CN106264722A - Window modulation combines Gauss polynomial matching monitoring radio frequency ablation device and method - Google Patents

Abstract

The present invention relates to Medical Image Processing, ultrasonic temperature imaging, for a kind of method proposing resolution improving monitoring radio-frequency ablation procedure electrode position and heated perimeter.The method can be effectively improved heated perimeter in monitoring radio-frequency ablation procedure, improves the accuracy killing cancerous cell scope of radio-frequency ablation procedure, is effectively improved monitoring effect.To this end, the technical solution used in the present invention is, window modulation combines Gauss polynomial matching monitoring radio frequency ablation device, is provided with such as lower module: (1) frame data arrange module on the transducer of ultrasonic instrument；(2) envelope image acquisition module；(3) the compound Nakagami image collection module of Nakagami image forming module (4)；(5) time combination picture Img_T(temporal compounding image) acquisition module；(6) to the time combination picture Img obtained_TApplying one-dimensional Gauss polynomial to carry out post processing, the image obtained is for monitoring radio-frequency (RF) ablation.Present invention is mainly applied to image procossing.

Description

Window modulation combines Gauss polynomial matching monitoring radio frequency ablation device and method

Technical field

The present invention relates to Medical Image Processing, ultrasonic temperature imaging, radio frequency burns technical field, particularly relate to for Radio frequency burns in operation, modulates Nakagami imaging and the monitoring radio frequency ablation device of Gauss polynomial matching and calculation based on window Method research.

Background technology

In medical research, ultrasonography is mainly used in fetus and produces inspection, breast ultrasound detection [1], observation blood vessel The flowing of middle blood and the motion [2,3] etc. of observation internal organs；The transducer mid frequency of ultrasonic instrument arrives between 5MHz Between 15MHz.Wherein the main imaging mode for picture control includes mode A (amplitude modulation), B mould Formula (brightness modulation), M-mode (motion modulation) and doppler mode；The most especially with B-mode Be most widely used.The operation principle of ultrasonography is mainly sent ultrasonic signal by transducer, then ultrasound wave letter After number being reflected by organ-tissue, then by pulse-echo technique (pulse echo technique), these ultrasonic reflections are believed Number it is back to supersonic transducer, is then passed through imaging on instrument after algorithm process via transducer.

Primary hepatocarcinoma is the most current modal liver malignancy, and its sickness rate increases the most day by day, Because B-mode and infection with hepatitis C virus sickness rate is more and more higher.Excision and liver transplantation are considered as to treat liver at present The primary treatments of cell carcinoma.But, only small number of patients is suitable for operation.Therefore, effective and safe controlling is formulated Treat the therapeutic strategy of hepatocarcinoma it is critical that [1-3].

Radio frequency ablation therapy (radio frequency ablation) is that wound is little, cost is low and is applied to face for one One of the thermal therapeutical method of most successful treatment primary hepatocarcinoma of bed [4].During treatment, doctor can melt electricity one Pole is inserted into the cancer site of patient, then carrys out the cancer site to patient by instrument controlling power and treats.Therefore in treatment During observe heated by electrodes the size of scope most important, it is necessary to heated perimeter is controlled to most suitable scope to protect Patient is caused the post-surgical trauma of minimum by card.And Nakagami imaging mode is in the formation method that monitoring video melts now A kind of imaging algorithm that precision is higher.

The statistical property effectively describing hepatic tissue of ultrasonic Nakagami Parameter Map, research before the most fully demonstrates Use sliding window on Nakagami image is used to monitor the feasibility of the method for radio-frequency (RF) ablation.Meanwhile, based on The Nakagami imaging of Nakagami statistical distribution model is also a kind of method [5] effectively supplementing conventional B model.Had before Research uses Nakagami imaging to combine polynomial approximation method and is applied successfully to monitor radio-frequency (RF) ablation.But the method also exists The most limited place [6].First, although multinomial application Nakagami image can effectively matching radio-frequency (RF) ablation region, But inevitably affect the detection of radio frequency field.Therefore, we should to other approximating methods such as Gaussian approximation and A young waiter in a wineshop or an inn takes advantage of the feasibility of approximation to be estimated.Secondly, this research application is tradition Nakagami imaging algorithm, because use Window is relatively big, and traditional Nakagami image does not provide high image resolution and smooths, and causes monitoring in operation poor image quality.

