CN109965905A - A kind of radiography region detection imaging method based on deep learning - Google Patents
A kind of radiography region detection imaging method based on deep learning Download PDFInfo
- Publication number
- CN109965905A CN109965905A CN201910289375.7A CN201910289375A CN109965905A CN 109965905 A CN109965905 A CN 109965905A CN 201910289375 A CN201910289375 A CN 201910289375A CN 109965905 A CN109965905 A CN 109965905A
- Authority
- CN
- China
- Prior art keywords
- signal
- actual measurement
- imaging
- signal segment
- deep learning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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
-
- 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
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Biomedical Technology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
- Image Processing (AREA)
Abstract
The present invention proposes a kind of radiography region detection imaging method based on deep learning.Signal subsection in its every scan line is obtained several isometric one-dimensional RF signal segments by S1, several raw ultrasounds RF image randomly selected；S2, with the training of RF signal segment, test convolutional neural networks；S3, the RF signal for surveying the every scan line of object is divided into isometric one-dimensional actual measurement RF signal segment, and inputs the trained convolutional neural networks of institute；S4, label is set to 0 for P imaging point in the middle part of the one-dimensional actual measurement RF signal segment of tissue signal；S5, the one-dimensional actual measurement RF signal segment of recombination obtain pre-imaging；S6, it is converted using microvesicle morther wavelet and improves microvesicle brightness of image in pre-imaging；S7, picture contrast is improved using feature space minimum variation algorithm.The theory of deep learning is applied in the classification of ultrasonic RF signal by the present invention, can more effectively be filtered out tissue interference, further be improved the accuracy of medical clinic applications.
Description
Technical field
The present invention relates to image segmentation field, in particular to a kind of radiography region detection imaging side based on deep learning
Method.
Background technique
Acoustic contrast agent is made of a large amount of microbubbles, while ultrasonic backscattered signal can be enhanced, produces echo-signal
Raw harmonic components abundant.To human injection's acoustic contrast agent, ultrasonograph quality is helped to improve, assigns ultrasound diagnosis identification
The ability of tiny lesion.
Occur in recent years a series of contrast imaging new methods (harmonic imaging, dipulse transmitting imaging, coded pulse at
Picture, microvesicle morther wavelet technology), these methods are all based on greatly a principle: while sufficiently extracting the harmonic components of microvesicle, filter
The fundamental wave component for carrying out self-organizing is removed, to improve contrastographic picture contrast.
But the prior art, due to being to be completed at the same time extraction microbubble signals ingredient and filter out tissue signal components, method sheet
Body will take into account the two, therefore not strong to respective specific aim, to tissue signal filtration result ratio when tissue signal is too strong
It is poor, affect the quality of the contrast imaging of final output.
Summary of the invention
The object of the present invention is to provide a kind of radiography region detection imaging method based on deep learning, by deep learning
Theory is applied in the classification of ultrasonic RF (radio frequency Radio Frequency) signal, and by convolutional neural networks, preliminary distinguish is made
Microbubble signals and tissue signal in the raw ultrasound RF image of shadow imaging, obtain the pre-imaging of radiography；Then pass through microvesicle again
Morther wavelet imaging method, beamforming algorithm further increase the image quality of microbubble signals for pre-imaging.
In order to achieve the above object, the present invention provides a kind of radiography region detection imaging method based on deep learning, makes
The ultrasonic RF image of shadow imaging is made of the RF signal in several from left to right scan lines, radiography region detection imaging side
Method includes step:
S1, the raw ultrasound RF image for randomly selecting several contrast imagings are simultaneously divided into two groups, establish experimental data set；It will
RF signal in every scan line of the raw ultrasound RF image is sequentially segmented from top to bottom, and it is one-dimensional and isometric to obtain several
RF signal segment；The RF signal segment includes n continuous imaging points, adjacent interval m continuous imaging points of RF signal segment, and one
The corresponding RF signal of a imaging point；
S2, tag along sort set Y={ tissue signal, microbubble signals } are established；It will be mentioned from first group of raw ultrasound RF image
The RF signal segment input convolutional neural networks taken, obtain trained convolutional neural networks；By second group of raw ultrasound RF image
RF signal segment input the trained convolutional neural networks, test the trained convolutional neural networks；
S3, radiography is carried out to actual measurement object, and sequentially to the RF signal subsection in the actual measurement every scan line of object, if obtaining
Dry isometric one-dimensional actual measurement RF signal segment；Each actual measurement RF signal segment includes n continuous imaging points, adjacent actual measurement RF letter
Number section interval m continuous imaging points；
S4, the actual measurement RF signal segment is inputted into the trained convolutional neural networks, when point of actual measurement RF signal segment
When class label is tissue signal, P imaging point in the middle part of pulverised actual measurement RF signal segment；
S5, the actual measurement RF signal segment is recombinated by scan line, obtains the two-dimentional pre-imaging of actual measurement object；
S6, using microvesicle morther wavelet imaging method, improve the brightness of microvesicle imaging point in the pre-imaging；
S7, the contrast being imaged using beamforming algorithm, raising step S6 gained, obtain the two-dimensional ultrasound of actual measurement object
Image.
