CN114366164A - Ultrasonic quantitative evaluation and risk assessment method for atherosclerotic plaque - Google Patents
Ultrasonic quantitative evaluation and risk assessment method for atherosclerotic plaque Download PDFInfo
- Publication number
- CN114366164A CN114366164A CN202210053468.1A CN202210053468A CN114366164A CN 114366164 A CN114366164 A CN 114366164A CN 202210053468 A CN202210053468 A CN 202210053468A CN 114366164 A CN114366164 A CN 114366164A
- Authority
- CN
- China
- Prior art keywords
- plaque
- gamma
- bimodal
- distribution
- cumulative
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012502 risk assessment Methods 0.000 title claims abstract description 12
- 208000037260 Atherosclerotic Plaque Diseases 0.000 title claims abstract description 11
- 238000011158 quantitative evaluation Methods 0.000 title claims abstract description 9
- 238000009826 distribution Methods 0.000 claims abstract description 94
- 230000001186 cumulative effect Effects 0.000 claims abstract description 79
- 230000002902 bimodal effect Effects 0.000 claims abstract description 68
- 238000005315 distribution function Methods 0.000 claims abstract description 36
- 238000001514 detection method Methods 0.000 claims abstract description 13
- 238000003384 imaging method Methods 0.000 claims abstract description 7
- 238000013179 statistical model Methods 0.000 claims description 44
- 238000002604 ultrasonography Methods 0.000 claims description 18
- 210000001715 carotid artery Anatomy 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 4
- 230000006835 compression Effects 0.000 claims description 3
- 238000007906 compression Methods 0.000 claims description 3
- 210000004204 blood vessel Anatomy 0.000 claims description 2
- 238000012805 post-processing Methods 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims description 2
- 230000000740 bleeding effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 150000002632 lipids Chemical class 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 208000024172 Cardiovascular disease Diseases 0.000 description 2
- 208000007536 Thrombosis Diseases 0.000 description 2
- 230000002308 calcification Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 201000001320 Atherosclerosis Diseases 0.000 description 1
- 206010016654 Fibrosis Diseases 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000003143 atherosclerotic effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000002592 echocardiography Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000004761 fibrosis Effects 0.000 description 1
- 230000003176 fibrotic effect Effects 0.000 description 1
- 238000010562 histological examination Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001338 necrotic effect Effects 0.000 description 1
- 238000009828 non-uniform distribution Methods 0.000 description 1
- 230000007406 plaque accumulation Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000009885 systemic effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012285 ultrasound imaging Methods 0.000 description 1
- 230000002792 vascular Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
- A61B8/0833—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures
- A61B8/085—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures for locating body or organic structures, e.g. tumours, calculi, blood vessels, nodules
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
- A61B8/0891—Detecting organic movements or changes, e.g. tumours, cysts, swellings for diagnosis of blood vessels
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/44—Constructional features of the ultrasonic, sonic or infrasonic diagnostic device
- A61B8/4483—Constructional features of the ultrasonic, sonic or infrasonic diagnostic device characterised by features of the ultrasound transducer
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/46—Ultrasonic, sonic or infrasonic diagnostic devices with special arrangements for interfacing with the operator or the patient
- A61B8/467—Ultrasonic, sonic or infrasonic diagnostic devices with special arrangements for interfacing with the operator or the patient characterised by special input means
- A61B8/469—Ultrasonic, sonic or infrasonic diagnostic devices with special arrangements for interfacing with the operator or the patient characterised by special input means for selection of a region of interest
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/5215—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data
Abstract
The invention relates to an ultrasonic quantitative evaluation and risk assessment method for atherosclerotic plaques, which belongs to the field of ultrasonic imaging and comprises the following steps: reconstructing ultrasonic RF data by using TGC, and then carrying out envelope detection to obtain echo envelope data and a B ultrasonic image; selecting the whole plaque area in the B-ultrasonic image as an ROI, calculating a probability distribution histogram of echo envelope data, obtaining actual cumulative probability distribution, and establishing a bimodal gamma model; using echo envelope data of the whole plaque as input, initializing model parameters, calculating cumulative probability distribution, and estimating the model parameters by using a nonlinear least square fitting method until the maximum error between an estimated cumulative distribution function and actual cumulative probability distribution does not exceed a specified value, so as to obtain corresponding parameters of a fitting bimodal gamma model; and obtaining the estimated cumulative probability distribution by using the model, comparing the estimated cumulative probability distribution with the actual cumulative density function, calculating a characteristic value E, estimating the plaque type and carrying out risk assessment.
