CN111528912A - Ultrasonic elastography method, device and system - Google Patents
Ultrasonic elastography method, device and system Download PDFInfo
- Publication number
- CN111528912A CN111528912A CN202010447279.3A CN202010447279A CN111528912A CN 111528912 A CN111528912 A CN 111528912A CN 202010447279 A CN202010447279 A CN 202010447279A CN 111528912 A CN111528912 A CN 111528912A
- Authority
- CN
- China
- Prior art keywords
- data
- frame
- elastography
- matching
- calculating
- 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.)
- Pending
Links
- 238000002091 elastography Methods 0.000 title claims abstract description 85
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000003384 imaging method Methods 0.000 claims abstract description 61
- 238000012216 screening Methods 0.000 claims abstract description 49
- 238000011156 evaluation Methods 0.000 claims abstract description 21
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 238000006073 displacement reaction Methods 0.000 claims description 32
- 238000003825 pressing Methods 0.000 claims description 29
- 230000010363 phase shift Effects 0.000 claims description 20
- 238000005070 sampling Methods 0.000 claims description 18
- 238000002113 ultrasound elastography Methods 0.000 claims description 17
- 239000000523 sample Substances 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 9
- 238000002604 ultrasonography Methods 0.000 claims description 8
- 230000006835 compression Effects 0.000 claims description 5
- 238000007906 compression Methods 0.000 claims description 5
- 238000012285 ultrasound imaging Methods 0.000 claims description 5
- 230000004323 axial length Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 9
- 230000002159 abnormal effect Effects 0.000 abstract description 8
- 238000004364 calculation method Methods 0.000 description 32
- 238000001914 filtration Methods 0.000 description 17
- 230000004927 fusion Effects 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000007726 management method Methods 0.000 description 4
- 230000007704 transition Effects 0.000 description 4
- 239000000969 carrier Substances 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000004393 prognosis Methods 0.000 description 2
- 208000003028 Stuttering Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000003139 buffering effect Effects 0.000 description 1
- 229910010293 ceramic material Inorganic materials 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 239000000126 substance Substances 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
- A61B8/48—Diagnostic techniques
- A61B8/485—Diagnostic techniques involving measuring strain or elastic properties
-
- 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
-
- 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
- A61B8/5269—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving detection or reduction of artifacts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Heart & Thoracic Surgery (AREA)
- Surgery (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Biomedical Technology (AREA)
- Veterinary Medicine (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention relates to the technical field of ultrasonic imaging, and discloses an ultrasonic elastography method, which comprises the following steps: acquiring elastography original data, performing data preprocessing on the elastography original data to obtain IQ data, and caching the IQ data; extracting partial IQ data from the cached multi-frame IQ data to be used as an IQ data set to be screened; combining IQ data in an IQ data set to be screened pairwise to obtain a plurality of IQ data sets; calculating a structural similarity evaluation coefficient of each IQ data set, and taking the IQ data set with the structural similarity evaluation coefficient within a set threshold range as a primary screening IQ data set; calculating the imaging quality parameters of each IQ data set in the primary screened IQ data set, and screening out the optimal combination from the primary screened IQ data set according to the imaging quality parameters; and performing elastography according to the optimal combination to obtain an elastogram. The ultrasonic elastography method provided by the invention has the technical effects of stable elastography, no abnormal frame and low resource occupancy rate.
Description
Technical Field
The invention relates to the technical field of ultrasonic imaging, in particular to an ultrasonic elastography method, an ultrasonic elastography device, an ultrasonic elastography system and a computer storage medium.
Background
Ultrasonic elastography (ultrasound) is a new ultrasonic imaging technique that has developed relatively rapidly in recent years, and was proposed by j. Ultrasonic elastography provides richer functional information for clinical disease diagnosis, so that the method is widely concerned by the industry and related research institutions and has high commercial value.
The ultrasonic compression elastography algorithm is realized on the basis that displacement exists between two adjacent frames of data under the same stress effect, and the strain of tissues generated by different elastic sizes is different. The elastic size (strain size) of the tissue is reflected by different colors, and elastic imaging is realized. The ultrasound compression elastography technology requires manual operation, and the correlation between the imaging quality and the operation method is large. High-quality imaging is premised on continuity and stability of pressing force, but absolute stability of the pressing force degree is difficult to achieve in the actual operation process, and it is more difficult to ensure that the stress of each frame of acquired image is fixed and unchanged. The existence of stress differences between adjacent frame data in the acquired data may cause a large color change of the calculated elastic image (strain image). In addition, the variation of stress magnitude during operation may cause decorrelation between adjacent frames of acquired data, resulting in elastic image anomalies. These large color changes and abnormal results of the elastic image can cause the phenomenon of instability of the elastic image in the real-time display process, which affects clinical judgment.
At present to the unstable problem of elasticity imaging, the general way is, after elasticity imaging is accomplished, obtain the fusion result of the mean value of meeting an emergency and the matching parameter of elasticity image, the mean value of meeting an emergency and the matching parameter fusion result of matching are the formation of image quality parameter of elasticity image, whether select to output current elasticity image according to the fusion result of mean value of meeting an emergency and matching parameter, promptly: and screening out the elastic images with abnormal imaging, and outputting the elastic images meeting the quality requirement, thereby improving the quality of the elastic imaging. However, the elastic image is calculated according to the original data of the elastic imaging, which needs to be realized by an elastic imaging algorithm, the elastic imaging algorithm is an algorithm with a large calculation amount, and the method for judging whether to start the elastic image or not after the elastic image is calculated has the problems of low efficiency and large hardware resource occupation loss. Particularly, when more obsolete data appear, the real-time elastic image is easy to generate the stuttering false image, the obsolete data are also the elastic results obtained through the elastic imaging algorithm, the process occupies computing resources, but the obtained abnormal imaging results are useless, and the frame rate of the elastic imaging is slowed down to a certain extent.