M=5 in step S1 and step S3, and n=60.
P imaging point, in particular to by the measured signal section in the middle part of the actual measurement RF of pulverised described in step S4 signal segment
The RF signal value of 29th to the 33rd imaging point is set to 0, P=5.
The actual measurement RF signal segment is recombinated by scan line described in step S5, in particular to every after step S4 is scanned
Actual measurement RF signal segment on line, it is primary every 5 imaging point samplings；By all sampled points constitute actual measurement object two dimension in advance at
Picture；In the pre-imaging, each sampled point is in former scan line.
The convolutional neural networks are U-net convolutional neural networks.
The U-net convolutional neural networks are using cross entropy as cost function, using ReLU function as nonlinear activation letter
Number, using Adam algorithm as optimization algorithm.
First group of raw ultrasound RF image is four times of second group of raw ultrasound RF image.
Compared with the prior art, the advantages of the present invention are as follows: it is obtained in the ultrasonic RF signal for handling scanning by the prior art
Before obtaining ultrasonic RF image, using the method for deep learning, ultrasonic RF one-dimensional signal is classified, preliminary screening falls major part
Tissue signal obtains the pre-imaging of ultrasound RF image.Then empty using microvesicle morther wavelet imaging method, feature to the pre-imaging
Between minimum variation algorithm obtain final ultrasonic RF image.The present invention can more effectively filter out the interference of tissue signal, further
Improve the accuracy of medical clinic applications.
Detailed description of the invention
In order to illustrate more clearly of technical solution of the present invention, attached drawing needed in description will be made simply below
It introduces, it should be apparent that, the accompanying drawings in the following description is one embodiment of the present of invention, and those of ordinary skill in the art are come
It says, without creative efforts, is also possible to obtain other drawings based on these drawings:
Fig. 1 is the radiography region detection imaging method flow chart of the invention based on deep learning；
Fig. 2 is in step S3 of the invention, to the RF signal subsection schematic diagram in every scan line of actual measurement object；
Fig. 3 is the 29th to the 33rd imaging point schematic diagram of pulverised measured signal section in step S4 of the invention；
Fig. 4 is in step S5 of the invention to actual measurement RF signal segment sampled result schematic diagram；
Fig. 5 is the two-dimentional pre-imaging schematic diagram for recombinating all sampled points in step S5 of the invention and constituting actual measurement object.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The present invention provides a kind of radiography region detection imaging method based on deep learning, the ultrasonic RF image of contrast imaging
It is made of the RF signal in several from left to right scan lines.As shown in Figure 1, the radiography region detection imaging method includes step
It is rapid:
S1, the raw ultrasound RF image for randomly selecting several contrast imagings are simultaneously divided into two groups, establish experimental data set,
In first group of raw ultrasound RF image be four times of second group of raw ultrasound RF image.The raw ultrasound RF is swept for image every
The RF signal retouched on line is sequentially segmented from top to bottom, obtains several one-dimensional and isometric RF signal segments.The RF signal segment packet
Containing 60 continuous imaging points, the adjacent continuous imaging point in 5, RF signal segment interval, the corresponding RF letter of an imaging point
Number.
S2, tag along sort set Y={ tissue signal, microbubble signals } are established.It will be mentioned from first group of raw ultrasound RF image
The RF signal segment input U-net convolutional neural networks taken, obtain trained convolutional neural networks, by second group of raw ultrasound RF
The RF signal segment of image inputs the trained convolutional neural networks, tests the trained convolutional neural networks.It is described
U-net convolutional neural networks are using cross entropy as cost function, and with ReLU, (Rectified Linear Unit rectification is linear single
Member) function is as nonlinear activation function, using Adam algorithm as optimization algorithm.
S3, radiography is carried out to actual measurement object, and sequentially to the RF signal subsection in the actual measurement every scan line of object, if obtaining
Dry isometric one-dimensional actual measurement RF signal segment.Each actual measurement RF signal segment includes 60 continuous imaging points, adjacent actual measurement RF
The continuous imaging point in 5, signal segment interval.