List of references:

1.H.Tsukuma,T.Hiyama,S.Tanaka,M.Nakao,T.Yabuuchi,T.Kitamura, K.Nakanishi,I.Fujimoto,A.Inoue,H.Yamazaki and T.Kawashima,“Risk factors for hepatocellular carcinoma among patients with chronic liver disease,”N Engl J Med.,328(25):1797-801,1993。

2.F.X.Bosch,J.Ribes,M.Díaz and R.Cléries,“Primary liver cancer: worldwide incidence and trends,”Gastroenterology,127:S5-S16,2004。

3.F.X.Bosch,J.Ribes,M.Díaz and R.Cléries,“Primary liver cancer: worldwide incidence and trends,”Gastroenterology,127:S5-S16,2004。

4.Tony Hsiu-Hsi,C.,Chien-Jen,C.,Ming-Fang,Y.,Sheng-Nan,L.,Chien-An, S.,&GuanTarn,H.,et al.(2002).Ultrasound screening and risk factors for death from hepatocellular carcinoma in a high risk group in taiwan..International Journal of Cancer Journal International Du Cancer,98(2),257–261。

5.H.C.Holfman,Statistical Methods on Radio Wave Propagation,New York: Pergamon Press,1960。

6.C.Y.Wang,X.Geng,T.S.Yeh,H.L.Liu and P.H.Tsui,“Monitoring radiofrequency ablation with ultrasound Nakagami imaging,”Med.Phys.,40: 072901,2013。

Summary of the invention

For overcoming the deficiencies in the prior art, it is contemplated that propose one improve monitoring radio-frequency ablation procedure electrode position and The method of the resolution of heated perimeter.The method can be effectively improved heated perimeter in monitoring radio-frequency ablation procedure, improves and penetrates Frequently the accuracy killing cancerous cell scope of ablative surgery, is effectively improved monitoring effect.To this end, the technical solution used in the present invention Be, window modulation combine Gauss polynomial matching monitoring radio frequency ablation device, the transducer of ultrasonic instrument is arranged just like Lower module:

(1) frame data arrange module, obtain M frame ultrasonic signal data by ultrasonic imaging system, and arrange One frame data are j=1；

(2) envelope image acquisition module, by Hilbert conversion conversion, takes the absolute value of conversion signal, thus obtains The envelope image of the radiofrequency signal of ablation tissue；

(3) Nakagami image forming module, utilizes N number of window to form Nakagami image: assume that (x y) is W Nakagami image, x and y represents the transverse and longitudinal coordinate of Nakagami image respectively, and zero is the upper left corner, and W_j1(x,y), W_j2(x,y),W_j3(x, y) ..., and W_jN(x, y) represents in jth Frame, and sliding window is from the arteries and veins of 1 transducer increasing to N times Rush the Nakagami image that length computation obtains.Then, the window of jth frame is combined Nakagami image W_jm(x, y) by following Formula obtains:

$W_{jm}(x,y)=\frac{1}{N}\underset{i=1}{\overset{N}{\Σ}}{W}_{ji}(x,y)---\left(1\right)$

(4) compound Nakagami image collection module, for increasing to M at j from 2, repeats calling module (2) and (3), so The most just obtain M frame window and be combined Nakagami image: W_1m(x,y),W_2m(x,y),W_3m(x, y) ..., and W_Mm(x,y)；

(5) time combination picture Img_T(temporal compounding image) acquisition module, is by owning The window obtained by (4) step be combined Nakagami image addition and average and obtain time combination picture Img_T；

(6) to the time combination picture Img obtained_TApplying one-dimensional Gauss polynomial to carry out post processing, the image obtained is used Monitor radio-frequency (RF) ablation.