Description
Technical Field
The invention belongs to the technical field of ultrasonic imaging, and relates to a method for fitting plaque envelope characteristic distribution, quantitatively evaluating plaque stability and performing risk evaluation based on statistical modeling of plaque ultrasonic imaging echoes.
Background
Atherosclerosis is a systemic disease of the whole body and is a relatively common cause of cardiovascular disease. The main cause of atherosclerotic lesions is the rupture of vulnerable plaque to form a thrombus. Not all plaques eventually break down leading to thrombus, which is divided into stable and unstable plaques. Therefore, the early evaluation of plaque risk and identification of unstable vulnerable plaque are of great significance for the prevention and treatment of sudden cardiovascular diseases. Regular ultrasound examination of the carotid arteries has become the primary method for clinical general screening of atherosclerotic plaques. The ultrasonic echogenic features and histological examination of plaques indicate that hyperechoic plaques are mostly composed of fibrosis and calcification components and are stable plaques, while hypoechoic plaques are mostly rich in lipid components, bleeding and necrotic substances and are vulnerable plaques. The existing clinical diagnosis of the vascular plaque mainly depends on an ultrasonographer and a radiologist to subjectively judge the risk degree of the plaque through the echo characteristic of the plaque in a B-ultrasonic image. However, this method relies on the operator's proficiency in the equipment and experience with plaque identification, and diagnostic results are susceptible to subjective factors. Therefore, a method for quantitatively analyzing the ultrasonic echo characteristics of the plaque area is developed, which can objectively analyze the echo characteristics of the plaque and reduce errors caused by human factors in plaque stability judgment. The method greatly improves the existing plaque detection method, improves the reliability of vulnerable plaque detection, and provides an automatic ultrasonic quantitative evaluation method for vulnerable plaque detection.
Quantitative analysis of plaque is typically a quantitative measurement of conventional grayscale images. The median gray level (GSM) of plaque ultrasound images was found to be linearly related to plaque type. However, the complexity of the ultrasound gray scale image is high, and the GSM only considers the gray scale median, so that the method is too many facets. The mathematical statistic model can embody the characteristics of the image more deeply, and can possibly enlarge the characteristics of the image to achieve the effect of image enhancement, and can be used as an objective evaluation method for the plaque classification. The ultrasound backscatter signal is typically modeled with a Rayleigh (Rayleigh) distribution or a non-Rayleigh (non-Rayleigh) distribution. Later studies found that other statistical distribution models, such as K-distribution, can also be used for statistical modeling of the ultrasound envelope signal. However, these models are not sufficiently matched in describing the plaque echo signal, and therefore a new statistical model needs to be established to describe the ultrasound envelope signal. It was found that the gamma distribution can be used for classification of radar targets, the ultrasound backscatter signal is similar to radar signals, and therefore the probability density function (pdf) of the gamma distribution can also be used for medical ultrasound signals. However, this single type of distribution is not suitable for non-uniform tissue. Atherosclerotic plaques generally contain different histological components such as lipids, bleeding, fibrous tissue, and calcification. Therefore, a single gamma distribution is also not suitable for statistical modeling of plaque. Therefore, there is a need to develop a multimodal gamma distribution to realize modeling statistics for different types of plaques, to realize classification of different plaques and identification of vulnerable plaques.
Disclosure of Invention
Aiming at the practical problems and the technical defects, the objective classification of the plaque and the automatic identification of the vulnerable plaque are realized, and the invention provides an ultrasonic quantitative evaluation and risk assessment method for the atherosclerotic plaque.
The invention adopts the following specific scheme:
an ultrasonic quantitative assessment and risk assessment method for atherosclerotic plaques comprises the following steps:
firstly, reconstructing the obtained carotid artery ultrasonic RF data by using a time depth gain compensation function (TGC), and then carrying out envelope detection to obtain echo envelope data and a B ultrasonic image;
selecting the whole plaque area in the B-mode ultrasonic image as a region of interest (ROI), calculating a probability distribution histogram of plaque echo envelope data of the ROI, then obtaining actual cumulative probability distribution, and establishing a bimodal gamma statistical model;
thirdly, using echo envelope data of the whole plaque as input, initializing parameters of the bimodal gamma statistical model, calculating a Cumulative Distribution Function (CDF), and estimating the parameters of the bimodal gamma statistical model by using a nonlinear least square fitting method until the maximum error between the fitted cumulative distribution function and the actual cumulative probability distribution does not exceed a specified value, so as to obtain corresponding parameters of the estimated bimodal gamma statistical model;
and step four, obtaining an estimated cumulative distribution function by using the obtained bimodal gamma statistical model, comparing the estimated cumulative distribution function with the actual cumulative probability distribution, calculating a characteristic value E, estimating the plaque type and carrying out risk assessment.
