CN104990993A - Ultrasound slowness difference tomography algorithm for weak scattering mediums - Google Patents

Abstract

The present invention relates to an ultrasound slowness difference tomography algorithm for weak scattering mediums. According to the method, the slowness difference value on the sound wave path in the presence or absence of a measured object is used to carry out algebraic iteration so as to obtain a reconstructed image of a weak scattering medium; an annular array is immersed in a coupling medium, and in the absence of the measured object, a single ultrasonic transducer is used to transmit pulse sound waves while a transducer positioned on the opposite side of the transmitting transducer is used to receive the sound waves so as to complete one data acquisition; the transmitting transducers are sequentially replaced and the same method is used to complete multiple data acquisitions until all the transducers in the annular array transmit the sound waves; all waveform data are recorded and are adopted as background signals; a weak scattering material having sight refraction coefficient difference from the background medium is immersed in an array coupling agent, and the same way as the background data acquisition is used to acquire a detection signal; and after all detection signals are acquired, the image reconstruction is completed in two steps through comparison with the background signals so as to finally obtain the tomographic image of the measured weak scattering medium, wherein the two steps comprise calculating the interval transit time difference and the slowness difference between the background sound signals and the detection signals, and carrying out algebraic reconstruction by using the slowness difference as the iterative object.

Description

A kind of ultrasonic slowness difference CT calculating method for weak scattering medium

One, technical field

The present invention relates to a kind of ultrasonic slowness difference CT calculating method being applicable to weak scattering medium, this algorithm will record the slowness in acoustic propagation path when there is no measured object, then the slowness in same paths is recorded when there being measured object, both are done difference, algebraically iteration is carried out, the tomographic reconstruction images of measured object under acquisition weak scattering regime by the slowness difference obtained.So-called weak scattering regime, refers to that the sound deflection coefficient difference between measured object and background media is little, and sound wave is smaller in the direction deflection by occurring during interface between measured object and background media.

Two, background technology

Ultrasonic ct technology by from different perspectives launch sound wave irradiate measured object, and receive and observed matter interaction after scattering wave, rebuild the structure of measured object through certain algorithm.This technology has been applied in medical tissue detection, and main realization rate is rebuild based on the ultrasound computed tomography of acoustic wave energy decay.

The periodical literature that retrieved by middle National IP Network " based on the ultrasound tomography algorithm [J] of the minimum residual error of broad sense. acoustic technique " propose the innovatory algorithm that the minimum residual sum mean filter of broad sense combines; Periodical literature " Iterative reconstruction algorithm for ultrasonic reflection mode diffraction tomography [J]. acoustic technique " Non Uniform Fourier conversion and process of iteration are combined, establish reflection-type ultrasonic diffraction CT calculating method; Periodical literature " sound diffraction tomography progress [J]. acoustic technique " mention some scholars under weak scattering regime by measuring intensity in multiple plane to replace phase place, the object section reconstruction of Diffraction CT is only carried out according to intensity; Periodical literature " iterative approximation [J] of ultrasound tomography. acoustic technique " adopt the diffuse sound under Born approximate condition to have a meeting, an audience, etc. well under one's control to rebuild the image of measured object; Periodical literature " tomography ultrasonoscopy is in conjunction with the ultrasonic clinical value to diagnosis on ovarian tumors of three-dimensional energy [J]. Qiqihar Medical College's journal " discuss the ultrasonic application to diagnosis on ovarian tumors aspect of three-dimensional energy.Related content in above-mentioned document is all rebuild means with energy, and the estimation that the energy attenuation on measured object interface can make the mistake to the pad value on the acoustic propagation path of measured object own, affect Detection job, the reconstruction figure metamorphopsic distortion of measured object may be made.

The periodical literature that retrieved by middle National IP Network " research and implementation [J] of ultrasound tomography detection system. electronic design engineering " have employed Joint iteration reconstruction technique (SIRT), the reconstruction slowness matrix of measured object is obtained based on linear interpolation ray-tracing algorithm; Periodical literature " the ultrasound computed tomography reconstruction algorithm based on Joint iteration reconstruction technique " is deduced SIRT algorithm based on least squared criterion, have also been obtained measured object and rebuilds slowness matrix.The algorithm completed when utilizing sound has been related in above-mentioned document, but adopt SIRT algorithm to there is the slow problem of iteration speed, and the slowness projection only utilizing measured object to exist in situation carries out iteration, insensitive to the differentiation between measured object and background area, the pixel artifact that measured object is rebuild around figure is serious.

The Chinese patent that retrieved by the patent retrieval of national patent board web and service system " utilize frequency spectrum extrapolation technique to improve the ultrasound computed tomography image rebuilding method [P] of image quality. publication number: CN102651126A " and Chinese patent " based on the ultrasound tomography method [P] of Nonuniform fast Fourier transform. publication number: CN102652680A " have employed spatial domain to frequency domain, image reconstruction is completed again by frequency domain to the transform method of spatial domain, the frequency spectrum Extrapolation method used is carried out in domain space, the slight error that this conversion produces can cause very large error when turning back to spatial domain imaging.Chinese patent " concrete ultrasound tomography algorithm [P]. publication number: CN1908651 " method that employs tower ART algorithm amendment slowness value rebuilds concrete inner structure, also can exist this problem insensitive of the differentiation between measured object and background area; Chinese patent " method of biological organism optical and ultrasonic collection and tomography and device [P] thereof. publication number: CN1555764 " mention and the ultrasonic signal collected is rebuild measured object image in the mode of linear projection, but do not provide concrete CT calculating method process.

This patent adopts new parameter from different being of above-mentioned periodical literature and patent---the algebraically iteration that slowness difference is carried out, the impact of measured object interface on energy attenuation can not be considered, compare energy method, image reconstruction more accurately can be completed, and when using the brand-new couplant of the unknown velocity of sound, the image of measured object still correctly can be rebuild

Three, summary of the invention

The object of this invention is to provide a kind of ultrasonic slowness difference CT calculating method for weak scattering medium, to solve the image fault of energy method ultrasonic ct algorithm on measured object structural remodeling.

For reaching the testing goal to weak scattering medium, the ultrasonic slowness difference CT calculating method that the present invention proposes comprises: slowness difference when immersion annular array data acquisition and employing exist with or without measured object on sound wave path carries out the ultrasonic ct algorithm of iterative approximation.

Described immersion annular array data acquisition is: the immersion annular array be made up of multiple ultrasonic transducer is not having measured object, when only having couplant, adopt single ultrasonic transducer transponder pulse sound wave, use the receive MUT sound wave of transmitting transducer offside simultaneously, complete a data acquisition.Change transmitting transducer in turn afterwards and adopt identical method to complete repeatedly data acquisition, until transducers all in annular array all launched sound wave.This only have the data collected during couplant to be called background signal.Be dipped in by measured object in the couplant of annular array, the transducer adopting the mode identical with background extraction data to travel through in turn in array carries out ultrasonic transmitting-receiving and data acquisition, and the data at this moment obtained are called detection signal.So far, data acquisition completes.

The embodiment of described slowness difference iteration ultrasonic ct algorithm is: the background signal obtain data acquisition and detection signal carry out calculating and contrasting, observe in process transmitting-receiving transducer between relation one to one, obtain slowness difference as iteration object by the acoustic propagation path harmony time difference, the mode of last resort algebraically iteration completes the image reconstruction of weak scattering medium measured object about acoustic properties.

The ultrasonic slowness difference CT calculating method that the present invention proposes, solve the measured object that the ultrasonic ct algorithm based on energy method causes due to interfacial energy loss well and rebuild image fault, for realizing radiationless weak scattering medium, to detect (such as medical soft tissue detection) significant.

Four, accompanying drawing explanation

The transmitted wave path schematic diagram of Fig. 1 annular array slowness difference iterative algorithm

The sound time difference schematic diagram of Fig. 2 background signal and detection signal

Fig. 3 weak scattering medium that 36 array element annular arrays complete rebuilds image

Five, embodiment

Annular array is dipped in couplant, when there is not measured object, adopts single ultrasonic transducer transponder pulse sound wave, using the receive MUT sound wave of transmitting transducer offside simultaneously, completing a data acquisition.Change transmitting transducer in turn afterwards and adopt identical method to complete repeatedly data acquisition, until transducers all in annular array all launched sound wave.Record all Wave datas, as a setting signal.

The weak scattering material be more or less the same with background media refraction coefficient being dipped in array couples agent, adopting the mode identical with gathering background signal to obtain detection signal.Accompanying drawing 1 is shown in beam path contrast without acoustic beam during measured object and when having measured object, dotted line in figure between transmitting transducer T1 to receiving transducer T22 and between T1 to receiving transducer T16 represents beam path when not having measured object, and solid line then indicates beam path during measured object.

After obtaining all detection signals, by the contrast with background signal, complete image reconstruction.Ultrasonic slowness difference iteration CT calculating method for rebuilding has two steps, is that sound time difference between background signal and detection signal and slowness difference are asked for respectively, and adopts slowness difference as the algebraic reconstruction of iteration object.

With regard to a certain transmitting-receiving transducer to, the time that sound wave is propagated in background media is:

$t_b=\frac{d_b}{c_b}$

D _bthe distance between two transducers, c _bit is the velocity of sound of background media.When measured object is placed in array, transducer between the ultrasonic transit time can be expressed as:

$t_p=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}\frac{l_i}{c_i}+\frac{d_b-\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{l}_i}{c_b}$

L _irepresent the distance of sound wave process in i-th piece of measured object, c _irepresent the velocity of sound of i-th piece of measured object.I=1 ~ N represents this bundle ultrasound wave altogether through the measured object of N number of different velocity of sound.

Two formulas on simultaneous, obtain the sound time difference (see accompanying drawing 2) detecting ripple and background wave:

$\mathrm{\Δt}=t_p-t_b=\underset{i=1}{\overset{N}{\mathrm{\Σ}}}\frac{l_i}{c_i}-\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{l}_i}{c_b}$

L on the right of above formula _iand c _iall unknown quantity, need by transducer between permutation and combination, obtain the solving equations of sufficient amount.The sound time difference is on average just obtained slowness difference in sound wave path:

$\mathrm{\Δs}=\frac{\mathrm{\Δt}}{d_b}=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\frac{l_i}{c_i}-\frac{\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{l}_i}{c_b}}{d_b}$

Finally utilize the slowness difference obtained to be brought in the iterative process of algebraic reconstruction technique (Algebraic Reconstruction Technique, ART) to calculate:

${\stackrel{\→}{x}}^{(k+1)}=\left\{\begin{array}{cc}{\stackrel{\→}{x}}^{\left(k\right)}& {\stackrel{\→}{r}}_{\mathrm{ik}}^T{\stackrel{\→}{x}}^{\left(k\right)}\≤{\mathrm{\Δs}}_{\mathrm{ik}}\\ {\stackrel{\→}{x}}^{\left(k\right)}+{\mathrm{\λ}}^{\left(k\right)}\frac{{\mathrm{\Δs}}_{\mathrm{ik}}-{\stackrel{\→}{r}}_{\mathrm{ik}}^T{\stackrel{\→}{x}}^{\left(k\right)}}{{\left|\right|{\stackrel{\→}{r}}_{\mathrm{ik}}\left|\right|}^2}{\stackrel{\→}{r}}_{\mathrm{ik}},& {\stackrel{\→}{r}}_{\mathrm{ik}}^T{\stackrel{\→}{x}}^{\left(k\right)}\≥{\mathrm{\Δs}}_{\mathrm{ik}}\end{array}\right.$

K in formula refers to iterations, and λ is called relaxation factor, and its value, usually between 0 ~ 1, changes less between less expression twice iterative value, also more stable, but convergence is also slower.R _ijrepresent whether i ray crosses j pixel, is equal to weight coefficient, then represent vector 2 norms.The final pixel that above formula obtains be exactly that slowness difference rebuilds image.

Six, application example

In order to the correctness of verification algorithm, emulation mode is adopted to obtain data.In this verification model, use water to do couplant, by slowness difference algorithm, image reconstruction is carried out to the liver material be immersed in water and fat material two kinds of weak scattering media.

The material properties of the measured object 1 (rectangle) in accompanying drawing 1 is set as liver, and density is 1079kg/m ³, the velocity of sound is 1560m/s; The material properties of measured object 2 (ellipse) is set as fat, and density is 911kg/m ³, the velocity of sound is 1470m/s.The quantity of annular array array element is 36, and ultrasound wave centre frequency is 5MHz, and acoustic beam spread angle is 60 degree.

As shown in Figure 1 the annular array of 36 array elements is dipped in water, set up rectangular coordinate system using array center as initial point, the radius of setting annular array is R=40mm, and the center point coordinate of measured object 1 is (0,8mm), longly and wide 15mm and 8mm is respectively; The center point coordinate of measured object 2 is (0 ,-8mm), and oval major semi-axis and minor semi-axis length are respectively 15mm and 8mm.Utilize the array element in array excite and receive ultrasound wave successively, the reconstruction image adopting slowness difference algorithm to detect to obtain as shown in Figure 3.

Claims (10)

1. the ultrasonic slowness difference CT calculating method for weak scattering medium, it is characterized in that: adopt immersion annular array to obtain ultrasound data, need to gather background signal when not having measured object and detection signal when having measured object to exist, the cross section being completed weak scattering medium measured object by slowness difference iterative algorithm is detected.

2. the ultrasonic slowness difference CT calculating method described in claim 1 is characterised in that: complete fault detection for the weak scattering medium be more or less the same with background media refraction coefficient.

3. its data acquisition of immersion annular array described in claim 1 is characterised in that: do not having measured object and having measured object to deposit in case respectively, obtaining array signal.

4. its pulsating wave of the immersion annular array transmitting-receiving mode described in claim 1 is characterised in that: all receive transmitted wave respectively by single transducer excitation ultrasound ripple, multiple transducer at every turn, and obtain all transmission waveforms along 360 degree of direction order replacing transmitting transducers around measured object, thus the background wave graphic data obtained needed for reconstruction and detection waveform data.

5. its array element of immersion annular array described in claim 4 is rectangle array element, and the ratio of width to height is less than 1/5.

6. its array element of immersion annular array described in claim 4 arranges in turn along local width direction, and interval is less than array element width.

7. its array element acoustic beam spread angle of the immersion annular array described in claim 4 is greater than 60 degree.

8. the mode that the slowness difference iterative algorithm described in claim 1 adopts background signal and detection signal to contrast obtains slowness difference as iteration object.

9. the slowness difference iterative algorithm described in claim 1 carries out iteratively faster based on algebraic reconstruction technique (ART).

10. the slowness difference iterative algorithm described in claim 1, when using the brand-new couplant of the unknown velocity of sound (prerequisite need meet weak scattering relation between measured object), still correctly can rebuild the image of measured object.