Window modulation combines Gauss polynomial matching monitoring radio-frequency (RF) ablation method, and step is,

(1) obtain M frame ultrasonic signal data by ultrasonic imaging system, and to arrange the first frame data be j=1；

(2) changed by Hilbert, take the absolute value of conversion signal, thus obtain the bag of the radiofrequency signal of ablation tissue Network image；

(3) utilize N number of window to form Nakagami image: to assume that (x, y) is Nakagami image to W, and W_j1(x,y), W_j2(x,y),W_j3(x, y) ..., and W_jN(x, y) representing jth Frame window length of side Nakagami image increases to N times from 1 The pulse length of transducer, W represents the Nakagami image of jth Frame, and increases to from 1 along with window width The transducer pulse length of N times, then, the window of jth frame is combined Nakagami image W_jm(x y) can pass through formula (1) and obtain .

(4) j increases to M from 2, repeats step (2) and (3), has the most just obtained M frame window and has been combined Nakagami image: W_1m(x,y),W_2m(x,y),W_3m(x, y) ..., and W_Mm(x,y)；

(5) time combination picture Img_T(temporal compounding image) be by by all of by (4) step The window obtained is combined Nakagami image addition averaging and obtains；

(6) to the time combination picture Img obtained_TApplying one-dimensional Gauss polynomial to carry out post processing, the image obtained is used Monitor radio frequency to melt.

Utilize N number of window to form the compound Nakagami imaging (WMC of Nakagami image i.e. window modulation Nakagamiimaging) step is specifically, in Nakagami is distributed, and determining of the probability density function of scattering signals envelope R Justice is as follows:

$f\left(R\right)=\frac{2m^m{R}^{2m-1}}{\mathrm{\Γ}\left(m\right){\mathrm{\Ω}}^m}\exp (-\frac{m}{\mathrm{\Ω}}{R}^2)U\left(R\right)---\left(2\right)$

Γ and U is gamma function and unit steplike function respectively, and Ω is zooming parameter, and m is relevant to Nakagami distribution Custom parameter, if E represents expectation, then Ω and being obtained by following formula with m:

Ω=E (R²) (3)

With

$m=\frac{E^2\left(R^2\right)}{Var\left(R^2\right)}---\left(4\right)$

After obtaining zooming parameter Ω and Nakagami parameter m, Nakagami model is used to describe ultrasound wave back scattering The distribution situation of son.

A square window is used to scan whole ultrasonic signal envelope image, then the value of calculation of square window As the pixel of square window position, length of window is 3 times of transducer pulse length can obtain preferable effect.

Specifically, Nakagami image has numerous row and columns in Gauss polynomial matching, for any row or column bag Containing n data point, it is expressed as coordinate form: (x_i,y_i), i=1,2,3 ..., n, wherein x_i,y_iIt is respectively i-th data point transverse and longitudinal The coordinate of coordinate axes, if a stochastic variable X Gaussian distributed, and average and variance be respectively μ and σ², μ ∈ R and σ > 0, and X has the probability density function y of continuous distribution_i:

$y_i=A\×\exp (-\frac{{(x_i-\mathrm{\μ})}^2}{2{\mathrm{\σ}}^2})---\left(5\right)$

A represents the amplitude of Gaussian function, (5) formula both sides is taken natural logrithm and obtains:

$\ln y_i=(\ln A-\frac{{\mathrm{\μ}}^2}{2{\mathrm{\σ}}^2})+\frac{x_i\mathrm{\μ}}{{\mathrm{\σ}}^2}-\frac{x_i^2}{2{\mathrm{\σ}}^2}---\left(6\right)$

Order

Then equation (7) can represent with matrix form:

$\left(\begin{array}{c}{z}_1\\ z_2\\ .\\ .\\ .\\ z_n\end{array}\right)=\left(\begin{array}{ccc}1& x_1& x_1^2\\ 1& x_2& x_2^2\\ .& .& .\\ .& .& .\\ .& .& .\\ 1& x_n& x_n^2\end{array}\right)\left(\begin{array}{c}{b}_0\\ b_1\\ b_2\end{array}\right)---\left(8\right)$

It is abbreviated as:

Z=XB (9)

Wherein matrix Z, X and B is respectively the matrix shorthand for formula (7).The least square solution of linear equation (9) is:

B=(X^TX)^-1X^TZ (10)

μ and σ of one-dimensional Gauss distribution can be obtained by the combination of formula (7) and (10).

The feature of the present invention and providing the benefit that:

Thus the present invention can be effectively improved heated perimeter in monitoring radio-frequency ablation procedure, improve killing of radio-frequency ablation procedure The accuracy of dead cancerous cell scope, monitors good image quality, is effectively improved monitoring effect in operation.

Accompanying drawing illustrates:

Fig. 1 is: the monitoring radio-frequency (RF) ablation algorithm flow chart of improvement.

Fig. 2 is: experiment frame composition.

Detailed description of the invention

The present invention adopts the technical scheme that the prison that window modulation Nakagami imaging and Gauss polynomial matching combine Survey radio-frequency (RF) ablation algorithm.

The 1.1 compound Nakagami imagings of window modulation

Nakagami distributed model is initially used to describe the statistics of radar return, and afterwards, Hampshire proposes utilization The ultrasonic scattering signals of Parameter analysis of Nakagami statistical model.In Nakagami is distributed, scattering signals envelope R's is general Rate density function is defined as follows:

$f\left(R\right)=\frac{2m^m{R}^{2m-1}}{\mathrm{\Γ}\left(m\right){\mathrm{\Ω}}^m}\exp (-\frac{m}{\mathrm{\Ω}}{R}^2)U\left(R\right)---\left(11\right)$

Γ and U is gamma function and unit steplike function respectively, and Ω is zooming parameter, and m is relevant to Nakagami distribution Custom parameter.If E represents expectation, then Ω and being obtained by following formula with m:

Ω=E (R²) (3)

With

$m=\frac{E^2\left(R^2\right)}{Var\left(R^2\right)}---\left(4\right)$

The statistical property of scattering signals depends on the characteristic of scattering.When the resolution elements of sensor contain in a large number with Scattering of machine distribution, then probability density function meets rayleigh distributed.If resolution elements comprise scattering son have with The scattering section of machine change and the variance of higher degree, then probability density function meets front rayleigh distributed.Work as resolution elements Comprise the scattering of periodically scattering and random distribution when, then probability density function meets rear rayleigh distributed. Nakagami parameter presents the probability density function rayleigh distributed in the past of envelope to the change of rear rayleigh distributed from the change of 0 to 1 Change process.Therefore Nakagami model can be used to describe the distribution situation of ultrasound wave back scattering completely.

In research before, all can use a square window to scan whole ultrasonic signal envelope image, then The value of calculation of square window is as the pixel of square window position.Big window can have stable Nakagami parameter (m) Estimate and improve the flatness of image, but simultaneously can the image resolution ratio of large losses, wicket can improve comparatively speaking The resolution of image.From the point of view of comprehensively, length of window is 3 times of transducer pulse length can obtain comparatively speaking preferably effect.

1.2 Gauss polynomial matchings

After obtaining the Nakagami image of window modulation, we use one-dimensional Gauss polynomial to be fitted it, the phase Prestige obtains more preferable imaging effect, thus completes to estimate ablation areas more accurately.Gauss polynomial matching is based on many The method of item formula matching, is included in axially and transversely direction and finds optimal Gaussian function respectively.Nakagami image has crowd Many row and columns, comprise n data point for any a line (or row), and it can be expressed as: (x_i,y_i), (i=1,2,3 ..., n).Wherein x_i,y_iThe matrix coordinate being respectively Nakagami image represents.If a variable X Gaussian distributed, and all Value and variance are respectively μ and σ²(μ ∈ R and σ > 0), and X has a probability density function of continuous distribution:

$y_i=A\×\exp (-\frac{{(x_i-\mathrm{\μ})}^2}{2{\mathrm{\σ}}^2})---\left(14\right)$

A represents the amplitude of Gaussian function.(14) formula both sides are taken natural logrithm obtain:

$\ln y_i=(\ln A-\frac{{\mathrm{\μ}}^2}{2{\mathrm{\σ}}^2})+\frac{x_i\mathrm{\μ}}{{\mathrm{\σ}}^2}-\frac{x_i^2}{2{\mathrm{\σ}}^2}---\left(15\right)$

Order

$\ln y_i=z_i,\ln A-\frac{{\mathrm{\μ}}^2}{2{\mathrm{\σ}}^2}=b_0,\frac{\mathrm{\μ}}{{\mathrm{\σ}}^2}=b_1,-\frac{1}{2{\mathrm{\σ}}^2}=b_2---\left(16\right)$

Then equation (16) can represent with matrix form:

$\left(\begin{array}{c}{z}_1\\ z_2\\ .\\ .\\ .\\ z_n\end{array}\right)=\left(\begin{array}{ccc}1& x_1& x_1^2\\ 1& x_2& x_2^2\\ .& .& .\\ .& .& .\\ .& .& .\\ 1& x_n& x_n^2\end{array}\right)\left(\begin{array}{c}{b}_0\\ b_1\\ b_2\end{array}\right)---\left(17\right)$

Can be abbreviated as:

Z=XB (18)

Wherein matrix Z, X and B is respectively the matrix shorthand for formula (17).The least square solution of then linear equation (18) For:

B=(X^TX)^-1X^TZ (19)

μ and σ of the most one-dimensional Gauss distribution can be obtained by the combination of formula (17) and (19).

The 1.3 monitoring radio-frequency (RF) ablation algorithm flows improved

The present invention is further described below in conjunction with the accompanying drawings.Monitoring radio-frequency (RF) ablation algorithm (Fig. 1) improved can be divided into six Step illustrates:

1. obtain M frame ultrasonic signal data by ultrasonic imaging system, and to arrange the first frame data be j=1.

2. changed by Hilbert, take the absolute value of conversion signal, thus obtain the bag of the radiofrequency signal of ablation tissue Network image.

3.Nakagami image forming module, utilizes N number of window to form Nakagami image: to assume that (x y) is W Nakagami image, x and y represents the transverse and longitudinal coordinate of Nakagami image respectively, and zero is the upper left corner.And W_j1(x,y), W_j2(x,y),W_j3(x, y) ..., and W_jN(x, y) represents in jth Frame, and sliding window is from the arteries and veins of 1 transducer increasing to N times Rush the Nakagami image that length computation obtains.Then, the window of jth frame is combined Nakagami image W_jm(x, y) by following Formula obtains:

$W_{jm}(x,y)=\frac{1}{N}\underset{i=1}{\overset{N}{\Σ}}{W}_{ji}(x,y)---\left(20\right)$

4.j increases to M from 2, repeats step (2) and (3).Then we just obtained M frame window be combined Nakagami figure Picture: W_1m(x,y),W_2m(x,y),W_3m(x, y) ..., and W_Mm(x,y)。

5. time combination picture Img_TIt is by all of window obtained by (4) step is combined Nakagami image addition And average and obtain.

6. pair time combination picture obtained (temporal compounding image) Img_TApply one-dimensional Gauss many Item formula carries out post processing, and the image obtained is for monitoring radio-frequency (RF) ablation.

An instantiation is presented herein below.

Before radiofrequency ablation therapy, pig liver sample is cut into appropriately sized, puts into the plastic casing of a full normal saline, Then ablating electrode is inserted into inside in vitro liver by an aperture.Clay material is used to avoid saline solution to leak.Super Sonic probe is placed on liver and is immersed in saline solution, and the distance between transducer and sample depends on Jiao of transducer Away from, this is adjustable.Therefore, sample may be located at the focus area in ultrasonic scanning.Then ultrasonic system is opened, so that it may To find the position of electrode.During radio-frequency (RF) ablation, the transducer of excusing from death wave system system ceaselessly captures and is dissipated by pig liver The scattering signals penetrated, after capturing via transducer, ultrasound wave it is two to enter that the software system carried automatically saves the most again Data processed.Then we just can be by obtaining ultrasonography to the process of these binary data.

Radio frequency ablation system, in the automatic mode operation of acquiescence, starts at 50W/min, the most automatically increases 10W/ minute, Because the output of high impedance and radio system is identical to different electrode lengths.In heating period (12 minutes), original RF data obtain respectively from tissue, scan ultrasonic scattering signals line including 256.Sample frequency and pulse length are set It it is 30 megahertzs and 0.7 millimeter.After heating period, radio frequency ablation system is automatically stopped work.The most each electrode length (0.5,1, With 1.5 centimetres) and 5 Pig Livers (n=15) do the most respectively and once test.

Obtained ultrasound data is read in Matlab program, then according at algorithm mentioned above Reason, then obtains last ultrasonography, then contrasts with the actual situation of burning in process of the test, then verify that this calculates The feasibility of method.

Claims (5)

1. window modulation combines a Gauss polynomial matching monitoring radio frequency ablation device, it is characterized in that, at ultrasonic instrument It is provided with on transducer such as lower module:

(1) frame data arrange module, obtain M frame ultrasonic signal data by ultrasonic imaging system, and arrange the first frame Data are j=1；

(2) envelope image acquisition module, by Hilbert conversion conversion, takes the absolute value of conversion signal, thus is melted The envelope image of the radiofrequency signal of tissue；

(3) Nakagami image forming module, utilizes N number of window to form Nakagami image: assume that (x y) is Nakagami to W Image, x and y represents the transverse and longitudinal coordinate of Nakagami image respectively, and zero is the upper left corner, and W_j1(x,y),W_j2(x, y),W_j3(x, y) ..., and W_jN(x, y) represents in jth Frame, and sliding window is long from the pulse of 1 transducer increasing to N times Spend calculated Nakagami image.Then, the window of jth frame is combined Nakagami image W_jm(x y) passes through below equation Obtain:

$W_{jm}(x,y)=\frac{1}{N}\underset{i=1}{\overset{N}{\Σ}}{W}_{ji}(x,y)---\left(1\right)$

(4) compound Nakagami image collection module, for increasing to M at j from 2, repeats calling module (2) and (3), the most just Obtain M frame window and be combined Nakagami image: W_1m(x,y),W_2m(x,y),W_3m(x, y) ..., and W_Mm(x,y)；

(5) time combination picture Img_T(temporal compounding image) acquisition module, be by by all of by (4) window that step obtains is combined Nakagami image addition and averages and obtain time combination picture Img_T；

(6) to the time combination picture Img obtained_TApplying one-dimensional Gauss polynomial to carry out post processing, the image obtained is for monitoring Radio-frequency (RF) ablation.

2. window modulation combines a Gauss polynomial matching monitoring radio-frequency (RF) ablation method, it is characterized in that, comprises the steps:

(1) obtain M frame ultrasonic signal data by ultrasonic imaging system, and to arrange the first frame data be j=1；

(2) changed by Hilbert, take the absolute value of conversion signal, thus obtain the envelope diagram of the radiofrequency signal of ablation tissue Picture；

(3) utilize N number of window to form Nakagami image: to assume that (x, y) is Nakagami image to W, and W_j1(x,y),W_j2(x, y),W_j3(x, y) ..., and W_jN(x y) represents jth Frame window length of side Nakagami image from 1 transducing increasing to N times The pulse length of device, W represents the Nakagami image of jth Frame, and increases to N times along with window width from 1 Transducer pulse length, then, the window of jth frame is combined Nakagami image W_jm(x y) can pass through formula (1) and obtain.

(4) j increases to M from 2, repeats step (2) and (3), has the most just obtained M frame window and be combined Nakagami image: W_1m(x, y),W_2m(x,y),W_3m(x, y) ..., and W_Mm(x,y)；

(5) time combination picture Img_T(temporal compounding image) is by being obtained all of by (4) step Window is combined Nakagami image addition averaging and obtains；

(6) to the time combination picture Img obtained_TApplying one-dimensional Gauss polynomial to carry out post processing, the image obtained is for monitoring Radio frequency melts.

3. window modulation as claimed in claim 2 combines Gauss polynomial matching monitoring radio-frequency (RF) ablation method, it is characterized in that, profit Compound Nakagami imaging (the WMC Nakagami imaging) step of Nakagami image i.e. window modulation is formed with N number of window Specifically, in Nakagami is distributed, the probability density function of scattering signals envelope R is defined as follows:

$f\left(R\right)=\frac{2m^m{R}^{2m-1}}{\mathrm{\Γ}\left(m\right){\mathrm{\Ω}}^m}\exp (-\frac{m}{\mathrm{\Ω}}{R}^2)U\left(R\right)---\left(2\right)$

Γ and U is gamma function and unit steplike function respectively, and Ω is zooming parameter, and m is oneself relevant to Nakagami distribution Defined parameters, if E represents expectation, then Ω and being obtained by following formula with m:

Ω=E (R²) (3)

$m=\frac{E^2\left(R^2\right)}{Var\left(R^2\right)}---\left(4\right)$

After obtaining zooming parameter Ω and Nakagami parameter m, Nakagami model is used to describe ultrasound wave back scattering Distribution situation.

4. window modulation as claimed in claim 2 combines Gauss polynomial matching monitoring radio-frequency (RF) ablation method, it is characterized in that, makes Whole ultrasonic signal envelope image is scanned, then using the value of calculation of square window as square by a square window The pixel of window position, length of window is 3 times of transducer pulse length can obtain preferable effect.

5. window modulation as claimed in claim 2 combines Gauss polynomial matching monitoring radio-frequency (RF) ablation method, it is characterized in that, high Specifically, Nakagami image has numerous row and columns to this fitting of a polynomial, comprises n data for any row or column Point, is expressed as coordinate form: (x_i,y_i), i=1,2,3 ..., n, wherein x_i,y_iIt is respectively i-th data point transverse and longitudinal coordinate axes Coordinate, if a stochastic variable X Gaussian distributed, and average and variance be respectively μ and σ², μ ∈ R and σ > 0, and X There is the probability density function y of continuous distribution_i:

$y_i=A\×\exp (-\frac{{(x_i-\mathrm{\μ})}^2}{2{\mathrm{\σ}}^2})---\left(5\right)$

A represents the amplitude of Gaussian function, (5) formula both sides is taken natural logrithm and obtains:

$\ln y_i=(\ln A-\frac{{\mathrm{\μ}}^2}{2{\mathrm{\σ}}^2})\×\frac{x_i\mathrm{\μ}}{{\mathrm{\σ}}^2}-\frac{x_i^2}{2{\mathrm{\σ}}^2}---\left(6\right)$

Order

Then equation (7) can represent with matrix form:

$\left(\begin{array}{c}{z}_1\\ z_2\\ \·\\ \·\\ \·\\ z_n\end{array}\right)=\left(\begin{array}{ccc}1& x_1& x_1^2\\ 1& x_2& x_2^2\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ 1& x_n& x_n^2\end{array}\right)\left(\begin{array}{c}{b}_0\\ b_1\\ b_2\end{array}\right)---\left(8\right)$

It is abbreviated as:

Z=XB (9)

Wherein matrix Z, X and B is respectively the matrix shorthand for formula (7).The least square solution of linear equation (9) is:

B=(X^TX)^-1X^TZ (10)

μ and σ of one-dimensional Gauss distribution can be obtained by the combination of formula (7) and (10).