Fig. 2 is the RF signal subsection schematic diagram in every scan line to actual measurement object；Include k scan line in Fig. 2, leads to
Actual measurement object is imaged in the RF signal for crossing the k scan line.Every scan line is divided into M actual measurement RF signal segment, RFij
Indicate j-th of actual measurement RF signal segment in i-th scan line, i ∈ [1, k], j ∈ [1, M], RFijInclude 60 imaging points.RFijWith
RFi(j+1)Between be spaced 5 imaging points (j ∈ [1, M-1]), RFijWith RFi(j-1)Between be spaced 5 imaging points (j ∈ [2, M]).Step
In rapid S1, according to identical method shown in Fig. 2, from top to bottom sequentially to the RF signal in raw ultrasound RF every scan line of image
Segmentation.
S4, the actual measurement RF signal segment is inputted into the trained convolutional neural networks, when point of actual measurement RF signal segment
When class label is tissue signal, the RF signal value of the 29th to the 33rd imaging point of the measured signal section is set to 0.
As shown in figure 3,To survey RF signal segment RFij60 imaging points.As actual measurement RF signal segment RFij
Tag along sort be tissue signal when, willThe RF signal value of totally 5 imaging points is set to 0.
S5, the actual measurement RF signal segment is recombinated by scan line, in particular in every scan line after step S4
RF signal segment is surveyed, it is primary every 5 imaging point samplings；As shown in figure 4, to as actual measurement RF signal segment RFijIt is obtained after sampling
RFij', RFij' it include RFij's
Totally 12 imaging points.
The two-dimentional pre-imaging that actual measurement object is made up of all sampled points, in the pre-imaging, each sampled point is swept in original
It retouches on line.As shown in figure 5, passing through RF11'~RFkMThe two-dimentional pre-imaging of ' composition actual measurement object, and RF11'~RFkM' still respective
In affiliated scan line.
S6, using microvesicle morther wavelet imaging method, improve the brightness of microvesicle imaging point in the pre-imaging；
S7, the contrast being imaged using beamforming algorithm, raising step S6 gained, obtain the two-dimensional ultrasound of actual measurement object
Image.
Compared with the prior art, the advantages of the present invention are as follows: it is obtained in the ultrasonic RF signal for handling scanning by the prior art
Before obtaining ultrasonic RF image, using the method for deep learning, ultrasonic RF one-dimensional signal is classified, preliminary screening falls major part
Tissue signal obtains the pre-imaging of ultrasound RF image.Then empty using microvesicle morther wavelet imaging method, feature to the pre-imaging
Between minimum variation algorithm obtain final ultrasonic RF image.The present invention can more effectively filter out the interference of tissue signal, further
Improve the accuracy of medical clinic applications.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (8)
1. a kind of radiography region detection imaging method based on deep learning, if the ultrasonic RF image of contrast imaging is by from left to right
RF signal composition in dry scan line, which is characterized in that the radiography region detection imaging method includes step:
S1, the raw ultrasound RF image for randomly selecting several contrast imagings are simultaneously divided into two groups, establish experimental data set；It will be described
RF signal in raw ultrasound RF every scan line of image is sequentially segmented from top to bottom, obtains several one-dimensional and isometric RF letters
Number section；The RF signal segment includes n continuous imaging points, adjacent interval m continuous imaging points of RF signal segment, one at
Picture point corresponds to a RF signal；
S2, tag along sort set Y={ tissue signal, microbubble signals } are established；It will be from first group of raw ultrasound RF image zooming-out
RF signal segment inputs convolutional neural networks, obtains trained convolutional neural networks；By the RF of second group of raw ultrasound RF image
Signal segment inputs the trained convolutional neural networks, tests the trained convolutional neural networks；
S3, radiography is carried out to actual measurement object, and several sequentially is obtained to the RF signal subsection in the actual measurement every scan line of object
Isometric one-dimensional actual measurement RF signal segment；Each actual measurement RF signal segment includes n continuous imaging points, adjacent actual measurement RF signal segment
It is spaced m continuous imaging points；
S4, the actual measurement RF signal segment is inputted into the trained convolutional neural networks, when the contingency table of actual measurement RF signal segment
When label are tissue signal, P imaging point in the middle part of pulverised actual measurement RF signal segment；
S5, the actual measurement RF signal segment is recombinated by scan line, obtains the two-dimentional pre-imaging of actual measurement object；
S6, using microvesicle morther wavelet imaging method, improve the brightness of microvesicle imaging point in the pre-imaging；
S7, the contrast being imaged using beamforming algorithm, raising step S6 gained, obtain the two-dimensional ultrasound figure of actual measurement object
Picture.
2. the radiography region detection imaging method based on deep learning as described in claim 1, which is characterized in that step S1 and
M=5 in step S3, and n=60.
3. the radiography region detection imaging method based on deep learning as described in claim 1, which is characterized in that in step S4
In the middle part of the pulverised actual measurement RF signal segment P imaging point, in particular to by the 29th of the measured signal section to the 33rd at
The RF signal value of picture point is set to 0, P=5.
4. the radiography region detection imaging method based on deep learning as described in claim 1, which is characterized in that step S5 institute
It states and recombinates the actual measurement RF signal segment by scan line, in particular to for the actual measurement RF signal in every scan line after step S4
Section, it is primary every 5 imaging point samplings；The two-dimentional pre-imaging of actual measurement object is made up of all sampled points；In the pre-imaging,
Each sampled point is in former scan line.
5. the radiography region detection imaging method based on deep learning as described in claim 1, which is characterized in that the convolution
Neural network is U-net convolutional neural networks.
6. the radiography region detection imaging method based on deep learning as claimed in claim 5, which is characterized in that the U-
Net convolutional neural networks are using cross entropy as cost function, using ReLU function as nonlinear activation function, with Adam algorithm work
For optimization algorithm.
7. the radiography region detection imaging method based on deep learning as described in claim 1, which is characterized in that the wave beam
Formation algorithm is characterized space minimum variation algorithm.
8. the radiography region detection imaging method based on deep learning as described in claim 1, which is characterized in that first group former
Beginning ultrasound RF image is four times of second group of raw ultrasound RF image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910289375.7A CN109965905B (en) | 2019-04-11 | 2019-04-11 | Contrast region detection imaging method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910289375.7A CN109965905B (en) | 2019-04-11 | 2019-04-11 | Contrast region detection imaging method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109965905A true CN109965905A (en) | 2019-07-05 |
| CN109965905B CN109965905B (en) | 2020-02-11 |
Family
ID=67084140
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910289375.7A Active CN109965905B (en) | 2019-04-11 | 2019-04-11 | Contrast region detection imaging method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109965905B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110399915A (en) * | 2019-07-23 | 2019-11-01 | 王英伟 | A kind of Ultrasound Image Recognition Method and its system based on deep learning |
| WO2021007989A1 (en) * | 2019-07-12 | 2021-01-21 | 华南理工大学 | High-contrast minimum variance imaging method based on deep learning |
| CN113436109A (en) * | 2021-07-08 | 2021-09-24 | 清华大学 | Ultrafast high-quality plane wave ultrasonic imaging method based on deep learning |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0760625A4 (en) * | 1994-02-04 | 1999-02-03 | Long Island Jewish Medical Ct | Method of tissue characterization by ultrasound |
| US20020040187A1 (en) * | 2000-07-20 | 2002-04-04 | Alam Sheikh Kaisar | Methods for estimating tissue strain |
| CN1757381A (en) * | 2005-09-21 | 2006-04-12 | 北京市肿瘤防治研究所 | Method for improving resolution of ultrasonic image-forming image, and ultrasonic contrast image-forming apparatus |
| CN103330576A (en) * | 2013-06-09 | 2013-10-02 | 西安交通大学 | Micro-elasticity imaging method based on tissue microbubble dynamics model |
| CN103381096A (en) * | 2013-04-19 | 2013-11-06 | 西安交通大学 | Blood perfusion separation detecting and imaging method for bone surface capillary |
| WO2015092667A1 (en) * | 2013-12-20 | 2015-06-25 | Koninklijke Philips N.V. | System and method for tracking a penetrating instrument |
| CN105574820A (en) * | 2015-12-04 | 2016-05-11 | 南京云石医疗科技有限公司 | Deep learning-based adaptive ultrasound image enhancement method |
| CN106897573A (en) * | 2016-08-01 | 2017-06-27 | 12西格玛控股有限公司 | Use the computer-aided diagnosis system for medical image of depth convolutional neural networks |
| CN106991445A (en) * | 2017-04-05 | 2017-07-28 | 重庆大学 | A kind of ultrasonic contrast tumour automatic identification and detection method based on deep learning |
| US20180177461A1 (en) * | 2016-12-22 | 2018-06-28 | The Johns Hopkins University | Machine learning approach to beamforming |
-
2019
