CN109394263B

CN109394263B - A Multiscale Imaging Method of Ultrasonic Scatterer Diameter Based on Backscattering Coefficient

Info

Publication number: CN109394263B
Application number: CN201811115577.1A
Authority: CN
Inventors: 周著黄; 欧阳亚丽; 吴水才; 高宏建; 丁琪瑛
Original assignee: Beijing University of Technology
Current assignee: Beijing University of Technology
Priority date: 2018-09-25
Filing date: 2018-09-25
Publication date: 2021-06-18
Anticipated expiration: 2038-09-25
Also published as: CN109394263A

Abstract

The invention discloses a multi-scale imaging method of ultrasonic scatterer diameter based on backscattering coefficient. Slide the sliding windows of different scales on the ultrasonic radio frequency signal, calculate the ultrasonic scatterer diameter parameters in each sliding window based on the backscattering coefficient, and obtain the ultrasonic scatterer diameter parameter value matrix at each scale. The scatterer parameter value matrix is interpolated to the size of the ultrasonic radio frequency signal, and superimposed and averaged to obtain a multi-scale ultrasonic scatterer diameter parameter matrix. The ultrasonic scatterer diameter multi-scale imaging method of the present invention can be used for ultrasonic tissue characterization of biological tissues such as breast and liver.

Description

Ultrasonic scatterer diameter multi-scale imaging method based on backscattering coefficient

Technical Field

The invention belongs to the technical field of signal processing, and particularly relates to a medical ultrasonic signal processing method, in particular to a method for calculating a backscattering coefficient by using an ultrasonic backscattering signal (radio frequency signal), then calculating an ultrasonic scatterer diameter parameter and carrying out multi-scale imaging.

Background

Ultrasonic imaging is widely used in clinical diagnosis due to its characteristics of good real-time performance, low cost, no ionizing radiation and the like. An ultrasonic probe (transducer) transmits ultrasonic waves into tissue, and after the ultrasonic waves and the tissue generate a series of interactions such as scattering, reflection, diffraction and the like, the ultrasonic probe receives back scattering echoes of the tissue. The commonly used "B-mode ultrasound" (B-mode ultrasound imaging) utilizes amplitude information of a back-scattered signal (radio frequency signal) to perform imaging, but loses information such as frequency, so that diagnostic information of the ultrasound imaging is limited.

Biological soft tissue can be modeled as a series of ultrasound scatterers, i.e., a combination of tiny particles that scatter sound waves. The ultrasonic probe transmits ultrasonic waves into soft tissue and receives back scattering echoes from scatterers, so that ultrasonic radio frequency signals are also called back scattering signals. For liver and breast tissue, ultrasound scatterers include hepatocytes, breast cells (diffuse scatterers) and liver lobules, breast ducts/lobules (coherent scatterers), etc., which may directly reflect the microstructure of the tissue. The ultrasound backscatter signal implies important properties of scatterers, such as: scattering subvolume size, acoustic impedance, concentration and arrangement, etc. On the other hand, the ultrasonic scatterer characteristic parameter imaging algorithm based on the sliding window has a key problem to be solved, namely different window sizes can influence scatterer characteristic parameter images; specifically, a larger window may result in stable parameter estimation and better smoothness of the scatterometry sub-parametric image, and a smaller window may result in higher resolution of the scatterometry sub-parametric image. In order to take the advantages of a large window and a small window into consideration, the invention aims to provide an ultrasonic scatterer diameter multi-scale imaging method based on a backscattering coefficient.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provides an ultrasonic scatterer diameter multi-scale imaging method based on a backscattering coefficient, which can be used for ultrasonic tissue characterization of biological tissues such as mammary gland, liver and the like.

In order to achieve the purpose, the invention adopts the following technical scheme:

a backscattering coefficient-based ultrasonic scatterer diameter multi-scale imaging method comprises the following steps:

(1) sliding a rectangular window over an ultrasonic RF signalThe ultrasonic radio-frequency signal is M multiplied by N, namely M scanning lines, each scanning line comprises N sampling points, and the distance between every two adjacent scanning lines is Int_latMeter, the distance between two adjacent sampling points is Int_axiAnd (4) rice. The size of the rectangular window (i.e. sliding window) is M_w×N_wDenotes M_wScanning line N_wAnd (4) sampling points. The sliding window has sliding steps of delta in the X direction (scanning line direction) and Z direction (sampling point direction)_XAnd delta_ZTo obtain σ in total_X×σ_ZA sliding window, δ_XAnd delta_ZDenotes the distance between two adjacent sliding windows in the X-direction and Z-direction, respectively, 0<δ_X≤M_w，0<δ_Z≤N_w，σ_X＝<(M－M_w)/δ_X>，σ_Z＝<(N－N_w)/δ_Z>Wherein<>Indicating rounding up.

(2) For the sigma_X×σ_ZEach size is M_w×N_wRespectively calculating the ultrasonic scatterer diameter parameter value in each sliding window to obtain sigma_X×σ_ZValue of the ultrasonic scatterer diameter parameter, i.e. the size σ_X×σ_ZTwo-dimensional matrix SD of ultrasonic scatterer diameter parameters_orig. The ultrasonic scatterer diameter parameter calculation in the sliding window comprises the following steps:

(2.1) firstly calculating the back scattering coefficient BSC of the tissue to be measured in the sliding window_s：

Where ω denotes angular frequency, z denotes ultrasonic scanning depth, S_s(omega) is the power spectrum of the tissue to be examined, S_r(omega) is the power spectrum of the reference phantom, the backscattering coefficient BSC of the reference phantom_rAnd attenuation coefficient alpha_r(ω) is known. Attenuation coefficient alpha of tissue to be measured_s(ω) was calculated using a spectral difference method based on a reference phantom:

wherein γ (ω) represents ln [ S ]_s(ω)/S_r(ω)]The slope of the line is fitted with the ultrasound scan depth z. The power spectrum S is calculated in the following manner:

wherein p is_n(t) denotes the RF signal of the nth scan line in the sliding window, FT denotes the Fourier transform, N_swIs the number of scan lines within the sliding window.

(2.2) the scatterer diameter SD in the sliding window passes through the back scattering coefficient BSC of the tissue to be measured_sTheoretical backscattering coefficient BSC of spherical Gaussian scatterer_tThe least squares fit between yields:

the above formula represents

Minimum SD, where ω_minAnd ω_maxRespectively representing the minimum value and the maximum value of omega, n_ωIndicates the number of values of ω,. psi (ω, SD) and

calculated by the following formula:

ψ(ω,SD)＝10ln[BSC_s(ω)]-10ln[BSC_t(ω)]wherein BSC_s(omega) and BSC_t(omega) respectively represents the back scattering coefficient of the tissue to be detected under the angular frequency omega and the theoretical back scattering coefficient of the spherical Gaussian scatterer,

(3) for said size σ_X×σ_ZTwo-dimensional of the ultrasonic scatterer diameter parameterMatrix SD_origAnd interpolating the ultrasonic scatterer diameter parameter into an ultrasonic scatterer diameter parameter two-dimensional matrix SDM with the size of M multiplied by N.

(4) Size M of sliding window_wAnd N_wAre respectively set as M_w＝<ε×Len/Int_lat>，N_w＝<ε×Len/Int_axi>Where Len is the length of the ultrasonic transmit pulse, Len is in meters,<>meaning rounding up, epsilon in turn takes the values 1,2₁In which epsilon₁Is a positive integer more than or equal to 2. For each epsilon value, calculating an ultrasonic scatterer diameter parameter two-dimensional matrix SDM under each epsilon value, namely each scale by respectively using the steps 1 to 3_ε。

(5) Calculating two-dimensional matrix SDM (software description model) of multi-scale ultrasonic scatterer diameter parameters_mul：

(6) For multi-scale ultrasonic scatterer diameter parameter two-dimensional matrix SDM_mulAnd performing color mapping to obtain an ultrasonic scattering sub-diameter multi-scale image.

The invention has the advantages of

The ultrasonic scatterer diameter multi-scale imaging method based on the backscattering coefficient has the following beneficial effects:

1. the invention provides an ultrasonic scatterer diameter multi-scale imaging method based on a backscattering coefficient, which effectively solves a key problem in an ultrasonic scatterer characteristic parameter imaging algorithm based on a sliding window: the big window and the small window have advantages and disadvantages respectively, and the advantages of the big window and the small window cannot be considered at the same time. The ultrasonic scatterer diameter multi-scale imaging method based on the backscattering coefficient effectively solves the contradiction between the large window and the small window, and can effectively give consideration to the advantages of the large window and the small window.

2. The ultrasonic scatterer diameter multi-scale imaging method based on the backscattering coefficient can effectively make up for the defects of the traditional B-mode ultrasound, namely ultrasonic scatterer diameter information which cannot be provided by the traditional B-mode ultrasound is provided, and the scatterer diameter information has the characteristic of multi-scale, so that the ultrasonic tissue characterization and disease diagnosis of tissues such as liver, mammary gland and the like are facilitated.

Drawings

FIG. 1: a flow chart of the method of the invention;

FIG. 2: ultrasonic radio frequency signals and a sliding window schematic diagram.

Detailed Description

The invention discloses an ultrasonic scatterer diameter multi-scale imaging method based on a backscattering coefficient, which is a method for calculating the backscattering coefficient based on an ultrasonic backscattering signal (radio frequency signal) of a tissue to be detected, then calculating an ultrasonic scatterer diameter parameter and calculating a multi-scale image of the ultrasonic scatterer diameter parameter.

Without loss of generality, the ultrasonic radio frequency signal is composed of M scanning lines, each scanning line comprises N sampling points, and the distance between every two adjacent scanning lines is Int_lat(in meters) and the distance between two adjacent sampling points is Int_axi(unit is meter), the ultrasonic radio frequency signal is a two-dimensional matrix with the size of M multiplied by N; let Len (in meters) be the length of the ultrasound transmit pulse. FIG. 1 is a flow chart of the method of the present invention, which mainly comprises the following steps:

(1) sliding a rectangular window on the ultrasonic radio frequency signal, as shown in fig. 2, wherein the width and height of the sliding window are both epsilon × Len (unit is meter), and epsilon is a positive integer; the size of the sliding window expressed by the number of scanning lines and the number of sampling points is M_w×N_wDenotes M_wScanning line N_wA sampling point where M_w＝<ε×Len/Int_lat>，N_w＝<ε×Len/Int_axi>，<>Indicating rounding up. Let the step length of sliding window in X and Z directions (FIG. 2) be delta_XAnd delta_ZTo obtain σ in total_X×σ_ZA sliding window, δ_XAnd delta_ZRespectively representing the distance between two adjacent sliding windows in the X and Z directions, satisfying the following condition:

0<δ_X≤M_w，0<δ_Z≤N_w。

in this embodiment, δ_X＝<0.5×M_w>，δ_Z＝<0.5×N_w>。σ_XAnd σ_ZThe calculation method comprises the following steps:

σ_X＝<(M－M_w)/δ_X>，σ_Z＝<(N－N_w)/δ_Z>。

(2) for σ_X×σ_ZEach size is M_w×N_wRespectively calculating the ultrasonic scatterer diameter parameter value in each sliding window to obtain sigma_X×σ_ZValue of the ultrasonic scatterer diameter parameter, i.e. the size σ_X×σ_ZTwo-dimensional matrix SD of ultrasonic scatterer diameter parameters_orig. The method for calculating the diameter parameter of the ultrasonic scatterer in the sliding window comprises the following steps:

firstly, calculating the back scattering coefficient BSC of the tissue to be measured in the sliding window_s：

Where ω denotes angular frequency, z denotes ultrasonic scanning depth, S_s(omega) is the power spectrum of the tissue to be examined, S_r(omega) is the power spectrum of the reference phantom, the backscattering coefficient BSC of the reference phantom_rAnd attenuation coefficient alpha_r(ω) is known. Attenuation coefficient alpha of tissue to be measured_sThe (ω) can be calculated by a spectral shift (spectral shift) method or a spectral difference (spectral difference) method, and in this example, a reference phantom-based spectral difference method is adopted:

wherein γ (ω) represents ln [ S ]_s(ω)/S_r(ω)]The slope of the line is fitted with the ultrasound scan depth z. The medium of the reference phantom may be of any type, but it is required that its scattering type is incoherent. In addition, the ultrasonic imaging system and imaging parameters adopted by the tissue to be detected and the reference phantom are required to be consistent. The power spectrum S is calculated in the manner：

The scatterer diameter SD in the sliding window passes through the back scattering coefficient BSC of the tissue to be detected_sTheoretical backscattering coefficient BSC of spherical Gaussian scatterer_tThe least squares fit between yields:

the above formula represents

calculated by the following formula:

theoretical backscattering coefficient BSC of spherical Gaussian scatterer_tCalculated by the method reported in the following documents: faran Jr J.Sound scattering by soluble cyclines and spheres.journal of the environmental Society of America,1951,23(4): 405-.

(3) For said size σ_X×σ_ZUltrasonic scatterer diameter parameter twoDimension matrix SD_origAnd interpolating the ultrasonic scatterer diameter parameter into an ultrasonic scatterer diameter parameter two-dimensional matrix SDM with the size of M multiplied by N. The interpolation can adopt methods such as nearest neighbor interpolation, bilinear interpolation, cubic spline interpolation and the like. In this embodiment, cubic spline interpolation is adopted.

(4) For epsilon, values 1,2, epsilon are taken in sequence₁In which epsilon₁Is a positive integer more than or equal to 2, and the two-dimensional matrix SDM of the ultrasonic scatterer diameter parameter under each epsilon value, namely each scale is respectively calculated by utilizing the steps 1 to 3_ε. In this example,. epsilon₁And 10 is taken.

(6) For multi-scale ultrasonic scatterer diameter parameter two-dimensional matrix SDM_mulAnd performing color mapping to obtain an ultrasonic scattering sub-diameter multi-scale image. The color mapping can adopt methods such as Jet, Hot, Spring and the like in Matlab software. In this embodiment, Jet color mapping is employed.

The ultrasonic scatterer diameter multi-scale imaging method is a process for calculating ultrasonic scatterer diameter parameters and obtaining ultrasonic scatterer diameter multi-scale images. The ultrasonic scatterer diameter multi-scale imaging method can be used for ultrasonic tissue characterization of biological tissues such as mammary gland, liver and the like.

Claims

1. a method for multi-scale imaging of ultrasonic scatterer diameter based on backscattering coefficient, is characterized in that, comprises the following steps:

Step 1. Slide the rectangular window on the ultrasonic radio frequency signal of the tissue to be tested. The size of the ultrasonic radio frequency signal is M×N, that is, M scan lines, each scan line contains N sampling points, and two adjacent scan lines are scanned. The distance between lines is Int _lat meters, and the distance between two adjacent sampling points is Int _axi meters; the size of the rectangular window, that is, the sliding window, is M _w ×N _w , which means M _w scanning lines × N _w sampling points, M _w =<ε×Len/Int _lat >, N _w =<ε×Len/Int _axi >, where Len is the length of the ultrasonic emission pulse, the unit of Len is meters, and <> means rounded up, ε is a positive integer; the sliding steps of the sliding window in the X direction, that is, the scanning line direction and the Z direction, that is, the sampling point direction, are δ _X and δ _Z respectively, and a total of σ _X ×σ _Z sliding windows are obtained, δ _X and δ _Z respectively represent the distance between two adjacent sliding windows in the X and Z directions, 0<δ _X ≤M _w , 0<δ _Z ≤N _w , σ _X =<(M-M _w )/δ _X >, σ _Z =<(N-N _w )/δ _Z >, where <> represents rounding up;

Step 2. For the σ _X ×σ _Z sliding windows of size M _w × N _w , calculate the parameter value of the ultrasonic scatterer diameter in each sliding window respectively, and obtain σ _X ×σ _Z ultrasonic scatterer diameters in total. parameter value, that is, a two-dimensional matrix SD _orig of ultrasonic scatterer diameter parameters of size σ _X ×σ _Z ;

The calculation of the ultrasonic scatterer diameter parameter in the sliding window includes the following steps 2.1 to 2.2:

Step 2.1. Calculate the backscattering coefficient BSC _s of the tissue to be tested in the sliding window:

where ω is the angular frequency, z is the ultrasound scanning depth, S _s (ω) is the power spectrum of the tissue to be measured, S _r (ω) is the power spectrum of the reference phantom, the backscattering coefficient BSC _r and attenuation of the reference phantom The coefficient α _r (ω) is known; the attenuation coefficient α _s (ω) of the tissue under test is calculated using the spectral difference method based on the reference phantom:

Among them, γ(ω) represents the slope of the straight line fitted by ln[S _s (ω)/S _r (ω)] with the ultrasonic scanning depth z; the calculation method of the power spectrum S is:

Among them, p _n (t) represents the radio frequency signal of the nth scan line in the sliding window, FT represents the Fourier transform, and N _sw is the number of scan lines in the sliding window;

Step 2.2. The scatterer diameter SD in the sliding window is obtained by least squares fitting between the backscattering coefficient BSC _s of the tissue to be tested and the theoretical backscattering coefficient BSC _t of the spherical Gaussian scatterer:

The above formula expresses

SD with the smallest value, where ω _min and ω _max represent the minimum and maximum values of ω, respectively, n _ω represents the number of ω values, ψ(ω, SD) and

Calculated by the following formula:

ψ(ω,SD)=10ln[BSC _s (ω)]-10ln[BSC _t (ω)], where BSC _s (ω) and BSC _t (ω) represent the backscattering coefficient of the tissue to be measured at the angular frequency ω, respectively and the theoretical backscattering coefficient of spherical Gaussian scatterers,

Step 3. For the two-dimensional matrix SD _orig of ultrasonic scatterer diameter parameters, interpolate it into a two-dimensional matrix SDM of ultrasonic scatterer diameter parameters of size M×N;

Step 4. For ε, take the values 1, 2,...,ε ₁ in turn, where ε ₁ is a positive integer ≥ 2, and use the above steps 1 to 3 to calculate the ultrasound at each ε value, that is, at each scale. Two-dimensional matrix SDM _ε of scatterer diameter parameters;

Step 5. Calculate the two-dimensional matrix SDM _mul of multi-scale ultrasonic scatterer diameter parameters:

Step 6: Perform color mapping on the multi-scale ultrasonic scattering sub-diameter parameter two-dimensional matrix SDM _mul to obtain a multi-scale ultrasonic scattering sub-diameter image.