Firstly, reconstructing the obtained carotid artery ultrasonic RF data by using a time gain compensation function (TGC), and then carrying out envelope detection to obtain echo envelope data and a B ultrasonic image, wherein the specific steps are as follows:
(1) reading parameter settings in an ultrasonic imaging system to obtain a proper gain curve, selecting a proper gain curve beta according to the system settings, calculating gain as exp (beta multiplied by d) according to depth d, performing time gain compensation on echo amplitude to obtain an echo signal after depth compensation, and realizing gain compensation according to the depth of ultrasonic propagation to compensate ultrasonic attenuation caused by biological tissues;
(2) carrying out Hilbert transformation on the echo signals, taking absolute values, then solving square roots, realizing envelope detection demodulation, and obtaining low-frequency envelope signals from the echo signals to obtain echo envelope signals;
(3) and performing post-processing such as secondary sampling, logarithmic compression and the like on the echo envelope signals to obtain a B ultrasonic image of the carotid artery.
Selecting the whole plaque area in the B-mode ultrasonic image as a region of interest (ROI), calculating a probability distribution histogram of plaque echo envelope data of the ROI, then obtaining actual cumulative probability distribution, and establishing a bimodal gamma statistical model, wherein the method specifically comprises the following steps of:
(1) identifying a plaque area in the B-mode ultrasonic image, selecting the whole plaque area as an interested area, and extracting echo envelope data of the interested area;
(2) calculating probability density distribution of the plaque envelope data of the region of interest, normalizing to obtain a probability distribution histogram, and sequentially adding to obtain cumulative probability distribution;
(3) establishing a simplest probability density function of multimodal gamma statistical distribution, namely fitting the echo envelope of the non-uniform plaque by using bimodal gamma statistical distribution only containing two gamma distributions, wherein the parameters in the model are p and a respectively1，b1，a2，b2。
Thirdly, using echo envelope data of the whole plaque as input, initializing parameters of the bimodal gamma statistical model to obtain cumulative probability distribution, estimating the parameters of the bimodal gamma statistical model by using a nonlinear least square fitting method until the maximum error between a fitted cumulative distribution function and actual cumulative probability distribution does not exceed a specified value, and obtaining corresponding parameters of the estimated bimodal gamma statistical model; the method comprises the following specific steps:
(1) estimating 5 parameters in the bimodal model by using a cumulative distribution function of gamma random distribution, and firstly establishing a cumulative density function F (x) (p) gamma (x, a) of the bimodal model1,b1)+(1-p)*gamma(x,a2,b2) Then, substituting the echo envelope of the plaque as an input into a cumulative distribution function;
(2) selecting a proper value to initialize 5 parameters in the cumulative distribution function of the bimodal gamma model, and fitting the model parameters by using a nonlinear least square fitting method to obtain 5 parameters of the bimodal gamma statistical model;
(3) calculating the total error between the actual cumulative probability distribution and the fitting cumulative distribution function of the bimodal gamma statistical model, and judging the error, wherein if the error is less than the maximum allowable error of 0.0085, the parameter estimation of the bimodal gamma statistical model is completed;
(4) and if the error is larger than 0.0085, fitting again and calculating the total error until the total error is smaller than 0.0085, finishing the statistical modeling of the patch echo envelope, and obtaining a fitted bimodal gamma statistical model.
Fourthly, obtaining an estimated cumulative distribution function by using the obtained bimodal gamma statistical model, comparing the estimated cumulative distribution function with actual cumulative probability distribution, calculating a characteristic value E, estimating the plaque type and carrying out risk assessment, wherein the method specifically comprises the following steps:
(1) obtaining an estimated cumulative probability density distribution function by using the obtained bimodal gamma statistical model, comparing the estimated cumulative probability density distribution function with an original cumulative probability distribution curve, and evaluating the fitting degree;
(2) selecting a proper characteristic value as an evaluation standard of the plaque type, setting a characteristic value E of the bimodal gamma as p a based on that the mean value of the gamma distribution is a b1b1+(1-p)*a2b2As a parameter for subsequent plaque classification;
(3) and calculating a characteristic value E of the plaque in the blood vessel, and evaluating the plaque to be detected by combining the cumulative density distribution curve to realize plaque classification and risk evaluation.