- 2019-04-11 CN CN201910289375.7A patent/CN109965905B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0760625A4 (en) * | 1994-02-04 | 1999-02-03 | Long Island Jewish Medical Ct | Method of tissue characterization by ultrasound |
| US20020040187A1 (en) * | 2000-07-20 | 2002-04-04 | Alam Sheikh Kaisar | Methods for estimating tissue strain |
| CN1757381A (en) * | 2005-09-21 | 2006-04-12 | 北京市肿瘤防治研究所 | Method for improving resolution of ultrasonic image-forming image, and ultrasonic contrast image-forming apparatus |
| CN103381096A (en) * | 2013-04-19 | 2013-11-06 | 西安交通大学 | Blood perfusion separation detecting and imaging method for bone surface capillary |
| CN103330576A (en) * | 2013-06-09 | 2013-10-02 | 西安交通大学 | Micro-elasticity imaging method based on tissue microbubble dynamics model |
| WO2015092667A1 (en) * | 2013-12-20 | 2015-06-25 | Koninklijke Philips N.V. | System and method for tracking a penetrating instrument |
| CN105574820A (en) * | 2015-12-04 | 2016-05-11 | 南京云石医疗科技有限公司 | Deep learning-based adaptive ultrasound image enhancement method |
| CN106897573A (en) * | 2016-08-01 | 2017-06-27 | 12西格玛控股有限公司 | Use the computer-aided diagnosis system for medical image of depth convolutional neural networks |
| US20180177461A1 (en) * | 2016-12-22 | 2018-06-28 | The Johns Hopkins University | Machine learning approach to beamforming |
| CN106991445A (en) * | 2017-04-05 | 2017-07-28 | 重庆大学 | A kind of ultrasonic contrast tumour automatic identification and detection method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| BAIYING LEI等: "Segmentation of breast anatomy for automated whole breast ultrasound images with boundary regularized convolutional encoder–decoder network", 《NEUROCOMPUTING》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021007989A1 (en) * | 2019-07-12 | 2021-01-21 | 华南理工大学 | High-contrast minimum variance imaging method based on deep learning |
| CN110399915A (en) * | 2019-07-23 | 2019-11-01 | 王英伟 | A kind of Ultrasound Image Recognition Method and its system based on deep learning |
| CN113436109A (en) * | 2021-07-08 | 2021-09-24 | 清华大学 | Ultrafast high-quality plane wave ultrasonic imaging method based on deep learning |
| CN113436109B (en) * | 2021-07-08 | 2022-10-14 | 清华大学 | Ultrafast high-quality plane wave ultrasonic imaging method based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109965905B (en) | 2020-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11490877B2 (en) | System and method of identifying characteristics of ultrasound images | |
| CN109965905A (en) | A kind of radiography region detection imaging method based on deep learning | |
| DE102012108353B4 (en) | CLASSIFICATION PREPROCESSING IN MEDICAL ULTRASONIC SHEAR WAVE IMAGING | |
| EP3655917B1 (en) | Fetal ultrasound image processing | |
| DE102016108737A1 (en) | Knowledge-based ultrasound image enhancement | |
| DE102016100367A1 (en) | Sparse tracking in sound beam intensity impulse imaging | |
| CN105913432A (en) | Aorta extracting method and aorta extracting device based on CT sequence image | |
| DE102010061577A1 (en) | Method and system for processing ultrasound data | |
| DE102012022010A1 (en) | Adaptive image optimization in wave-induced ultrasound imaging | |
| JP2004033732A (en) | System and method for post-processing of ultrasonic color doppler imaging | |
| DE202013105253U1 (en) | imaging device | |
| Iula et al. | An ultrasound technique for 3D palmprint extraction | |
| CN108670303B (en) | Method and system for detecting uniformity of ultrasonic image | |
| CN107346541A (en) | A kind of tissue characterization method based on ultrasonic radio frequency time series wavelet analysis | |
| CN106725592A (en) | The automatic identifying method and system of ultrasonic radio frequency metadata | |
| CN106022369A (en) | Bone quality evaluation method based on ultrasonic backscatter signal imaging | |
| CN102800072B (en) | Diasonograph scan image quality optimization disposal route and device thereof | |
| DE202013105255U1 (en) | ultrasonic testing | |
| CN111829956B (en) | Photoacoustic endoscopic quantitative tomography method and system based on layered guidance of ultrasonic structure | |
| CN116058875A (en) | Ultrasonic imaging method and ultrasonic imaging system | |
| JP5959880B2 (en) | Ultrasonic diagnostic equipment | |
| CN206151437U (en) | Multimode microwave imaging device | |
| CN115937219B (en) | Ultrasonic image part identification method and system based on video classification | |
| JP2002125971A (en) | Ultrasonic diagnostic instrument | |
| CN105725965A (en) | Multi-mode microwave scanning and breast imaging method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |