hardness of tumor with respect to normal surrounding tissue, 4. the application > > (author: palace rosy clouds) of < < China ultrasound medicine magazine > > the 25th volume the 4th phase 362-364 page in 2009, Fudan University master thesis < < Real time Organization elastogram in 2010 in distinguishing between benign and malignant breast lump, by tumor region area ratio on elastic image, infer the hardness of tumor with respect to normal surrounding tissue, 5. < < J Ultrasound Med > > 2012 the 31st phase 281-287 page proposes on elastic image the lenth ratio of tumor region on tumor region and B pattern, the elastic characteristic of assess lesion region and normal tissue regions.

Technical scheme in sum, the elastic characteristic that 1. method is extracted is quantitative; 2. the elastic characteristic that 3. 4. 5. method is extracted is qualitative or sxemiquantitative.Concrete, 1. for adopting the instantaneous elasticity detection technique of one-dimensional image technology, although can quantitative organize average elasticity modulus value, cannot obtain two-dimension elastic distributed intelligence, be generally only applicable to detect dispersivity pathological changes.2. being that elastic modelling quantity is estimated to a kind of method substantially the most the earliest, is also current clinical diagnostic applications elastic characteristic the most widely, but also has following shortcoming: 1) due to tumor multiformity, standards of grading are improved and are not enough to be applicable to all tumor situations not; 2) subjective assessment that scoring process is made elastic image by sonographer, causes scoring process to be subject to the impact of doctor's subjective factors.3. be 5. 4. the feature of recent scholars for assessment of tumor elastic performance, compare the firmness change degree that 2. more reflects focus objective quantitative, but still exist and treat improvements: 1) feature cannot be carried out the lateral comparison between different focuses, the above-mentioned ratio interval obtaining under some pathological changes exists certain overlapping, only relies on single ratio identification easily to cause mistaken diagnosis; 2) in focus region soft or hard region define use fixed threshold, due to elastogram shortcoming, in different images, define the actual corresponding adaptability to changes difference of threshold value, cause elasticity number quantization error; 3) when benign from malignant tumors is classified, adopt single threshold value to image classification, the suitability is not strong, causes mistaken diagnosis, fails to pinpoint a disease in diagnosis.

In sum, apply the elastic characteristic value that above-mentioned the whole bag of tricks obtains, the elastic information of sign tumor that cannot be complete, exists that good Malignant mass nicety of grading is low, the low problem of tumor growth monitoring accuracy.

Summary of the invention

Based on this, be necessary also to exist the elastic information of sign tumor that cannot be complete, the low problem of good Malignant mass nicety of grading for above-mentioned elastic characteristic value, a kind of pathological tissues growth monitoring method and system is provided.

A kind of pathological tissues growth monitoring method, comprises the steps:

Obtain the induction shearing wave of diseased tissue area deformation, according to described induction shearing wave, obtain the elastic image of pathological tissues;

For the ROI region of described elastic image, carry out pixel in pointwise coupling tracing area, build the elastic information image in ROI region;

Determine the soft or hard attribute of pathological tissues in described elastic information image;

Described elastic information image is carried out to infiltration degree tracking, determine invade tissues regional extent;

According to described elastic information image, soft or hard attribute and invade tissues regional extent, obtain the infiltration growth characteristics parameter of pathological tissues;

According to the infiltration growth characteristics parameter of described pathological tissues, the growth of described pathological tissues is monitored.

A kind of pathological tissues growth monitoring system, comprising:

Elastic image acquisition module, for obtaining the induction shearing wave of diseased tissue area deformation, obtains the elastic image of pathological tissues according to described induction shearing wave;

Elastic information image creation module, carries out pixel in pointwise coupling tracing area for the ROI region for described elastic image, builds the elastic information image in ROI region;

Organize soft or hard attribute determination module, for determining the soft or hard attribute of described elastic information image pathological tissues;

Infiltration degree tracking module, for described elastic information image is carried out to infiltration degree tracking, determines invade tissues regional extent;

Elastic characteristic parameter extraction module, for obtaining the infiltration growth characteristics parameter of pathological tissues according to described elastic information image, soft or hard attribute and invade tissues regional extent;

Pathological tissues growth monitoring module, monitors the growth of described pathological tissues for the infiltration growth characteristics parameter according to described pathological tissues.

Above-mentioned pathological tissues growth monitoring method and system, merge acoustic radiation force elastogram technology and image processing techniques and obtain two-dimension elastic modulus figure, utilize image processing techniques to carry out pretreatment to image and obtain tumor-infiltrated degree profile, realize auto Segmentation and the soft or hard region deviding of wetted area on two-dimension elastic image, then combining image is processed knowledge and clinical experience data, according to the growth characteristics of wetted area, from image, extract identification ability high, the elasticity number argument sequence of strong robustness, adopt the Infiltrating of mechanics imaging means assess lesion tissue, the elastic information of complete sign pathological tissues, can instruct the good of pathological tissues, pernicious differentiation, good/Malignant mass nicety of grading is high, pathological tissues growth monitoring accuracy is high, for pathological research and the early diagnosis treatment of tumor provide important evidence, for early diagnosis and the pathological study treatment of tumor provide new effective way.

Accompanying drawing explanation

Fig. 1 is the pathological tissues growth monitoring method flow diagram of an embodiment;

Fig. 2 is elastic information image reconstruction result schematic diagram;

Fig. 3 is elastic image soft or hard region deviding result schematic diagram;

Fig. 4 is tissue infiltration's degree border tracking results schematic diagram;

Fig. 5 is that schematic diagram is selected in invade tissues neighboring area;

Fig. 6 is the pathological tissues growth monitoring system structure schematic diagram of an embodiment;

Fig. 7 is the zone boundary tracking results schematic diagram of fibroadenoma (optimum);

Fig. 8 is IDC (pernicious) zone boundary tracking results schematic diagram.

The specific embodiment

Below in conjunction with accompanying drawing, the specific embodiment of pathological tissues growth monitoring method and system of the present invention is described in detail.

Pathological tissues growth monitoring method and system of the present invention, can be applied to the various pathological tissues growths of monitoring, for example, the extent of elasticity assessments such as such as liver, pulmonary, thyroid, breast tumor infiltrates the assessment of growth characteristics extent of elasticity, also can be for lesion region extent of elasticity assessment on mammary gland and other position of human body.

Shown in Figure 1, Fig. 1 is the pathological tissues growth monitoring method flow diagram of an embodiment, mainly comprises the steps:

Step S10, obtains the induction shearing wave of diseased tissue area deformation, obtains the elastic image of pathological tissues according to described induction shearing wave.

In this step, can, by utilizing the ultrasonic acoustic radiation field of force to induce the propagation of the shearing wave producing in mechanics of biological tissue, obtain the induction shearing wave of diseased tissue area deformation, and then according to induction shearing wave, obtain the elastic image of pathological tissues.

Consider that the shearing wave that ultrasonic acoustic radiation field of force induction produces organizes at complex biological the propagation that (comprises fat, connective tissue etc. in as mammary gland) in different structure, can introduce the impact of the imperfectization factors such as heterogeneity, velocity of sound decay, border, simultaneously because electronic noise and quantizing noise, the tissue compression of ultrasonic system cause the reasons such as echo-signal axial deformation, can produce the effect of signal Correlaton, make Displacement Estimation bring larger error.Therefore, need to provide one more accurate elastograph imaging method, to obtain high-quality breast tumor two-dimension elastic modulus scattergram.

In one embodiment, the elastic image process of obtaining pathological tissues of step S10 specifically comprises the steps:

Step S101, the radiant force that utilizes focused ultrasound beams to produce obtains breast tumor regional deformation induction shearing wave.

Particularly, adopt the Ultrasonic Elasticity Imaging based on acoustic radiation force, the radiant force that utilizes ultrasonic system to produce obtains the deformation induced shearing wave of diseased tissue area.

Step S102, follows the trail of the propagation of shearing crest face, and calculation side, to the relation curve of each point position on propagation path and shearing wave crest time of advent, calculates shear wave velocity according to described relation curve.

Particularly, first the RF signal that obtains shearing wave is carried out to Hilbert transform, becomes analytic signal:

$\hat{r}(n,s)=r(m,s)*h\left(m\right)=\frac{1}{\mathrm{\&pi;}}{\&Integral;}_{-\&infin;}^{\&infin;}\frac{r(\mathrm{\&tau;},s)}{m-\mathrm{\&tau;}}\mathrm{d\&tau;}---\left(1\right)$

In formula (1), m is the sample point coordinate of RF signal, and s represents the time coordinate of sampled point; * represent convolution algorithm, h (m) is the FIR Hilbert transform factor.

Choose the long L of suitable observation window, above-mentioned analytic signal carried out to computing cross-correlation:

$\hat{k}(n,s)={\mathrm{\&Sigma;}}_{1=-\frac{L}{2}}^{\frac{L}{2}-1}\hat{r}(m+1,s){\hat{r}}^{*}(m+1+L,s+1)---\left(2\right)$

In formula (2), k^ (n, s) is cross-correlation coefficient, and subscript * represents conjugate complex number.The ultrasound echo signal time shift Δ t causing when micro-displacement is less than sampling period t _stime, the phase information that can comprise by cross-correlation coefficient is tried to achieve time shift:

In formula (3), n _maxrepresent the peaked coordinate of k^ (n, s), ∠ is angle computing, and ∠ (a+jb)=arctan (b/a) obtains after time shift value, and micro-displacement can be tried to achieve by ultrasonic velocity c:

$d(m,s)=\frac{1}{2}\mathrm{c\&Delta;t}---\left(4\right)$

In formula (4), d (m, s) represents micro-displacement.

Step S103, obtains Young's modulus of elasticity according to described shear wave velocity, according to described Young's modulus of elasticity, obtains the high-resolution elastic image of pathological tissues.

Particularly, by tracking, shear the propagation of crest face (being displacement maximum), calculate the relation curve of the time of advent of each point position and shearing wave crest on lateral propagation path, utilize Radon transform to find optimum linearity matching scheme, calculate shear wave velocity, substitution formula (5) is calculated Young's modulus of elasticity:

$E=2(1+v)\mathrm{\&rho;}{V}_s^2---\left(5\right)$

When meeting real time implementation requirement, make jitter effect minimize and approach the lower limit of cram é r-rao, obtain the elastic image of the quality, high resolution of pathological tissues.

In the present embodiment, adopt the Ultrasonic Elasticity Imaging based on acoustic radiation force, utilized ultrasonic system to obtain the deformation induced shearing wave of diseased tissue area.Based on the non-homogeneous biological tissue elasticity formation method of acoustic radiation force, obtain the two-dimension elastic modulus figure of more accurate mammary gland tissue; With respect to traditional cross correlation algorithm based on time-domain signal coupling, be that the RF signal data collecting is divided into some segments according to depth direction, by every a bit of and adjacent RF signal data, make computing cross-correlation, and then obtain the displacement (peak value by cross-correlation function is definite) of tissue with the technological means of the variation of the degree of depth and time.What the method can be adapted to that acoustic radiation force causes organizes side-play amount very little, generally only has tens microns even to only have the situation of several microns.By conventional cross correlation algorithm is improved, thereby micro-displacement can be more accurately solved, the elastic image of quality, high resolution can be obtained.

Step S20, carries out pixel in pointwise coupling tracing area for the ROI region (Region Of Interest, area-of-interest) of described elastic image, builds the elastic information image in ROI region.

In this step, due to the particularity of elasticity ultrasonoscopy, before tissue regions extent of elasticity being assessed according to elasticity ultrasonoscopy, need to be to image in the pretreatment operation that guarantees rebuild elastic information image in the situation that of not destroying amount of image information.

Elastic image is to utilize tissue ultrasonic imaging technique to tissue, to apply a small strain by probe apparatus, collect measured body echo-signal fragment and it is analyzed, property estimate that the deformation extent of tissue is tissue elasticity coefficient magnitude, then with GTG or coloud coding imaging.

In elastic image, the colouring information of characterizing tissues coefficient of elasticity size is superimposed upon on its B mode image and shows, for assessment accurately, organizes extent of elasticity, first will reconstruct elastic information image.The present invention utilizes the mode of pointwise coupling to follow the tracks of ROI area pixel point on elastic image, reconstructs tissue elasticity frame, and detailed process is as follows:

If elasticity vision-mix ROI region is I _c, B mode image ROI region is I _b, after rebuilding, image ROI region is I _r, according to elastogram principle, elastic information image can be set up by following formula:

I _r(i,j)=I _c(i,j)-I _b(i,j) (6)

Wherein, (i, j) is pixel coordinate in ROI region, by rebuilding elastic information image, quantize to extract and analyze tissue elasticity characteristic sequence on the image after rebuilding, the interference of removing B mode image, makes the extent of elasticity assessment of follow-up invade tissues more reliable.Concrete effect can be with reference to shown in figure 2, Fig. 2 is elastic information image reconstruction result schematic diagram, in figure take mammary gland tissue as example, provide mammary gland elastic image, B mode image and rebuild rear elastic information image, figure a is original elastic image, figure b is the elastic information image of rebuilding, and figure c is B mode image.

Step S30, determines the soft or hard attribute of pathological tissues in described elastic information image.

In this step, the soft or hard attribute of realizing in pathological tissues ROI region by elastic information image defines.

Traditional pathological tissues infiltrates growth characteristics elastic parameter sequential extraction procedures technology, soft or hard region Definition of Division is generally to realize by a fixed threshold, on elastic image, in different patient's images, the actual corresponding adaptability to changes of same color is not necessarily identical, but the suitability of fixed threshold cutting techniques is not strong, easily causes elasticity number quantization error.In fact on tissue elasticity frame, organization internal change color does not have obvious boundary, and the inner soft or hard of pathological tissues region is the fuzzy set of boundary fuzzy, and traditional extractive technique, obviously can not bring high-precision soft or hard attribute to define.

For this reason, the present invention also provides a kind of soft or hard attribute confining method of more accurate pathological tissues.

In one embodiment, step S30 determines that the process of the soft or hard attribute of pathological tissues in described elastic information image can comprise as follows:

First calculate the cluster centre of described elastic information image; Wherein, cluster centre comprises territory, hard area cluster centre and soft region clustering center.Then calculate on described elastic information image the fuzzy similarity matrix of each element and the square distance of cluster centre and, obtain the degree of membership that each element belongs to soft or hard interval.According to degree of membership, determine again the soft or hard attribute of ROI regional organization.Particularly, above-mentioned processing procedure is resolved as follows:

For elastic information image, establishing its gray level image is I, u _ijin presentation video, j pixel belongs to the degree of membership of i class, the sum of all pixels that n is image I, and cluster object function is as follows:

$E(U,V)=\underset{i=1}{\overset{2}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{\left(u_{\mathrm{ij}}\right)}^m{\left(d_{\mathrm{ij}}\right)}^2,(u_{\mathrm{ij}}\&Element;[\mathrm{0,1}],\underset{i=1}{\overset{2}{\mathrm{\&Sigma;}}}{u}_{\mathrm{ij}}=1)---\left(7\right)$

Wherein, U is initial degree of membership matrix, and m is weight index, and m ∈ [1 ,+∞), d _ijfor each pixel is to center vector distance, V is cluster centre, V=(v ₁, v ₂) ^t, the number of classifying is 2, i=1,2.

After degree of membership and cluster centre are determined, utilize Lagrangian multiplication, order:

$\left\{\begin{array}{c}F=\underset{i=1}{\overset{2}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{\left(u_{\mathrm{ij}}\right)}^m{\left(d_{\mathrm{ij}}\right)}^2+\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}(\underset{i=1}{\overset{2}{\mathrm{\&Sigma;}}}{u}_{\mathrm{ij}}-1)\\ \frac{\&PartialD;F}{{\&PartialD;x}_i^{\&prime;}}=0,\frac{\&PartialD;F}{{\&PartialD;u}_{\mathrm{ij}}}=0\end{array}\right.---\left(8\right)$

Solving equation group (8) can obtain:

$u_{\mathrm{ij}}=\frac{1}{\underset{l=1}{\overset{2}{\mathrm{\&Sigma;}}}{\left(\frac{d_{\mathrm{ij}}}{d_{\mathrm{lj}}}\right)}^{\frac{2}{(m-1)}}},(i=\mathrm{1,2};j=\mathrm{1,2},...,n)---\left(9\right)$

$v_i=\frac{\underset{n}{\overset{j=1}{\mathrm{\&Sigma;}}}{\left(u_{\mathrm{ij}}\right)}^m{x}_j}{\underset{j=1}{\overset{2}{\mathrm{\&Sigma;}}}{\left(u_{\mathrm{ij}}\right)}^m},(i=\mathrm{1,2})---\left(10\right)$

Degree of membership matrix U and cluster centre V obtain as follows:

1. meeting under the prerequisite of degree of membership constraints, between 0-1, initialize degree of membership matrix U.

2. application (10) formula solves 2 cluster centres (territory, hard area cluster centre, soft region clustering center).

3. according to (7), calculate E (U, V).

Repeat said process, until the variation of cluster centre V is less than certain threshold value or stops when substantially constant calculating, obtain best fuzzy classification matrix and cluster centre, thereby realize ROI regional organization soft or hard attribute, define.

The confining method of the soft or hard attribute of above-mentioned pathological tissues, by ask on elastic information image space the fuzzy similarity matrix of each element and the square distance of cluster centre and, obtain the interval degree of membership of element soft or hard, finally realizing ROI region inner tissue soft or hard attribute defines, this confining method adopts Fuzzy clustering techniques automatically to determine and organize soft or hard attribute, compared with traditional method, the accuracy that soft or hard attribute defines is higher, and the suitability is stronger.It defines result can be with reference to shown in figure 3, and Fig. 3 is elastic image soft or hard region deviding result schematic diagram, and in figure, still take mammary gland tissue as example, wherein black line is soft or hard regional edge boundary line.

Need statement, above-mentioned for preferably defining mode, for organizing soft or hard attribute to define, also can apply other two sorting algorithm, do not repeat them here.

Step S40, carries out infiltration degree tracking to described elastic information image, determines invade tissues regional extent.

In this step, according to the bounds of the pathological tissues of elastic information image, obtain and infiltrate growth characteristics argument sequence, the extent of elasticity of analysis and evaluation pathological tissues, and for instructing its growth monitoring, therefore, before extent of elasticity assessment, elastic information image is carried out to infiltration degree tracking, thereby determine the bounds of pathological tissues.

In one embodiment, consider complexity and the scrambling of elastic information image, first the tracking of tissue infiltration's degree is completed under B mode image, then be mapped under elastic information image, realizing tissue infiltration's degree image boundary on elastic image follows the tracks of, concrete, the process of step S40 can comprise as follows:

First, obtain the B mode image of elastic information image, and on B mode image, the zone boundary of pathological tissues is followed the tracks of and obtained tissue infiltration's degree border; Then extract the marginal information of B mode image and be fused on elastic information image, obtaining the invade tissues regional extent on elastic information image.

The infiltration degree tracking scheme of above-described embodiment, has utilized B pattern undertissue to infiltrate the complicated irregular feature of edge shape, adopts the tissue infiltration's degree border tracking based on Chan-Vese model, and concrete grammar is described below:

If the elastic image I of B pattern ₀coordinate set be Ω, ω is the subgraph being defined in Ω, the border C of ω is inside (C) and two regions of outside (C) by image division.

If c ₁and c ₂represent respectively the average gray in these two regions, level set function represent inside (C) and outside (C), when following energy function hour, level set function zero level collection be the object boundary curve C of expectation:

In above formula, μ, v, λ ₁, λ ₂all constant coefficient, be Heaviside function, for each step, develop c ₁and c ₂can be calculated by two formulas below:

$\left\{\begin{array}{c}{c}_1\left(\mathrm{\&phi;}\right)=\frac{{\&Integral;}_{\mathrm{\&Omega;}}{u}_0(x,y)H\left(\mathrm{\&phi;}\right)\mathrm{dxdy}}{{\&Integral;}_{\mathrm{\&Omega;}}H\left(\mathrm{\&phi;}\right)\mathrm{dxdy}}\\ c_2\left(\mathrm{\&phi;}\right)=\frac{{\&Integral;}_{\mathrm{\&Omega;}}{u}_0(x,y)(1-H\left(\mathrm{\&phi;}\right))\mathrm{dxdy}}{{\&Integral;}_{\mathrm{\&Omega;}}(1-H\left(\mathrm{\&phi;}\right))\mathrm{dxdy}}\end{array}\right.---\left(12\right)$

Set function in time t develop, by formula (12) can derive about euler-Lagrange equation:

$\left\{\begin{array}{c}\frac{\&PartialD;\mathrm{\&phi;}}{\&PartialD;t}={\mathrm{\&delta;}}_{\mathrm{\&epsiv;}}\left(\mathrm{\&phi;}\right)[\mathrm{\&mu;}\mathrm{div}\left(\frac{\&dtri;\mathrm{\&phi;}}{|\&dtri;\mathrm{\&phi;}|}\right)-v-{\mathrm{\&lambda;}}_1{(u_0-c_1)}^2+{\mathrm{\&lambda;}}_2{(u_0-c_2)}^2]=0,\mathrm{in}(0,\&infin;)\&times;\mathrm{\&Omega;}\\ \mathrm{\&phi;}(0,x,y)={\mathrm{\&phi;}}_0(x,y),\mathrm{in\&Omega;}\\ \frac{{\mathrm{\&delta;}}_{\mathrm{\&epsiv;}}\left(\mathrm{\&phi;}\right)}{|\&dtri;\mathrm{\&phi;}|}\frac{\&PartialD;\mathrm{\&phi;}}{\&PartialD;\stackrel{\&OverBar;}{n}}=0,\mathrm{on}\&PartialD;\mathrm{\&Omega;}\end{array}\right.---\left(13\right)$

In above formula, the approximate of Dirac function, represent the outer normal vector on border, δ ₀(x, y) is the symbolic distance function being obtained by initial boundary.

Solving equation (13) obtains object boundary curve C, thereby complete tissue infiltration's degree border follow the tracks of, above-mentioned tracking mode, in conjunction with auto Segmentation technology from motion tracking invade tissues border, avoid the subjectivity impact of manually cutting apart, improved and cut apart accuracy; Border tracking results can be with reference to shown in figure 4, Fig. 4 is tissue infiltration's degree border tracking results schematic diagram, in figure still take mammary gland tissue as example, figure a is tissue infiltration's degree border tracking results on B mode image, figure b is tissue infiltration's degree border tracking results on the elastic information image after rebuilding, and figure c is elastic information gray level image undertissue's infiltration degree border tracking results.

Need statement, above-mentioned is preferred infiltration degree tracking mode, for tissue infiltration's degree, follows the tracks of and also can apply other conventional image partition method, does not repeat them here.

Step S50, obtains the infiltration growth characteristics parameter of pathological tissues according to described elastic information image, soft or hard attribute and invade tissues regional extent.

In this step, elastic information and the carve information in the Nei Ge region, ROI region of Main Analysis pathological tissues, obtain one group of tumor-infiltrated growth characteristics argument sequence, reflects objectively the extent of elasticity situation of invade tissues and surrounding tissue thereof.

In one embodiment, infiltrating growth characteristics parameter comprises: wetted area hardness ratio, elastic characteristic ratio, elasticity average, elasticity variance and elasticity mark.

1) for diseased tissue area hardness ratio; Invade tissues region hardness Ratio Features Inva _hardbe expressed as:

${\mathrm{Inva}}_{\mathrm{hard}}=\frac{N_{\mathrm{hard}}}{N_{\mathrm{inva}}}---\left(14\right)$

In formula, N _hardthat in wellability tissue regions, attribute is territory, hard area pixel number; N _invafor pixel number in wellability tissue regions.Inva _hardcharacterized the relative resilient degree of certain limit inner tissue, it shows more greatly tissue elasticity less (organizing harder), on the contrary elasticity larger (organizing softer).

2) for diseased tissue area elasticity average; Invade tissues region elastic modelling quantity meansigma methods Inva _avgbe expressed as:

${\mathrm{Inva}}_{\mathrm{avg}}=\frac{1}{N_{\mathrm{inva}}}\underset{i=1}{\overset{N_{\mathrm{inva}}}{\mathrm{\&Sigma;}}}{I}_i---\left(15\right)$

In formula, N _invait is pixel number in invade tissues region; I _igray value corresponding to pixel on elastic image.Inva _avgfor characterizing the average elasticity degree of certain limit inner tissue.

3) for diseased tissue area elasticity variance; The elastic modelling quantity variance Inva in invade tissues region _stdbe expressed as:

${\mathrm{Inva}}_{\mathrm{std}}=\sqrt{\frac{1}{N_{\mathrm{inva}}-1}\underset{i=1}{\overset{N_{\mathrm{inva}}}{\mathrm{\&Sigma;}}}{(I_i-{\mathrm{Inva}}_{\mathrm{avg}})}^2}---\left(16\right)$

In formula (16), N _invait is pixel number in invade tissues region; I _igray value corresponding to pixel on elastic image; Inva _avgit is average elasticity average in invade tissues region; Inva _stdfor characterizing the uniformity of certain limit inner tissue extent of elasticity.

4) for pathological tissues territory elasticity ratio; The elasticity ratio I nva in invade tissues region _eratiobe expressed as:

${\mathrm{Inva}}_{\mathrm{eratio}}=\frac{{\mathrm{Inva}}_{\mathrm{avg}}}{{\mathrm{ROI}}_{\mathrm{savg}}}---\left(17\right)$

In formula (17), Inva _avgit is the elasticity average in invade tissues region; ROI _savgit is the elasticity average of area-of-interest.Inva _eratiocharacterized the relative resilient intensity of variation of invade tissues and its periphery certain limit inner tissue.

5) for pathological tissues elasticity mark; Elasticity mark is main according to the current elasticity point system (5 point-score) extensively adopting clinically, analyzes in wetted area and the distribution of the hardness ratio of neighboring area, with modeling pattern, standards of grading is quantified as to discriminant parameter Esc, and calculation procedure is as follows:

1. ask for invade tissues region hardness ratio Inva _hard;

2. according to invade tissues region contour, establishing apart from region, this profile 30mm place is invade tissues neighboring area, and asking in this region is hardness ratio Peri in invade tissues neighboring area _hard; As shown in Figure 6, Fig. 5 is that schematic diagram is selected in invade tissues neighboring area to neighboring area profile, and wherein, the region between solid white line and Lycoperdon polymorphum Vitt solid line is invade tissues neighboring area, can be used for calculating elastic grading parameters.

3. elasticity Rating Model is as follows:

Table 1 elasticity Rating Model

Elasticity scoring has characterized the relative distribution situation of wetted area elasticity, and in elastic image, with change color display organization firmness change information, colouring information is from red to indigo plant, correspondingly be the DE of lump by soft to firmly.Binding of pathological is gained knowledge, and generally, optimum focus is labeled as indigo plant, green, and the inner elasticity distribution of focus is more even, less with the elastic contrast of surrounding tissue; Pernicious focus shows as red, yellow, and the elasticity distribution inhomogeneous (as blood vessel, calcification etc.) of focus inside is larger with the elastic contrast of surrounding tissue.Aforesaid way creates the profit growth characteristics argument sequence of pathological tissues, can avoid the limitation of single parameter, makes the analysis of growth characteristics more comprehensive.

Step S60, monitors the growth of described pathological tissues according to the infiltration growth characteristics parameter of described pathological tissues.

In this step, based on infiltrating growth characteristics parameter, the growth of pathological tissues is monitored, can be according to the infiltration growth characteristics parameter being formed by hardness ratio, elasticity average, elasticity variance, flexible ratio and elasticity mark of said extracted and quantification, this infiltrates growth characteristics argument sequence and reflects from different perspectives the extent of elasticity of having described invade tissues.And then can be strong, objective according to this identification ability, repeatable strong characteristic sequence judge the growth conditions of pathological tissues, for analyzing the elastic performance of invade tissues, thereby realizes the monitoring that pathological tissues is grown.

Shown in Figure 7, Fig. 7 is the pathological tissues growth monitoring system structure schematic diagram of an embodiment, comprising:

Elastic image acquisition module 10, for obtaining the induction shearing wave of diseased tissue area deformation, obtains the elastic image of pathological tissues according to described induction shearing wave;

Elastic information image creation module 20, carries out pixel in pointwise coupling tracing area for the ROI region for described elastic image, builds the elastic information image in ROI region;

Organize soft or hard attribute determination module 30, for determining the soft or hard attribute of described elastic information image pathological tissues;

Infiltration degree tracking module 40, for described elastic information image is carried out to infiltration degree tracking, determines invade tissues regional extent;

Elastic characteristic parameter extraction module 50, for obtaining the infiltration growth characteristics parameter of pathological tissues according to described elastic information image, soft or hard attribute and invade tissues regional extent;

Pathological tissues growth monitoring module 60, monitors the growth of described pathological tissues for the infiltration growth characteristics parameter according to described pathological tissues.

In one embodiment, described elastic image acquisition module 10 is further used for:

The radiant force that utilizes focused ultrasound beams to produce obtains breast tumor regional deformation induction shearing wave; Follow the trail of the propagation of shearing crest face, calculation side, to the relation curve of each point position on propagation path and shearing wave crest time of advent, calculates shear wave velocity according to described relation curve; According to described shear wave velocity, obtain Young's modulus of elasticity, according to described Young's modulus of elasticity, obtain the high-resolution elastic image of pathological tissues.

In one embodiment, the described soft or hard attribute determination module 30 of organizing is further used for:

Calculate the cluster centre of described elastic information image; Wherein, cluster centre comprises territory, hard area cluster centre and soft region clustering center; Calculate on described elastic information image the fuzzy similarity matrix of each element and the square distance of cluster centre and, obtain the degree of membership that each element belongs to soft or hard interval; According to degree of membership, determine the soft or hard attribute of ROI regional organization.

In one embodiment, described infiltration degree tracking module 40 is further used for:

Obtain the B mode image of elastic information image; On B mode image, the zone boundary of pathological tissues is followed the tracks of and obtained tissue infiltration's degree border; Extract the marginal information of B mode image and be fused on elastic information image, obtaining the invade tissues regional extent on elastic information image.

In one embodiment, described infiltration growth characteristics parameter comprises: wetted area hardness ratio, elastic characteristic ratio, elasticity average, elasticity variance and elasticity mark.

Pathological tissues growth monitoring system of the present invention is corresponding one by one with pathological tissues growth monitoring method of the present invention, technical characterictic and the beneficial effect thereof of at the embodiment of above-mentioned pathological tissues growth monitoring method, setting forth are all applicable in the embodiment of pathological tissues growth monitoring system, do not repeat them here, hereby statement.

Each embodiment in sum, the pathological tissues growth monitoring method and system that the present invention proposes:

(1) the two-dimension elastic image acquiring method of the non-homogeneous biological tissue based on acoustic radiation force: by analyzing the offset characteristic of biological tissue under acoustic radiation force effect, the formation mechanism of research ultrasound detection echo-signal, sets up the echo signal model of carrying displacement of tissue change information (shearing velocity of wave propagation).According to calculating shearing velocity of wave propagation, obtain elastic parameter, finally obtain the high-quality two-dimension elastic modulus figure of pathological tissues.

(2) extracting method of the pathological tissues lubricant nature growth characteristics argument sequence based on elastic image: by analyzing the elastic information of invade tissues on color elastic image, combining image is processed and empirical data, from image, extract that some identification abilitys are high, strong robustness, do not rely on elasticity number parameter, morphological characteristic and the textural characteristics of acquisition system and operator's maneuver, the extent of elasticity of comprehensive, scientifical, the invade tissues that describes in a systematic way.

(3) in conjunction with the tumor growth method for monitoring and analyzing of acoustic radiation force technology, image processing, mode identification technology: the present invention carries out accurate quantitative Analysis and analysis to the extent of elasticity feature of tissue biological's mechanical characteristics on elastic image, help clinician to analyze tumor-infiltrated growth characteristics, thereby realize the growth monitoring of pathological tissues tumor, can be used for assisting doctor's clinical diagnosis.

In order to verify technical scheme feasibility of the present invention and effectiveness, below in conjunction with an application example, set forth technical scheme of the present invention, the ultrasonic breast tumor image (comprising conventional B ultrasonic image and the elastic modelling quantity image of focus) that adopts certain hospital's Ultrasonography to gather by HITACHI Vision8500 Ultrasound Instrument has carried out quantification and the extraction of elastic characteristic sequence.Double width shows two-dimentional B ultrasonic image and elastic image simultaneously, keeps the stable of probe during operation as far as possible, maintains 2 seconds, the static B ultrasonic image of Cryopreservation and elastic image.

Utilize the technical scheme that the present invention proposes to monitor tumor growth, shown in figure 7, Fig. 8 and table 2, provide respectively fibroadenoma (optimum) and the tracking of IDC (pernicious) zone boundary and infiltrated growth characteristics argument sequence and extracted result schematic diagram, wherein, Fig. 7 a is the B ultrasonic image of mammary gland fibroadenoma, Fig. 7 b is lesion segmentation result schematic diagram, and Fig. 7 c is springform spirogram; Fig. 8 a is the B ultrasonic image of infiltration ductal carcinomas of breast, and Fig. 8 b is lesion segmentation result schematic diagram, and Fig. 8 c is elastic modelling quantity gray-scale map.

From above-mentioned accompanying drawing and table 2, can find out, fibroadenoma (optimum) hardness rate is less, and the inner elasticity distribution of focus is more even, less with the elastic contrast of surrounding tissue; IDC (pernicious) hardness rate is larger, and the elasticity distribution of focus inside is inhomogeneous, larger with the elastic contrast of surrounding tissue.From clinical experiment result, the tumor growth monitoring system based on ultrasonic elastograph imaging that the present invention proposes, can be used for tumor growth monitoring, instruct good, the pernicious differentiation of tumor, for the pathological research of invade tissues and the treatment of the early diagnosis of malignant tumor provide new foundation.

Table 2 elastic characteristic sequence:

Technical scheme of the present invention, can such as, for various human soma (liver, pulmonary, thyroid etc.) extent of elasticity, assess: the elastic characteristic sequence that invention proposes not only can infiltrate the assessment of growth characteristics extent of elasticity for breast tumor, also can be for lesion region extent of elasticity assessment on mammary gland and other position of human body.

In addition, not only can be applied to springform spirogram, can also be applied to two-dimension displacement figure and two-dimentional strain figure: the focus zone boundary tracing into is fused to two-dimension displacement figure and springform spirogram, also can extracts a series of tumor-infiltrated growth characteristics parameters, realize the monitoring of tumor growth characteristic.

Finally, technical scheme of the present invention, be adapted to especially the growth monitoring of breast tumor, also can be applied to the growth monitoring (such as liver, pulmonary, thyroid etc.) of other human body organ-tissue: also can extract a series of growth characteristics parameters of other position tissue regions, realize the monitoring of growth simultaneously.

Technical scheme of the present invention, merge acoustic radiation force elastogram technology and image processing techniques, utilize ultrasonic acoustic radiation force system, deeply dissect the offset characteristic of biological tissue under acoustic radiation force effect, by calculating, shear velocity of wave propagation and obtain elastic parameter, obtain the two-dimension elastic modulus figure of mammary gland tissue; For the two-dimension elastic modulus figure getting, utilize image processing techniques to carry out pretreatment to image, the tumor-infiltrated degree profile of automatic acquisition, realizes auto Segmentation and the soft or hard region deviding of wetted area on two-dimension elastic image; Then combining image is processed knowledge and doctor's clinical experience, analyzes the growth characteristics of wetted area, extracts that some identification abilitys are high, the elasticity number argument sequence of strong robustness from image, comprehensive, scientifical, systematically assess tumor-infiltrated growth characteristics.In addition, can also combine with clinical practice, by clinical sample, test, inquire into the feasibility and the effectiveness that adopt mechanics imaging means assessment (mammary gland) tumor-infiltrated situation, solve the actual difficult point in clinical practice, for early diagnosis and the pathological study treatment of (mammary gland) tumor provide new effective way.

The above embodiment has only expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but can not therefore be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.

Claims (10)

1. a pathological tissues growth monitoring method, is characterized in that, comprises the steps:

Obtain the induction shearing wave of diseased tissue area deformation, according to described induction shearing wave, obtain the elastic image of pathological tissues;

For the ROI region of described elastic image, carry out pixel in pointwise coupling tracing area, build the elastic information image in ROI region;

Determine the soft or hard attribute of pathological tissues in described elastic information image;

Described elastic information image is carried out to infiltration degree tracking, determine invade tissues regional extent;

According to described elastic information image, soft or hard attribute and invade tissues regional extent, obtain the infiltration growth characteristics parameter of pathological tissues;

According to the infiltration growth characteristics parameter of described pathological tissues, the growth of described pathological tissues is monitored.

2. pathological tissues growth monitoring method according to claim 1, it is characterized in that, the described radiant force that utilizes focused ultrasound beams to produce obtains the induction shearing wave of diseased tissue area deformation, and the step of obtaining the elastic image of pathological tissues according to described induction shearing wave comprises:

The radiant force that utilizes focused ultrasound beams to produce obtains breast tumor regional deformation induction shearing wave;

Follow the trail of the propagation of shearing crest face, calculation side, to the relation curve of each point position on propagation path and shearing wave crest time of advent, calculates shear wave velocity according to described relation curve;

According to described shear wave velocity, obtain Young's modulus of elasticity, according to described Young's modulus of elasticity, obtain the high-resolution elastic image of pathological tissues.

3. pathological tissues growth monitoring method according to claim 1, is characterized in that, the described step of determining the soft or hard attribute of pathological tissues in described elastic information image comprises:

Calculate the cluster centre of described elastic information image; Wherein, cluster centre comprises territory, hard area cluster centre and soft region clustering center;

Calculate on described elastic information image the fuzzy similarity matrix of each element and the square distance of cluster centre and, obtain the degree of membership that each element belongs to soft or hard interval;

According to degree of membership, determine the soft or hard attribute of ROI regional organization.

4. pathological tissues growth monitoring method according to claim 1, is characterized in that, described described elastic information image is carried out to infiltration degree tracking, determines that the step of invade tissues regional extent comprises:

Obtain the B mode image of elastic information image;

On B mode image, the zone boundary of pathological tissues is followed the tracks of and obtained tissue infiltration's degree border;

Extract the marginal information of B mode image and be fused on elastic information image, obtaining the invade tissues regional extent on elastic information image.

5. pathological tissues growth monitoring method according to claim 1, is characterized in that, described infiltration growth characteristics parameter comprises: wetted area hardness ratio, elastic characteristic ratio, elasticity average, elasticity variance and elasticity mark.

6. a pathological tissues growth monitoring system, is characterized in that, comprising:

Elastic image acquisition module, for obtaining the induction shearing wave of diseased tissue area deformation, obtains the elastic image of pathological tissues according to described induction shearing wave;

Elastic information image creation module, carries out pixel in pointwise coupling tracing area for the ROI region for described elastic image, builds the elastic information image in ROI region;

Organize soft or hard attribute determination module, for determining the soft or hard attribute of described elastic information image pathological tissues;

Infiltration degree tracking module, for described elastic information image is carried out to infiltration degree tracking, determines invade tissues regional extent;

Elastic characteristic parameter extraction module, for obtaining the infiltration growth characteristics parameter of pathological tissues according to described elastic information image, soft or hard attribute and invade tissues regional extent;

Pathological tissues growth monitoring module, monitors the growth of described pathological tissues for the infiltration growth characteristics parameter according to described pathological tissues.

7. pathological tissues growth monitoring system according to claim 6, is characterized in that, described elastic image acquisition module is further used for:

The radiant force that utilizes focused ultrasound beams to produce obtains breast tumor regional deformation induction shearing wave; Follow the trail of the propagation of shearing crest face, calculation side, to the relation curve of each point position on propagation path and shearing wave crest time of advent, calculates shear wave velocity according to described relation curve; According to described shear wave velocity, obtain Young's modulus of elasticity, according to described Young's modulus of elasticity, obtain the high-resolution elastic image of pathological tissues.

8. pathological tissues growth monitoring system according to claim 6, is characterized in that, the described soft or hard attribute determination module of organizing is further used for:

Calculate the cluster centre of described elastic information image; Wherein, cluster centre comprises territory, hard area cluster centre and soft region clustering center; Calculate on described elastic information image the fuzzy similarity matrix of each element and the square distance of cluster centre and, obtain the degree of membership that each element belongs to soft or hard interval; According to degree of membership, determine the soft or hard attribute of ROI regional organization.

9. pathological tissues growth monitoring system according to claim 6, is characterized in that, described infiltration degree tracking module is further used for:

Obtain the B mode image of elastic information image; On B mode image, the zone boundary of pathological tissues is followed the tracks of and obtained tissue infiltration's degree border; Extract the marginal information of B mode image and be fused on elastic information image, obtaining the invade tissues regional extent on elastic information image.

10. pathological tissues growth monitoring system according to claim 6, is characterized in that, described infiltration growth characteristics parameter comprises: wetted area hardness ratio, elastic characteristic ratio, elasticity average, elasticity variance and elasticity mark.