Compared with the prior art, the invention has the following beneficial effects:
1. the statistical modeling is carried out on the echo envelope data of the plaque by adopting a statistical model to analyze the echo characteristics of the plaque and amplify the echo characteristics of the plaque, so that the objective division of the plaque types can be realized, and errors caused by human factors can be reduced.
2. During radio frequency data processing, a time depth gain compensation design TGC amplifier is adopted to make up ultrasonic attenuation caused by biological tissues and improve the accuracy of characteristic value E estimation.
3. The echo envelope of the non-uniform plaque is modeled by the simplest bimodal gamma distribution in the multimodal gamma distribution model, and the echo characteristics of different components of the atherosclerotic plaque can be better fitted by utilizing two distribution components.
4. The characteristic parameter E is provided as a characteristic value of plaque classification, the quantitative evaluation of plaque echo characteristics is realized by combining a cumulative distribution function, the risk of the plaque is evaluated, and objective reference basis is provided for the subsequent treatment of the plaque.
Drawings
FIG. 1 is a flow chart of the overall algorithm for statistical modeling evaluation of bimodal gamma distribution of plaque;
FIG. 2 is a plot fitting result of plaque accumulation distribution curves; in the figure, the left side (a) (c) (e) is the original ultrasonic image of calcified plaque, mixed echo plaque and hypoechoic plaque, and the right side (b) (d) (f) is the corresponding plaque actual cumulative probability distribution curve and the fitted cumulative probability distribution curve;
FIG. 3 is a cumulative probability density distribution curve obtained by bimodal gamma statistical modeling of different types of patches;
fig. 4 is a statistical analysis (n-20) of the feature value E of calcified plaque, mixed plaque, and hypoechoic plaque.
Detailed Description
The present invention will be described in detail with reference to the following embodiments and accompanying drawings. The invention relates to the programming realization of algorithms on an MATLAB platform.
The invention provides an ultrasonic quantitative evaluation and risk assessment method for carotid atherosclerotic plaques based on bimodal gamma statistical distribution. Fig. 1 is an overall flow chart of the measurement method, which includes the following steps:
firstly, reconstructing the obtained carotid artery ultrasonic RF data by using a time gain compensation function (TGC), and then carrying out envelope detection to obtain echo envelope data and a B ultrasonic image; the method comprises the following specific steps:
(1) reading parameter setting of an ultrasonic imaging system by using MATLAB to obtain imaging depth d and imaging center frequency f0And adopting time depth gain compensation, and selecting a gain curve as follows:
β＝ln 10αf/20 (1)
where α is the attenuation coefficient of the biological tissue and f is the center frequency of the transducer in ultrasound imaging of the target tissue.
Then the gain compensation curve is calculated according to the depth:
TGC＝eβd (2)
time gain compensation is carried out on the echo amplitude to obtain an echo signal after depth compensation:
A1＝A0×TGC
because the ultrasonic wave can generate transmission attenuation when propagating in the tissue, the echo signal of the deep tissue of the human body is smaller than the signal of the superficial part, therefore, the echo signal of the deep part needs to be compensated, so that the time gain compensation that the gain is increased along with the increase of the distance is adopted to compensate the ultrasonic attenuation caused by the biological tissue, and more accurate echo data is obtained.
(2) And (2) carrying out envelope detection processing on the echo data by adopting a Hilbert transform demodulation method, extracting low-frequency tissue physiological signals, carrying out Hilbert transform on the echo signals subjected to time gain compensation obtained in the step one, taking absolute values, and then solving square roots to realize envelope detection demodulation so as to obtain echo envelope signals.
(3) The echo envelope signal is subjected to secondary sampling by using an average value method, then 1 is added to the envelope signal subjected to secondary sampling, then the natural logarithm is taken (the addition of 1 is mainly used for avoiding the situation that the logarithm is taken for 0), the logarithm compression processing is realized, then the dynamic range of image display is determined, the image is displayed, and the B-mode ultrasonic image is obtained.
Selecting the whole plaque area in the B-mode ultrasonic image as a region of interest (ROI), calculating a probability distribution histogram of plaque echo envelope data of the ROI, then obtaining actual cumulative probability distribution, and establishing a bimodal gamma statistical model, wherein the method specifically comprises the following steps of:
(1) an operator identifies a plaque area in a carotid artery blood vessel through a B-mode ultrasound image, selects the whole plaque area as an ROI, identifies the abscissa and the ordinate of the whole ROI area, and then extracts echo envelope data of the plaque according to the abscissa and the ordinate of the ROI area as input of a bimodal gamma statistical model;
(2) calculating probability density distribution of the region-of-interest plaque envelope data, selecting the minimum value and the maximum value of ROI envelope data, taking 1 as an interval as the interval number of gray level histogram division, respectively counting the frequency number in a group, dividing by the total frequency number to obtain probability density distribution, and sequentially accumulating the probability density to obtain cumulative probability distribution.
(3) And selecting the simplest bimodal gamma distribution in the multimodal gamma statistical model as a statistical model for fitting the ultrasonic envelope signal. The probability density function of the gamma statistic of the ultrasound signal X is:
where Γ is the gamma distribution of the a-order. The parameters a and b are shape parameters and scale parameters, respectively. The mean of the ultrasound signal is then:
E(X)＝ab (4)
although the gamma distribution can accurately describe most of the ultrasound images, it cannot satisfy the modeling of the ultrasound signals of non-uniform distribution. Plaque in carotid ultrasound images generally contains four main components: lipid, bleeding, fibrotic and calcific components. Therefore, a unimodal gamma distribution is not suitable for modeling carotid ultrasound images. Therefore, we chose a bimodal gamma statistical model that is more suitable for non-homogeneous tissue to model the ultrasound envelope signal of plaque. The bimodal gamma distribution can be expressed as:
f(x)＝pf1(x)+(1-p)f2(x) (5)
wherein f is1(x) And f2(x) Is two probability density functions that can describe two classes of tissue. p is the probability that a measurement is required. Thus, we use a bimodal gamma statistical distribution of:
wherein a is1，b1And a2，b2Are parameters of two gamma distributions.
Thirdly, using echo envelope data of the whole plaque as input, initializing parameters of the bimodal gamma statistical model, calculating a Cumulative Distribution Function (CDF), and estimating the parameters of the bimodal gamma statistical model by using a nonlinear least square fitting method until the maximum error between the fitted cumulative distribution function and the actual cumulative probability distribution does not exceed a specified value, so as to obtain corresponding parameters of the estimated bimodal gamma statistical model; the method comprises the following specific steps:
(1) obtaining a cumulative distribution function of the bimodal gamma distribution according to the bimodal gamma distribution in the step two:
equation (7) can be simplified as:
F(xi,p,a1,b1,a2,b2)＝p gamcdf(xi,a1,b1)+(1-p)gamcdf(xi,a2,b2) (8)
(2) substituting the ultrasonic echo envelope data of the plaque into a cumulative distribution function of bimodal gamma distribution, selecting a proper array to carry out initialization setting on 5 parameters in the cumulative distribution function, then fitting the model parameters by using a nonlinear least square fitting method, and outputting 5 parameters of the bimodal gamma statistical model after fitting.
(3) Substituting the parameters into the bimodal gamma distribution to obtain a fitted cumulative distribution function, and then calculating an error epsilon between the actual cumulative probability distribution and the fitted cumulative distribution function, wherein the calculation formula is as follows:
where e _ cdf is the actual cumulative probability distribution and F is the cumulative probability distribution resulting from a bimodal gamma fit.
In order to ensure the accuracy of fitting, the error epsilon is required to be less than 0.0085, if the obtained total error is less than 0.0085, 5 parameters of bimodal gamma distribution are obtained, and the envelope statistical modeling is completed.
(4) If the estimation error of the cumulative distribution function is larger than 0.0085, fitting by using the nonlinear least square method again, calculating the total estimation error, and circulating in sequence until the cumulative probability distribution error is smaller than the allowable error of 0.0085, finishing the parameter estimation of the bimodal gamma distribution, and obtaining the bimodal gamma statistical model conforming to the plaque ultrasonic data.