Disclosure of Invention
The invention aims to overcome the technical defects, provides an ultrasonic elastography method, an ultrasonic elastography device, an ultrasonic elastography system and a computer storage medium, and solves the technical problems of low efficiency and high resource occupancy rate when the quality is improved aiming at the elastography abnormal problem caused by unstable stress in the prior art.
In order to achieve the technical purpose, the technical scheme of the invention provides an ultrasonic elastography method, which comprises the following steps:
acquiring elastography original data, performing data preprocessing on the elastography original data to obtain IQ data, and caching the IQ data;
extracting partial IQ data from the cached multi-frame IQ data to be used as an IQ data set to be screened; combining IQ data in the IQ data set to be screened pairwise to obtain a plurality of IQ data sets;
calculating a structural similarity evaluation coefficient of each IQ data set, and taking the IQ data set with the structural similarity evaluation coefficient within a set threshold range as a primary screening IQ data set;
calculating the imaging quality parameters of each IQ data set in the primary screening IQ data set, and screening the optimal combination from the primary screening IQ data set according to the imaging quality parameters;
and performing elastography according to the optimal combination to obtain an elastic image.
The invention also provides an ultrasonic elastography device, which comprises a processor and a memory, wherein the memory is stored with a computer program, and the computer program is executed by the processor to realize the ultrasonic elastography method.
The invention also provides an ultrasonic elastography system, which comprises the ultrasonic elastography device, an ultrasonic probe, a B-ultrasonic imaging device and a display device;
the ultrasonic probe is used for detecting the ultrasonic elasticity original data and the B ultrasonic original data in a time-sharing manner;
the B-ultrasonic imaging device is used for carrying out B-ultrasonic imaging according to the B-ultrasonic original data to obtain a B-ultrasonic image;
the display device is used for displaying the B-ultrasonic image and the elastic image.
The invention also provides a computer storage medium having a computer program stored thereon, which, when executed by a processor, implements the ultrasound elastography method.
Compared with the prior art, the invention has the beneficial effects that: firstly, acquiring IQ data through elastic imaging original data; then, the IQ data are not directly subjected to elastography calculation, the IQ data are firstly subjected to combined screening, a primary screening IQ data set with a larger structural similarity evaluation coefficient is screened out, IQ data sets with significant differences are screened out, the stability of acquired data is ensured, an optimal combination with better imaging quality is further screened out, the elastography effect is subjected to prognosis judgment, and the imaging effect of the screened optimal combination is ensured; and finally, performing elastography calculation based on the optimal combination to obtain a stable elastogram. The invention relates to a method for preparing a high-temperature-resistant ceramic material. According to the method, before the elastic image is calculated, structural similarity analysis and judgment are carried out on the acquired data, stable display of the output elastic image is guaranteed, the phenomenon that the elastic image is obviously different due to operation method difference or improper operation method is reduced, redundant calculation is not carried out, and calculation resources are saved; the optimal combination of effective frame data for calculating the elastography is determined before calculating the elastography, so that the output elastography is ensured to have no abnormal frame, the elastography frame rate is improved, redundant calculation is not performed, and the calculation resources are saved.
Drawings
FIG. 1 is a flow chart of one embodiment of a method of ultrasound elastography provided by the present invention;
FIG. 2 is a diagram of sample block sampling locations for generating a real-time compression profile according to one embodiment of the present invention;
FIG. 3 is a schematic diagram of an embodiment of an ultrasound elastography system provided by the present invention;
reference numerals:
101. a B-mode ultrasonic imaging device; 102. an ultrasonic elastography device; 103. a display device; 104. an ultrasound probe.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example 1
As shown in fig. 1, embodiment 1 of the present invention provides an ultrasonic elastography method including the steps of:
s1, acquiring elastography original data, performing data preprocessing on the elastography original data to obtain IQ data, and caching the IQ data;
s2, extracting partial IQ data from the cached multiframe IQ data to be used as an IQ data set to be screened; combining IQ data in the IQ data set to be screened pairwise to obtain a plurality of IQ data sets;
s3, calculating the structure similarity evaluation coefficient of each IQ data set, and taking the IQ data set with the structure similarity evaluation coefficient within a set threshold range as a primary screening IQ data set;
s4, calculating the imaging quality parameters of each IQ data set in the primary screening IQ data set, and screening the optimal combination from the primary screening IQ data set according to the imaging quality parameters;
and S5, performing elastography according to the optimal combination to obtain an elastogram.
In the embodiment, IQ data are obtained through elastic imaging original data; specifically, the IQ data refers to data obtained by performing IQ modulation on the elastography original data, and the elastography original data is divided into two paths: the I path and the Q path are respectively subjected to carrier modulation, the two paths of carriers are mutually orthogonal, the two paths of mutually orthogonal carriers are carriers with the phase difference of 90 degrees, and the I path and the Q path are respectively subjected to carrier modulation and then are transmitted together to obtain IQ data, so that the improvement of the frequency spectrum utilization rate is facilitated. Then, before the elastic image is calculated, a certain amount of cached IQ data is combined and screened, a primary screening IQ data set with a larger structural similarity evaluation coefficient is screened out, IQ data sets with significant differences are screened out, the IQ data set used for calculating the elastic image is ensured to have a better strain displacement result, and the stability of acquired data is ensured; on the basis of primary screening, further screening is carried out, an optimal combination with better imaging quality is screened out, prognosis judgment is carried out on the elasticity imaging effect, and the optimal combination with the best imaging effect is guaranteed to be screened out; and finally, performing elastography calculation based on the optimal combination to obtain a stable and high-quality elastogram.
The ultrasonic elastography method provided by the embodiment avoids calculating an abnormal elastography result, and saves the calculation time for calculating the abnormal result; the obtained elastic images are stable in transition and small in difference, and the phenomenon that the elastic images are obviously different due to difference or improper operation methods is reduced; meanwhile, redundant calculation is not performed, and operation resources are saved.
Specifically, before performing the elastography, the raw elastography data needs to be signal processed and buffered, which is described in detail below.
Preferably, the data preprocessing is performed on the elastography original data to obtain IQ data, and the IQ data is cached, specifically:
sequentially carrying out band-pass filtering, quadrature demodulation, low-pass filtering, secondary sampling and low-pass filtering on the elastic imaging original data to obtain IQ data;
and caching the latest IQ data with the first set frame number in real time to obtain a cached IQ data set.
The raw elastic imaging data is subjected to band-pass filtering, quadrature demodulation, low-pass filtering, secondary sampling and low-pass filtering to obtain continuous IQ data, wherein the size of the IQ data is 384 × 800 in the embodiment, 384 represents a wiring harness, 800 represents depth, and the imaging depth is 5 cm. The probe center frequency was 7.5 MHz.
In this embodiment, the first set frame number is set to 100 frames, and the latest 100 frames of IQ data are cached in real time, including I data m _ pSrc _ I _ data _ int [100] and Q data m _ pSrc _ Q _ data _ int [100 ]; and managing the storage address of the latest 100 frames of IQ data in the first buffer by using a frame number management method. In this embodiment: the latest frame has an address of m _ nFrameNO [0] and a value of m _ nFrameNO [0] 85 in the first buffer, and the corresponding latest frame IQ data is stored in m _ pSrc _ I _ data _ int [85] and m _ pSrc _ Q _ data _ int [85 ]. The storage mode of the IQ data in the first buffer is incremental storage, that is: when imaging is just started, a first frame is stored in m _ pSrc _ I _ data _ int [0] and m _ pSrc _ Q _ data _ int [0], a second frame is stored in m _ pSrc _ I _ data _ int [1] and m _ pSrc _ Q _ data _ int [1], …, and a 100 th frame is stored in m _ pSrc _ I _ data _ int [99] and m _ pSrc _ Q _ data _ int [99 ]; and then replacing the 1 st frame by the 101 st frame, storing the replaced frames in m _ pSrc _ I _ data _ int [0] and m _ pSrc _ Q _ data _ int [0], and sequentially and circularly storing the replaced frames.
After the IQ data are cached, combined screening and elastography can be performed, however, in the embodiment, while the elastography is performed, a real-time pressing curve is generated by using the cached IQ data, and simultaneous acquisition of the elastography and the real-time pressing curve is realized. The details are as follows.
Preferably, a parallel thread is adopted, and a real-time pressing curve is generated according to the cached IQ data while performing elastography according to the cached IQ data.
In the embodiment, the real-time pressing curve is generated while the elastography calculation is performed, and the calculation of the real-time pressing curve and the elastography calculation are realized by adopting parallel threads. The parallel threads are used for realizing the data screening of the real-time pressing curve and the elastic imaging and the parallel work of the imaging calculation, so that the continuity and the stability of the elastic image are ensured, and the real-time performance of the pressing curve is also ensured. The computational generation of the real-time compression curve is specifically set forth below.
Preferably, the real-time pressing curve is generated according to the cached IQ data, specifically:
extracting the latest frame IQ data and the next-to-new frame IQ data from the cached IQ data;
setting a plurality of sampling blocks with the same depth and the same size;
sampling a plurality of pairs of corresponding data blocks from the latest frame IQ data block and the next-new frame IQ data block respectively through the sampling block to obtain a plurality of groups of data block groups;
calculating the matching coefficient of each group of the data block groups;
extracting the relative phase shift between the latest frame IQ data and the next-new frame IQ data according to the matching coefficient;
acquiring the relative displacement between the latest frame IQ data and the next-new frame IQ data according to the relative phase shift;
carrying out weighted summation on the relative displacement corresponding to each group of data block groups to obtain real-time relative displacement;
and generating the real-time pressing curve according to the real-time relative displacement.
In this embodiment, when calculating the real-time pressing curve, first, it is determined whether the first buffer updates data, if yes, the latest frame and the next new frame are extracted from the first buffer in real time, and the values of m _ nFrameNO [0] and m _ nFrameNO [1] are the addresses of the latest frame and the next new frame in the first buffer. Extracting a matching coefficient of a latest frame and a next new frame through a fixed sampling block, extracting a phase shift between the latest frame and the next new frame through the matching coefficient, acquiring a relative displacement through the phase shift, wherein the relative displacement reflects the size and the direction of stress, the relative displacement is stored in m _ pdplace [0] of a pressing curve storage variable m _ pdplace _ line [100], and corresponding displacements in the pressing curve storage variable are sequentially shifted backwards by a value to generate a real-time pressing curve.
The matching coefficients of the corresponding data block groups in the latest frame and the next-to-new frame are obtained as follows.
Preferably, the calculating the matching coefficient of each group of data block groups specifically includes:
and performing cross-correlation matching in the data block group by adopting a correlation window to obtain a cross-correlation coefficient:
R12＝∫∫DS1(t+v,x+w)*S2(t+v,x+w)*dvdw
wherein R is12Is the cross-correlation coefficient of the data block group, t is the axial coordinate of the center of the data block to be matched, x is the transverse coordinate of the center of the data block to be matched, v is the transverse length of the correlation window, w is the axial length of the correlation window, S1(t + v, x + w) is the IQ data value of the first block in the group of blocks at the (t + v, x + w) pixel point, S2(t + v, x + w) is the IQ data value of the second data block in the data block group at the (t + v, x + w) pixel point, D ═ v × w;
extracting the relative phase shift between the latest frame IQ data and the next-to-new frame IQ data according to the matching coefficient, which specifically comprises the following steps:
wherein the content of the first and second substances,
for the relative phase shift of the latest frame IQ data and the next-to-new frame IQ data, I1(i, j) is the imaginary part, Q, of the quadrature demodulated data of the first data block1(I, j) is the real part of the quadrature demodulated data of the first data block, I2(i, j) is the imaginary part, Q, of the quadrature demodulated data of the second data block2(i, j) is the real part of the quadrature demodulated data for the second data block;
obtaining the relative displacement between the latest frame IQ data and the next new frame IQ data according to the relative phase shift:
wherein D is the relative displacement of the latest frame IQ data and the next-to-new frame IQ data, Δ t is the time required for phase shift, c is the propagation speed of the ultrasonic wave in the tissue, ω is the angular velocity corresponding to the center frequency of the ultrasonic wave, and f0Is the center frequency.
Specifically, in this embodiment, three samples with the same depth and the same size are used, the positions of the sample blocks are shown in fig. 2, which are respectively denoted by P1, P2, and P3, and the units of the sizes indicated in the drawing are pixels. The matching coefficients of the matching points represented by the three sampling blocks are respectively calculated, then phase shift is extracted according to the matching coefficients, then corresponding displacement is extracted according to the phase shift, finally, weighted summation is carried out on the displacement corresponding to the three matching points, and the corresponding weight value can be but is not limited to 0.3 × D1+0.4 × D2+0.3 × D3, wherein D1, D2 and D3 respectively correspond to the displacement acquired by the three sampling blocks.
In this embodiment, the real-time pressing curve calculation is performed, meanwhile, the elastography calculation is performed, and before the elastography, the cached IQ data needs to be combined and screened to ensure the imaging quality, which is specifically described as follows.
Preferably, part of IQ data is extracted from the buffered multi-frame IQ data to be used as an IQ data set to be screened, and the IQ data in the IQ data set to be screened are combined pairwise to obtain a plurality of IQ data sets, specifically:
extracting IQ data with a second set frame number from the cached IQ data set to serve as an IQ data set to be screened;
combining the IQ data in the IQ data set to be screened pairwise to obtain
An IQ data set, wherein N2And setting the frame number for the second setting frame number.
After the IQ data caching is completed, the IQ data in the cached IQ data set needs to be combined and screened. First, IQ data of a second set frame number, which is 5 in this embodiment, is extracted from the buffered IQ data set, so that 10 IQ data sets can be extracted.
After the IQ data combination is completed, an IQ data set needs to be primarily screened, the 10 combined data are respectively subjected to structural similarity evaluation, and an IQ data set with a structural similarity evaluation coefficient between set threshold ranges is screened out, which is specifically described as follows.
Preferably, the calculating the structure similarity evaluation coefficient of each IQ data set specifically includes:
calculating envelope data of each frame of IQ data in the IQ data group:
wherein, the data is envelope data of IQ data, I is an imaginary part of orthogonal demodulation data of the IQ data, and Q is a real part of the orthogonal demodulation data of the IQ data;
calculating a structural similarity evaluation coefficient between two frames of IQ data in the IQ data group according to the envelope data:
wherein x and y are envelope data of two frames of IQ data in the same IQ data set, SSIM (x, y) is a structural similarity evaluation coefficient between two frames of IQ data in the same IQ data set, UxIs the mean, U, of the envelope data xyIs the mean, σ, of the envelope data yx 2Is the variance, σ, of the envelope data xy 2Is the variance, σ, of the envelope data yxyCovariance for envelope data x and for envelope data y, C1、C2Are all constants.
In this example, C1Is 6.5025, C2Is 58.5225. The range of SSIM (x, y) obtained in this embodiment is 0-1, when x and y are the same, the coefficient is 1, and the larger the difference between x and y is, the smaller SSIM (x, y) is. In this embodiment, the threshold range is set to be 0.8-0.95, i.e. the IQ data set with the structure similarity evaluation coefficient in the range of 0.8-0.95 is selected for the next screening step.
After the primary screening is completed, secondary screening is performed on the basis of the primary screening, as described in detail below.
Preferably, the imaging quality parameters of each IQ data set in the preliminary screening IQ data set are calculated, and an optimal combination is screened out from the preliminary screening IQ data set according to the imaging quality parameters, specifically:
the imaging quality parameters comprise matching coefficients and strain mean values, and the matching coefficients and the strain mean values of all IQ data sets in the primary screening IQ data sets are calculated;
and screening IQ data sets with matching coefficients larger than relevant threshold values, and selecting a set number of IQ data sets with the maximum strain mean value from the screened IQ data sets as the optimal combination.
The secondary screening comprises two screening links: and a cross-correlation matching link and a strain mean matching link, wherein each IQ data set in the primary screening IQ data set is subjected to sampling detection to obtain a corresponding strain mean and a corresponding matching coefficient, and the first N3 IQ data sets with the maximum strain mean in the IQ data sets with the matching coefficients larger than 0.8 are taken as the optimal combination to perform elastography calculation. Two screening processes are specifically described below.
Preferably, the calculating the matching coefficient of each IQ data set in the preliminary screening IQ data set specifically includes:
setting two frames of IQ data in the IQ data group as a reference frame and a matching frame respectively;
dividing the reference frame into a set number of reference data blocks, and dividing a retrieval frame corresponding to the data block positions in the matching frame;
calculating the cross correlation coefficient of each reference data block in the reference frame and a plurality of matched data blocks in the corresponding retrieval frame:
RA12(t,x,n,m)＝∫∫DS1(t+v,x+w)*S2(t+n+v,x+m+w)*dvdw
wherein RA is12(t, x, n, m) is a cross-correlation coefficient, v is the transverse length of a correlation window, w is the axial length of the correlation window, a plurality of matched data blocks in the retrieval frame are sequentially moved and selected according to a set step length, n is the transverse stepping length, m is the axial stepping length, t is the axial coordinate of the center of the matched data block, x is the transverse coordinate of the center of the matched data block, and S is the transverse coordinate of the center of the matched data block1(t + v, x + w) is the IQ data value of the reference block at the (t + v, x + w) pixel point, S2(t + n + v, x + m + w) is the IQ data value of the matched data block at the (t + n + v, x + m + w) pixel point, D ═ v × w;
calculating the autocorrelation coefficient of the reference frame:
RA11(t,x,0,0)＝∫∫D|S1(t+v,x+w)|2dvdw
wherein RA is11(t, x,0,0) is the autocorrelation coefficient of the reference frame;
calculating the autocorrelation coefficient of the matched frame:
RA22(t,x,n,m)＝∫∫D|S2(t+n+v,x+m+w)|2dvdw
wherein RA is22(t, x, n, m) is the autocorrelation coefficient of the matched frame;
calculating the matching coefficient of the reference frame and the matching frame according to the cross correlation coefficient corresponding to each group of matched data blocks, the autocorrelation coefficient of the reference frame and the autocorrelation coefficient of the matching frame:
wherein C (t, x, n, m) is a matching coefficient of the reference frame and the matching frame;
and calculating the average value of the matching coefficients corresponding to the multiple groups of data blocks as a final matching coefficient.
A cross-correlation matching step, wherein for each IQ data group, a cross-correlation matching method is used for matching the best frame in the matched frame, and the position with the maximum matching coefficient corresponds to the best frame; in this embodiment, the IQ data set is divided into 4 × 12 data blocks for matching calculation in order to simplify the calculation; 4 × 12 data blocks are uniformly distributed in the whole frame of IQ data, wherein 4 is in transverse distribution, and 12 is in axial distribution; retrieving retrieval frames at corresponding positions of 4 x 12 data blocks in the matched reference frame; the size of the block P in the reference frame is 5 × 73, where 5 is the lateral size and 73 is the axial size; the size of the search box S at the same position in the matching frame is 9 × 81, wherein 9 is the transverse size, and 91 is the axial size; matching and stepping are 1 pixel point, and the sizes of the data block P and the retrieval frame S are set according to requirements; and calculating the matching coefficient of the reference data block and the matching data block in the retrieval frame by adopting the formula.
And converting n and m of the best matching data block into displacement coordinates with the central position of the reference data block as an origin.
And after matching is finished, acquiring 4 × 12 optimal matching points aiming at the optimal matching data block of each reference data block, so as to acquire 4 × 12 matching coefficients, finally, averaging the 4 × 12 matching coefficients, and screening an IQ data group with the matching coefficient larger than 0.8 for next strain average detection.
In order to further simplify the operation, the strain mean calculation is performed on the basis of equally dividing the IQ data group into a plurality of data blocks, and meanwhile, since the data blocks are divided already during the matching coefficient calculation, the strain mean calculation is performed in the same division mode, and the IQ data group is equally divided into 4 × 12 data blocks to extract corresponding strain mean values, so that the increase of the operation amount due to repeated division is avoided. The calculation of the strain mean is specifically described below.
Preferably, calculating a strain mean value of each IQ data set in the preliminary screening IQ data set specifically includes:
equally dividing two frames of IQ data in the IQ data group into a set number of data blocks respectively to obtain a plurality of groups of data block groups;
calculating the matching coefficient of each group of the data block groups;
extracting the relative phase shift between two frames of IQ data in the IQ data group according to the matching coefficient;
acquiring the relative displacement between two frames of IQ data in the IQ data group according to the relative phase shift;
calculating the axial gradient of the relative displacement to obtain a strain distribution value;
and calculating the average value of the strain distribution values corresponding to the data blocks to obtain the strain average value.
Specifically, in this embodiment, when the relative displacement of the data block group in the IQ data set is calculated, the calculation method of the relative displacement of the data block group of the latest frame and the next latest frame when the real-time pressing curve is calculated is adopted. And after the relative displacement is obtained through calculation, calculating the axial gradient of the strain by using a Sobel gradient operator to obtain a strain distribution value. The gradient operator in this embodiment is: [ 10-1; 20-2; 10-1].
And obtaining a strain distribution value for each data block group, and averaging the strain distribution values of a plurality of data blocks in the same IQ data group to obtain a strain average value. And each IQ data set can acquire a strain mean value, and finally, the IQ data sets are sorted from large to small according to the strain mean value, and the IQ data sets with the largest strain mean values in the first 5 groups are taken as the optimal combination for elastic calculation. If the number of IQ data groups entering the link is less than 5 frames, all the IQ data groups are output or the IQ data groups return to be null.
After the optimal combination is obtained, elastography can be performed according to the optimal combination, which is described in detail below.
Preferably, performing elastography according to the optimal combination to obtain an elastogram, specifically:
and calculating the strain distribution of the optimal combination, and mapping the strain distribution to a corresponding color table to obtain an elastic image.
When the optimal combination is calculated to obtain an elastic image, a strain distribution value needs to be calculated. In this embodiment, the strain distribution value of the optimal combination is calculated by calculating the strain distribution value of the IQ data set in the initially screened IQ data set during the secondary screening. And similarly, the optimal combination is divided into a plurality of grid blocks (namely data blocks) for calculation, the phase shift distribution of the grid blocks is firstly obtained, then the displacement distribution is obtained, finally the strain distribution is obtained through a gradient calculation module, and the strain distribution result is mapped to a corresponding color table and is output as an elastic image. The process is similar and will not be repeated here.
Before obtaining the phase shift distribution, it is preferable to perform gaussian filtering on the optimal combination, specifically, in this embodiment, two frames of IQ data in the optimal combination are subjected to gaussian filtering with a window of 3 × 5, and then are subjected to cross-correlation matching in a 9 × 81 block by using a block with a window size of 5 × 78, so as to obtain the phase shift distribution, the displacement distribution, and the strain distribution in sequence. After the phase shift distribution is extracted, filtering is also preferably performed, and a gaussian filter window with the size of 5 × 5 and a mean filter window with the size of 5 × 3 are set in the embodiment. After obtaining the strain distribution, it is also preferable to perform filtering processing, and in this embodiment, a 5 × 5 mean filtering window and a gaussian filtering window related to the smoothness of the image output result are set, the size is 5 × 9, and the filtering coefficient is 1.5, so as to obtain the elastic image.
In order to achieve good filtering of the output elastic image, the present embodiment also performs afterglow processing on the elastic image, as follows.
Preferably, performing elastography according to the optimal combination to obtain an elasticity image, further includes:
and performing weighted fusion on the current elastic image and the historical elastic image to obtain the final elastic imaging.
And performing weighted fusion on the current elastic image and the historical elastic image to obtain a final elastic imaging result, completing real-time transition of the elastic image, and outputting a real-time elastic image with smooth transition. Specifically, the weight value of the weighted fusion is set according to the gear selected by the user.
Example 2
Embodiment 2 of the present invention provides an ultrasound elastography device, comprising a processor and a memory, wherein the memory stores a computer program, and the computer program is executed by the processor to realize the ultrasound elastography method provided by embodiment 1.
The ultrasonic elastography device provided by the embodiment of the invention is used for realizing the ultrasonic elastography method, so that the ultrasonic elastography device has the technical effects of the ultrasonic elastography method, and the ultrasonic elastography device also has the technical effects, and the details are not repeated herein.
Example 3
As shown in fig. 3, embodiment 3 of the present invention provides an ultrasound elastography system, which is hereinafter referred to as the present system for short, and includes an ultrasound elastography device 102 provided in embodiment 2, an ultrasound probe 104, a B-ultrasound imaging device 101, and a display device 103;
the ultrasonic probe 104 is used for detecting the ultrasonic elasticity original data and the B-ultrasonic original data in a time-sharing manner;
the B-ultrasonic imaging device 101 is used for performing B-ultrasonic imaging according to the B-ultrasonic original data to obtain a B-ultrasonic image;
the display device 103 is used for displaying the B-mode ultrasound image and the elasticity image.
Because the B-mode ultrasound imaging is needed for guidance due to pressing of the elastography, the system simultaneously images the B-mode ultrasound image and the elastography, and the raw data of the B-mode ultrasound imaging and the elastography are acquired under different modes by sharing one ultrasonic probe 104. Because parameters required to be adjusted in a transmitting mode and a receiving mode of B-ultrasonic imaging and elastography are different, the system alternately controls the ultrasonic probe 104 to work in the two modes of the B-ultrasonic imaging mode and the elastography mode, the frame number ratio of the B-ultrasonic original data to the ultrasonic original data is 1:1 in the embodiment, the acquired ultrasonic original data and the B-ultrasonic original data enter two parallel devices, namely a B-ultrasonic imaging device 101 and an ultrasonic elastography device 102, and finally enter a display device 103 for display, and specific description is as follows.
The ultrasound probe 104 incorporates 10: a front end control device. Controlling the transmission and the reception of the ultrasonic probe 104, controlling the ultrasonic probe 104 to work between B-ultrasonic original data and elastography original data by switching frame by frame, wherein the frame number of the elastography original data and the B-ultrasonic original data is 1:1, and the ultrasonic probe is used for alternately acquiring the data.
The B-ultrasonic imaging device 101 is used for processing B-ultrasonic raw data acquired by the front end into a B-ultrasonic image. The B-mode ultrasound imaging apparatus 101 may be implemented by the prior art, and will not be described in detail herein.
The ultrasonic elastography device 102 comprises seven modules, namely a signal preprocessing module, a real-time pressing curve generation module, a first cache module, a second cache module, a combined screening module, an elastography module and an elastic afterglow module.
The signal preprocessing module is used for preprocessing the elastography original data, comprises a series of modules of band-pass filtering, quadrature demodulation, low-pass filtering, secondary sampling and low-pass filtering, and acquires IQ data which can be used for elastography calculation.
The first buffer module is a buffer module of the first buffer area and is used for buffering the IQ data output by the signal preprocessing module so as to be used for subsequent real-time pressing curve calculation, combined screening and elastography calculation. The module is used for caching IQ data with a first set frame number N1, managing stored data by using a frame number management module, indicating the sequence of data inflow through frame number management, circularly storing input data, replacing the position of the oldest data in a first buffer by the latest frame data when the storage is full, and only storing the current frame data and the historical N1-1 frame in the buffer forever. In this case, IQ data having a first buffer size of 100 frames is taken, and corresponding data storage addresses are stored in frame number management (a [100]) in the order of latest data and history data (for example, 0 to 99), where a [0] always stores the address of the latest frame data, and a [1], a [2], …, and a [99] sequentially store the previously input data.
The pressing curve generation module is used for extracting relative displacement according to the latest frame data and the next-to-latest frame data in the first cache module in real time, extracting the pressing displacement of the latest frame and storing the obtained pressing displacement value into a pressing curve cache variable.
The second cache module extracts IQ data of a second set frame number N2 from the updated data of the first cache module by judging whether the first cache module is updated or not, and combines the IQ data of the N2 frames pairwise;
the combined screening module is a key technical point of the invention, the combined screening module carries out structural similarity evaluation (SSIM) on a plurality of IQ data sets, the IQ data sets with structural similarity evaluation coefficients within a set threshold range are taken to enter the next link, IQ data sets with insufficient signal-to-noise ratio due to too small pressing force can be removed in the link, and IQ data sets with decorrelation due to too large pressing force can also be removed; then, carrying out sampling detection on the IQ data set, and uniformly sampling 4 x 12 data blocks in each frame, wherein 4 is transverse sampling, and 12 is axial sampling; respectively calculating the strain mean value and the matching coefficient of each data block group; different from the existing method, the combination with the matching coefficient larger than 0.85 is adopted; sequentially taking N3 IQ data groups with the maximum strain mean values in a set number from combinations with the correlation matching coefficient larger than 0.85, and outputting the IQ data groups as the optimal combinations to an elastic imaging module; n3 may be N2 or a natural integer less than N2 greater than 0;
the elastic imaging module is an elastic image calculation device and calculates the optimal combination output by the combination screening module to obtain the optimal elastic image.
The elastic afterglow module finishes caching the elastic image output by the elastic imaging module and weighting fusion of a certain ratio, and parameters of the weighting fusion and the number of participating frames are set by an afterglow gear; the afterglow treatment can realize the good transition of the elastic image, and cannot cause a sharp image phenomenon for a user. Finally, the elastic image is output to the display device 103.
And the display device 103 is used for performing fusion display on the B-ultrasonic image, the elastic image and the real-time pressing curve.
Example 4
Embodiment 4 of the present invention provides a computer storage medium having stored thereon a computer program which, when executed by a processor, implements the ultrasound elastography method provided in embodiment 1.
The computer storage medium provided by the embodiment of the invention is used for the ultrasonic elastography method, so that the computer storage medium has the technical effects of the ultrasonic elastography method, and the details are not repeated herein.
The above-described embodiments of the present invention should not be construed as limiting the scope of the present invention. Any other corresponding changes and modifications made according to the technical idea of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. An ultrasonic elastography method, comprising the steps of:
acquiring elastography original data, performing data preprocessing on the elastography original data to obtain IQ data, and caching the IQ data;
extracting partial IQ data from the cached multi-frame IQ data to be used as an IQ data set to be screened; combining IQ data in the IQ data set to be screened pairwise to obtain a plurality of IQ data sets;
calculating a structural similarity evaluation coefficient of each IQ data set, and taking the IQ data set with the structural similarity evaluation coefficient within a set threshold range as a primary screening IQ data set;
calculating the imaging quality parameters of each IQ data set in the primary screening IQ data set, and screening the optimal combination from the primary screening IQ data set according to the imaging quality parameters;
and performing elastography according to the optimal combination to obtain an elastic image.
2. The method of ultrasonic elastography according to claim 1, further comprising: and adopting a parallel thread to perform elastography according to the cached IQ data and generate a real-time pressing curve according to the cached IQ data.
3. The ultrasound elastography method of claim 2, wherein a real-time compression curve is generated from the cached IQ data, specifically:
extracting the latest frame IQ data and the next-to-new frame IQ data from the cached IQ data;
setting a plurality of sampling blocks with the same depth and the same size;
sampling a plurality of pairs of corresponding data blocks from the latest frame IQ data block and the next-new frame IQ data block respectively through the sampling block to obtain a plurality of groups of data block groups;
calculating the matching coefficient of each group of the data block groups;
extracting the relative phase shift between the latest frame IQ data and the next-new frame IQ data according to the matching coefficient;
acquiring the relative displacement between the latest frame IQ data and the next-new frame IQ data according to the relative phase shift;
carrying out weighted summation on the relative displacement corresponding to each group of data block groups to obtain real-time relative displacement;
and generating the real-time pressing curve according to the real-time relative displacement.
4. The method according to claim 1, wherein the calculating the structural similarity evaluation coefficient of each IQ data set comprises:
calculating envelope data of each frame of IQ data in the IQ data group:
wherein, the data is envelope data of IQ data, I is an imaginary part of orthogonal demodulation data of the IQ data, and Q is a real part of the orthogonal demodulation data of the IQ data;
calculating a structural similarity evaluation coefficient between two frames of IQ data in the IQ data group according to the envelope data:
wherein x and y are envelope data of two frames of IQ data in the same IQ data set, SSIM (x, y) is a structural similarity evaluation coefficient between two frames of IQ data in the same IQ data set, UxIs the mean, U, of the envelope data xyIs the mean, σ, of the envelope data yx 2Is the variance, σ, of the envelope data xy 2Is the variance, σ, of the envelope data yxyCovariance for envelope data x and for envelope data y, C1、C2Are all constants.
5. The ultrasound elastography method according to claim 1, wherein the imaging quality parameters of each IQ data set in the preliminary screening IQ data set are calculated, and an optimal combination is screened out from the preliminary screening IQ data set according to the imaging quality parameters, specifically:
the imaging quality parameters comprise matching coefficients and strain mean values, and the matching coefficients and the strain mean values of all IQ data sets in the primary screening IQ data sets are calculated;
and screening the IQ data sets with the matching coefficients larger than the matching threshold, and selecting a set number of IQ data sets with the maximum strain mean value from the IQ data sets with the matching coefficients larger than the matching threshold as the optimal combination.
6. The ultrasound elastography method according to claim 5, wherein the calculating of the matching coefficients of each IQ data set in the preliminary screening IQ data sets specifically comprises:
setting two frames of IQ data in the IQ data group as a reference frame and a matching frame respectively;
dividing the reference frame into a set number of reference data blocks, and dividing a retrieval frame corresponding to the data block positions in the matching frame;
calculating the cross correlation coefficient of each reference data block in the reference frame and a plurality of matched data blocks in the corresponding retrieval frame:
RA12(t,x,n,m)＝∫∫DS1(t+v,x+w)*S2(t+n+v,x+m+w)*dvdw
wherein RA is12(t, x, n, m) is a cross-correlation coefficient, v is the transverse length of a correlation window, w is the axial length of the correlation window, a plurality of matched data blocks in a retrieval frame are sequentially selected according to a set step length, n is the transverse stepping length, m is the axial stepping length, t is the axial coordinate of the center of the matched data block, x is the transverse coordinate of the center of the matched data block, and S is the transverse coordinate of the center of the matched data block1(t + v, x + w) is the IQ data value of the reference block at the (t + v, x + w) pixel point, S2(t + n + v, x + m + w) is the IQ data value of the matched data block at the (t + n + v, x + m + w) pixel point, D ═ v × w;
calculating the autocorrelation coefficient of the reference frame:
RA11(t,x,0,0)＝∫∫D|S1(t+v,x+w)|2dvdw
wherein RA is11(t, x,0,0) is the autocorrelation coefficient of the reference frame;
calculating the autocorrelation coefficient of the matched frame:
RA22(t,x,n,m)＝∫∫D|S2(t+n+v,x+m+w)|2dvdw
wherein RA is22(t, x, n, m) is the autocorrelation coefficient of the matched frame;
calculating the matching coefficient of the reference frame and the matching frame according to the cross correlation coefficient corresponding to each group of matched data blocks, the autocorrelation coefficient of the reference frame and the autocorrelation coefficient of the matching frame:
wherein C (t, x, n, m) is a matching coefficient of the reference frame and the matching frame;
and calculating the average value of the matching coefficients corresponding to the multiple groups of data blocks as a final matching coefficient.
7. The ultrasound elastography method of claim 5, wherein calculating the strain mean of each IQ data set in the preliminary screening IQ data set specifically comprises:
equally dividing two frames of IQ data in the IQ data group into a set number of data blocks respectively to obtain a plurality of groups of data block groups;
calculating the matching coefficient of each group of the data block groups;
extracting the relative phase shift between two frames of IQ data in the IQ data group according to the matching coefficient;
acquiring the relative displacement between two frames of IQ data in the IQ data group according to the relative phase shift;
calculating the axial gradient of the relative displacement to obtain a strain distribution value;
and calculating the average value of the strain distribution values corresponding to the data blocks to obtain the strain average value.
8. An ultrasound elastography device, characterized by a processor and a memory, said memory having stored thereon a computer program which, when executed by said processor, carries out an ultrasound elastography method as claimed in any one of claims 1 to 7.
9. The ultrasound elastography system of claim 1, comprising the ultrasound elastography device of claim 8, further comprising an ultrasound probe, a B-ultrasound imaging device, and a display device;
the ultrasonic probe is used for detecting the ultrasonic elasticity original data and the B ultrasonic original data in a time-sharing manner;
the B-ultrasonic imaging device is used for carrying out B-ultrasonic imaging according to the B-ultrasonic original data to obtain a B-ultrasonic image;
the display device is used for displaying the B-ultrasonic image and the elastic image.
10. A computer storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the ultrasound elastography method as claimed in any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010447279.3A CN111528912A (en) | 2020-05-25 | 2020-05-25 | Ultrasonic elastography method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010447279.3A CN111528912A (en) | 2020-05-25 | 2020-05-25 | Ultrasonic elastography method, device and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111528912A true CN111528912A (en) | 2020-08-14 |
Family
ID=71969537
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010447279.3A Pending CN111528912A (en) | 2020-05-25 | 2020-05-25 | Ultrasonic elastography method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111528912A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103845081A (en) * | 2012-11-28 | 2014-06-11 | 深圳迈瑞生物医疗电子股份有限公司 | System and method for ultrasonic elastography and method for real-time dynamic interframe processing |
| CN104739442A (en) * | 2013-12-25 | 2015-07-01 | 深圳迈瑞生物医疗电子股份有限公司 | Pressure elastic imaging displacement detection method, pressure elastic imaging displacement detection device and ultrasonic imaging device |
| CN109745073A (en) * | 2019-01-10 | 2019-05-14 | 武汉中旗生物医疗电子有限公司 | The two-dimentional matching process and equipment of elastogram displacement |
-
2020
- 2020-05-25 CN CN202010447279.3A patent/CN111528912A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103845081A (en) * | 2012-11-28 | 2014-06-11 | 深圳迈瑞生物医疗电子股份有限公司 | System and method for ultrasonic elastography and method for real-time dynamic interframe processing |
| CN104739442A (en) * | 2013-12-25 | 2015-07-01 | 深圳迈瑞生物医疗电子股份有限公司 | Pressure elastic imaging displacement detection method, pressure elastic imaging displacement detection device and ultrasonic imaging device |
| CN109745073A (en) * | 2019-01-10 | 2019-05-14 | 武汉中旗生物医疗电子有限公司 | The two-dimentional matching process and equipment of elastogram displacement |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8202220B2 (en) | Ultrasound diagnostic apparatus | |
| US20210236095A1 (en) | Ultrasound ct image reconstruction method and system based on ray theory | |
| CN112086197B (en) | Breast nodule detection method and system based on ultrasonic medicine | |
| US20210407371A1 (en) | Method, apparatus and device for acquiring luminance compensation data | |
| CN1134808A (en) | Method of displaying ultrasonic images and apparatus for ultrasonic diagnosis | |
| CN109191416A (en) | Image interfusion method based on sparse dictionary study and shearing wave | |
| CN112529870A (en) | Multi-scale CNNs (CNNs) lung nodule false positive elimination method based on combination of source domain and frequency domain | |
| CN111862038A (en) | Plaque detection method, device, equipment and medium | |
| CN112950480A (en) | Super-resolution reconstruction method integrating multiple receptive fields and dense residual attention | |
| CN113095358A (en) | Image fusion method and system | |
| CN1936926A (en) | Image blind separation based on sparse change | |
| CN214587004U (en) | Stereo matching acceleration circuit, image processor and three-dimensional imaging electronic equipment | |
| CN102760293B (en) | Image quality evaluation method based on distance matrix | |
| CN111528912A (en) | Ultrasonic elastography method, device and system | |
| Li | (Retracted) Infrared image filtering and enhancement processing method based upon image processing technology | |
| US20120157847A1 (en) | Ultrasound system and method for processing beam-forming based on sampling data | |
| CN109615591A (en) | A kind of three-dimensional Block- matching noise-reduction method accelerated parallel based on GPU | |
| Ma et al. | Glomerulus extraction by using genetic algorithm for edge patching | |
| CN114821493A (en) | Ship information display method and system based on computer vision, AIS and radar | |
| CN111192663B (en) | Magnetic resonance film imaging method, device, equipment and storage medium | |
| CN106251304A (en) | Dynamic image segmented system and method | |
| CN114298901A (en) | Portrait super-resolution reconstruction method, model training method, device, electronic equipment and readable storage medium | |
| CN114066726A (en) | Image processing method, device, equipment and storage medium | |
| Ye et al. | A Gray Scale Correction Method for Side-Scan Sonar Images Based on GAN | |
| CN104546008B (en) | A kind of first-harmonic/harmonic wave merges the formation method combined with space compound |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200814 |