Step four, obtaining an estimated cumulative distribution function by using the obtained bimodal gamma statistical model, comparing the estimated cumulative distribution function with actual cumulative probability distribution, calculating a characteristic value E, and estimating a plaque type and a risk type; the method comprises the following specific steps:
(1) substituting the finally obtained parameters of the bimodal gamma statistical model into a formula (8) by using the parameters of the bimodal gamma statistical model obtained in the step three to obtain an estimated cumulative distribution curve, comparing the estimated cumulative distribution curve with the actual cumulative probability distribution, and evaluating the consistency degree of bimodal gamma fitting;
(2) selecting a proper characteristic value as a classification standard of quantitative evaluation of plaques, selecting a characteristic value E as a standard for classifying the plaques by bimodal gamma fitting based on a mean value calculation formula of gamma distribution, wherein the calculation formula of the characteristic value E is as follows:
E＝p*a1b1+(1-p)*a2b2 (10)
(3) according to the relation between the echo type of the plaque and the plaque risk, the high-echo plaque generally belongs to the low-risk plaque, and the mixed echo plaque and the low-echo plaque belong to the high-risk plaque. Calculating a characteristic value E of the plaque to be detected and a fitted cumulative density distribution curve by using the method in the step, wherein the larger the characteristic value E is, the more stable the plaque is, and the smaller the risk is; the smaller the eigenvalue E, the more unstable the plaque and the greater the risk, as shown in table 1. The calcified plaque has the largest characteristic value and the strongest stability, while the hypoechoic plaque has the smallest characteristic value and the poorer stability, and belongs to vulnerable plaque.
Table 1 5 parameters and eigenvalues for different types of plaque fits.
To further test the feasibility of the method of the present invention, data of 20 calcified plaque groups, 20 mixed plaque groups and 20 hypoechoic plaque groups (60 cases in total) were collected, and the characteristic values of plaque regions were calculated, and the statistical results are shown in fig. 4. The distribution intervals of the characteristic values of the calcified plaque, the mixed plaque and the low-echo plaque are different, the characteristic value E of the calcified plaque is the largest, the characteristic value E of the low-echo plaque is the smallest, and the ranges of the characteristic values of the plaques of different types are not overlapped; the mean value of the characteristic values of the calcified plaque is 44.87, the mean value of the characteristic value of the mixed plaque is 29.92, and the mean value of the characteristic value of the low-echo plaque is 16.92, which are different from each other. Thus, the statistical results show that the method of the invention can be used for distinguishing different types of plaques and evaluating the risk of the plaques.
It should be noted that the above-mentioned embodiments illustrate rather than limit the scope of the invention, which is defined by the appended claims. It will be apparent to those skilled in the art that certain insubstantial modifications and adaptations of the present invention can be made without departing from the spirit and scope of the invention.
Claims (4)
1. The method for the ultrasonic quantitative evaluation and risk assessment of the atherosclerotic plaque is characterized by comprising the following steps:
firstly, reconstructing the obtained carotid artery ultrasonic RF data by using a time depth gain compensation function (TGC), and then carrying out envelope detection to obtain echo envelope data and a B ultrasonic image;
selecting the whole plaque area in the B-mode ultrasonic image as a region of interest (ROI), calculating a probability distribution histogram of plaque echo envelope data of the ROI, then obtaining actual cumulative probability distribution, and establishing a bimodal gamma statistical model;
thirdly, using echo envelope data of the whole plaque as input, initializing parameters of the bimodal gamma statistical model, calculating a Cumulative Distribution Function (CDF), and estimating the parameters of the bimodal gamma statistical model by using a nonlinear least square fitting method until the maximum error between the fitted cumulative distribution function and the actual cumulative probability distribution does not exceed a specified value, so as to obtain corresponding parameters of the estimated bimodal gamma statistical model;
step four, obtaining an estimated cumulative distribution function by using the bimodal gamma statistical model, comparing the estimated cumulative distribution function with the actual cumulative probability distribution, calculating a characteristic value E, estimating the plaque type and carrying out risk assessment; the method comprises the following specific steps:
(1) obtaining an estimated cumulative distribution function by using a bimodal gamma statistical model, comparing the estimated cumulative distribution function with an original cumulative probability distribution curve, and evaluating the fitting degree;
(2) selecting proper characteristic value asFor evaluation criteria of plaque type, a characteristic value E ═ p ×, a of bimodal gamma is set based on the mean value of gamma distribution as a × b1b1+(1-p)*a2b2As a parameter for subsequent plaque classification;
(3) and calculating a characteristic value E of the plaque in the blood vessel, and evaluating the plaque to be detected by combining an accumulative probability distribution curve to realize plaque classification and risk evaluation.
2. The method for ultrasonic quantitative assessment and risk assessment of atherosclerotic plaques according to claim 1, wherein the step one of reconstructing the obtained carotid artery ultrasonic RF data by using a time gain compensation function (TGC) and then carrying out envelope detection to obtain echo envelope data and B-mode ultrasound images comprises the following specific steps:
(1) reading parameter settings in an ultrasonic imaging system to obtain a proper gain curve, selecting a proper gain curve beta according to the system settings, calculating gain as exp (beta multiplied by d) according to depth d, performing time gain compensation on echo amplitude to obtain an echo signal after depth compensation, and realizing gain compensation according to the depth of ultrasonic propagation to compensate ultrasonic attenuation caused by biological tissues;
(2) carrying out Hilbert transformation on the echo signals, taking absolute values, then solving square roots, realizing envelope detection demodulation, and obtaining low-frequency envelope signals from the echo signals to obtain echo envelope signals;
(3) and performing post-processing such as secondary sampling, logarithmic compression and the like on the echo envelope signals to obtain a B ultrasonic image of the carotid artery.
3. The method according to claim 1, wherein the second step of selecting the whole plaque area in the B-mode ultrasound image as a region of interest (ROI), calculating a probability distribution histogram of plaque echo envelope data of the region of interest, obtaining an actual cumulative probability distribution, and establishing a bimodal gamma statistical model comprises the following specific steps:
(1) identifying a plaque area in the B-mode ultrasonic image, selecting the whole plaque area as an interested area, and extracting echo envelope data of the interested area;
(2) calculating probability density distribution of the plaque envelope data of the region of interest, normalizing to obtain a probability distribution histogram, and sequentially adding to obtain cumulative probability distribution;
(3) establishing a simplest probability density function of multimodal gamma statistical distribution, namely fitting the echo envelope of the non-uniform plaque by using bimodal gamma statistical distribution only containing two gamma distributions, wherein the parameters in the model are p and a respectively1，b1，a2，b2。
4. The method according to claim 1, wherein the echo envelope data of the whole plaque is used as input and the parameters of the bimodal gamma statistical model are initialized to obtain cumulative probability distribution, and the parameters of the bimodal gamma statistical model are estimated by using a nonlinear least square fitting method until the maximum error between the fitted cumulative distribution function and the actual cumulative probability distribution does not exceed a specified value, so as to obtain the corresponding parameters of the estimated bimodal gamma statistical model; the method comprises the following specific steps:
(1) estimating 5 parameters in the bimodal model by using a cumulative distribution function of gamma random distribution, and firstly establishing a cumulative density function F (x) (p) gamma (x, a) of the bimodal model1,b1)+(1-p)*gamma(x,a2,b2) Then, substituting the echo envelope of the plaque as an input into a cumulative distribution function;
(2) selecting a proper value to initialize 5 parameters in the cumulative distribution function of the bimodal gamma model, and fitting the model parameters by using a nonlinear least square fitting method to obtain 5 parameters of the bimodal gamma statistical model;
(3) calculating the total error between the actual cumulative probability distribution and the cumulative distribution function fitted by the bimodal gamma statistical model, and judging the error, wherein if the error is less than the maximum allowable error of 0.0085, the parameter estimation of the bimodal gamma statistical model is completed;
(4) and if the error is larger than 0.0085, fitting again and calculating the total error until the total error is smaller than 0.0085, finishing the statistical modeling of the patch echo envelope, and obtaining a fitted bimodal gamma statistical model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210053468.1A CN114366164B (en) | 2022-01-18 | 2022-01-18 | Ultrasonic quantitative evaluation and risk evaluation method for atherosclerosis plaque |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210053468.1A CN114366164B (en) | 2022-01-18 | 2022-01-18 | Ultrasonic quantitative evaluation and risk evaluation method for atherosclerosis plaque |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114366164A true CN114366164A (en) | 2022-04-19 |
| CN114366164B CN114366164B (en) | 2024-05-03 |
Family
ID=81144843
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210053468.1A Active CN114366164B (en) | 2022-01-18 | 2022-01-18 | Ultrasonic quantitative evaluation and risk evaluation method for atherosclerosis plaque |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114366164B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011118267A1 (en) * | 2010-03-26 | 2011-09-29 | 国立大学法人徳島大学 | Method for identifying nature of carotid plaque and evaluation device |
| CN102800088A (en) * | 2012-06-28 | 2012-11-28 | 华中科技大学 | Automatic dividing method of ultrasound carotid artery plaque |
| CN111265252A (en) * | 2020-02-25 | 2020-06-12 | 上海市同仁医院 | Method for evaluating stability of atherosclerotic plaque |
| CN111667467A (en) * | 2020-05-28 | 2020-09-15 | 江苏大学附属医院 | Clustering algorithm-based lower limb vascular calcification index multi-parameter accumulation calculation method |
| CN112328971A (en) * | 2020-11-10 | 2021-02-05 | 中国船舶重工集团海装风电股份有限公司 | Double-peak three-parameter Weibull distribution fitting method based on random number |
-
2022
- 2022-01-18 CN CN202210053468.1A patent/CN114366164B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011118267A1 (en) * | 2010-03-26 | 2011-09-29 | 国立大学法人徳島大学 | Method for identifying nature of carotid plaque and evaluation device |
| CN102800088A (en) * | 2012-06-28 | 2012-11-28 | 华中科技大学 | Automatic dividing method of ultrasound carotid artery plaque |
| CN111265252A (en) * | 2020-02-25 | 2020-06-12 | 上海市同仁医院 | Method for evaluating stability of atherosclerotic plaque |
| CN111667467A (en) * | 2020-05-28 | 2020-09-15 | 江苏大学附属医院 | Clustering algorithm-based lower limb vascular calcification index multi-parameter accumulation calculation method |
| WO2021238739A1 (en) * | 2020-05-28 | 2021-12-02 | 江苏大学附属医院 | Clustering algorithm-based multi-parameter cumulative calculation method for calcification index of lower limb blood vessels |
| CN112328971A (en) * | 2020-11-10 | 2021-02-05 | 中国船舶重工集团海装风电股份有限公司 | Double-peak three-parameter Weibull distribution fitting method based on random number |
Non-Patent Citations (1)
| Title |
|---|
| 杨天睿;撒亚莲;: "云南汉族γ-干扰素受体基因多态性与动脉粥样硬化斑块稳定性的相关性", 重庆医学, no. 33 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114366164B (en) | 2024-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7231074B2 (en) | Method for determining the efficacy of an anti-cancer treatment using image analysis | |
| Tsui et al. | Small-window parametric imaging based on information entropy for ultrasound tissue characterization | |
| Nam et al. | Quantitative assessment of in vivo breast masses using ultrasound attenuation and backscatter | |
| Oosterveld et al. | Ultrasound attenuation and texture analysis of diffuse liver disease: methods and preliminary results | |
| Hsu et al. | Breast tumor classification using different features of quantitative ultrasound parametric images | |
| CN105377145A (en) | System and method for classifying and characterizing tissues using first-order and second-order statistics of quantitative ultrasound parametric maps | |
| Tsui et al. | Effects of estimators on ultrasound Nakagami imaging in visualizing the change in the backscattered statistics from a Rayleigh distribution to a pre-Rayleigh distribution | |
| Vegas-Sanchez-Ferrero et al. | Gamma mixture classifier for plaque detection in intravascular ultrasonic images | |
| Lizzi et al. | On the statistics of ultrasonic spectral parameters | |
| Moon et al. | Computer-aided diagnosis based on speckle patterns in ultrasound images | |
| Yamaguchi | Basic concept and clinical applications of quantitative ultrasound (QUS) technologies | |
| Khairalseed et al. | Classification of thyroid nodules in H-scan ultrasound images using texture and prinicipal component analysis | |
| US10338033B2 (en) | Typing and imaging of biological and non-biological materials using quantitative ultrasound | |
| Shankar et al. | Application of the compound probability density function for characterization of breast masses in ultrasound B scans | |
| CN114366164B (en) | Ultrasonic quantitative evaluation and risk evaluation method for atherosclerosis plaque | |
| Zenteno et al. | Improving the quality of attenuation imaging using full angular spatial compounding | |
| Chang et al. | Automatic detection of microcalcifications in breast ultrasound | |
| Hu et al. | Ultrasound speckle reduction based on histogram curve matching and region growing | |
| Qi et al. | Detection of carotid intima and media thicknesses based on ultrasound B-mode images clustered with Gaussian mixture model | |
| Muhtadi et al. | Analyzing the texture of Nakagami parametric images for classification of breast cancer | |
| WO2021018107A1 (en) | Ultrasonic signal processing method and ultrasonic signal processing apparatus, and device and storage medium | |
| Zagzebski et al. | Quantitative ultrasound: Enhancing diagnosis using estimates of acoustic attenuation and backscatter | |
| Tsui et al. | The effect of logarithmic compression on estimation of the Nakagami parameter for ultrasonic tissue characterization: a simulation study | |
| James et al. | Attenuation compensation comparison for human carotid plaque characterization using spectral analysis of backscattered ultrasound | |
| Nair et al. | 'Blind'data calibration of intravascular ultrasound data for automated tissue characterization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |