CN109325925A - A kind of medicine dynamic image respiration motion compensation method based on sparse subspace clustering - Google Patents
A kind of medicine dynamic image respiration motion compensation method based on sparse subspace clustering Download PDFInfo
- Publication number
- CN109325925A CN109325925A CN201811112741.3A CN201811112741A CN109325925A CN 109325925 A CN109325925 A CN 109325925A CN 201811112741 A CN201811112741 A CN 201811112741A CN 109325925 A CN109325925 A CN 109325925A
- Authority
- CN
- China
- Prior art keywords
- dynamic image
- sparse subspace
- medicine dynamic
- motion compensation
- medicine
- 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
- 238000000034 method Methods 0.000 title claims abstract description 66
- 239000003814 drug Substances 0.000 title claims abstract description 52
- 230000029058 respiratory gaseous exchange Effects 0.000 title claims abstract description 37
- 239000011159 matrix material Substances 0.000 claims abstract description 44
- 230000000694 effects Effects 0.000 claims abstract description 6
- 230000003595 spectral effect Effects 0.000 claims abstract description 5
- 238000010606 normalization Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 description 6
- 230000000241 respiratory effect Effects 0.000 description 5
- 238000002595 magnetic resonance imaging Methods 0.000 description 4
- 230000002411 adverse Effects 0.000 description 3
- 210000001015 abdomen Anatomy 0.000 description 2
- 210000000038 chest Anatomy 0.000 description 2
- 238000002607 contrast-enhanced ultrasound Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000009792 diffusion process Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000035479 physiological effects, processes and functions Effects 0.000 description 2
- 238000011946 reduction process Methods 0.000 description 2
- 238000009394 selective breeding Methods 0.000 description 2
- 238000002604 ultrasonography Methods 0.000 description 2
- 206010020751 Hypersensitivity Diseases 0.000 description 1
- 208000026935 allergic disease Diseases 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000009610 hypersensitivity Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2136—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on sparsity criteria, e.g. with an overcomplete basis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10132—Ultrasound image
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The medicine dynamic image respiration motion compensation method based on sparse subspace clustering that the invention discloses a kind of, step S1 acquire medicine dynamic image, determine preprocess method;Step S2 selects sparse subspace representation method, and determines the corresponding penalty term constrained procedure of the sparse subspace representation method;Step S3, solves and obtains the sparse subspace expression coefficient matrix of the medicine dynamic image, and the sparse subspace indicates that coefficient matrix is used to indicate the relationship in medicine dynamic image between frame and frame;Step S4, the sparse subspace obtained according to the step S3 indicate that coefficient matrix is handled, construct measuring similarity matrix;Step S5 obtains the cluster result of medicine dynamic image using Spectral Clustering;Step S6 selects final respiration motion compensation image.Advantages of the present invention: improving the precision of image clustering, further improves the effect of dynamic image respiration motion compensation.
Description
Technical field
The present invention relates to a kind of crossing domains for belonging to computer technology and Medical Image Processing, and in particular to one kind is based on
The medicine dynamic image respiration motion compensation method of sparse subspace clustering.
Background technique
Medicine dynamic image is time-series image, provides the important diagnostic information of observed object physiology or pathology, is led to
The dynamic image presentation for crossing observation observed object difference phase in real time, can identify disease, state or physiology course, especially
It is that there is very hypersensitivity and specificity to good, malignant tumour diagnosis.Currently, medicine dynamic imaging methods mainly have ultrasound to make
Shadow (Contrast-Enhanced UltraSound, CEUS), enhanced CT (Computed Tomography, CT), Contrast-enhanced MRI
(Magnetic Resonance Imaging, MRI) etc..However, in progress chest, abdomen medicine dynamic image checking process,
With the breathing of human body, the shape of thorax abdomen organ, position can occur significantly to change, and cause the dynamic sequence picture frame obtained
Between frame object observing move back and forth, deformation, be unfavorable for the observation and diagnosis of dynamic image.
To reduce breathing bring adverse effect, existing respiration motion compensation method mainly has method for registering images and exhales
Inhale gate control method.Method for registering images is in dynamic sequence image, and wherein a frame image is reference picture for artificial selection, is used
Remaining frame image is registrated with reference picture by image registration algorithm frame by frame, and this method needs manual intervention, vulnerable to subjective factor
It influences.Respiration gate control method is then extracted first can reflect respirometric periodic amplitude breath signal, then be believed according to breathing
Number amplitude each respiratory cycle (primary exhale and an air-breathing is a cycle) is divided into several position of breathing, finally set
Counting certain gating strategy selects the subsequence image of a certain position of breathing as final compensation image.This method need to be extracted accurately
Breath signal and artificial selection breathing compensation position, however there are very big difference in amplitude, the period etc. of different patient respiratories, and
In medicine dynamic image checking process, other than respiratory movement ingredient, move into there is also heartbeat etc. is other
Point, the complicated and diversified selection for moving into the accurate extraction for being divided into breath signal and breathing compensation position brings extreme difficulties.
In conclusion when existing respiration motion compensation method is used for medicine dynamic image, exist needs it is artificial participate in, vulnerable to
The problems such as subjective factor influences.
Summary of the invention
In order to solve the above-mentioned technical problems, the present invention provides a kind of medicine dynamic images based on sparse subspace clustering
Respiration motion compensation method, when adopting said method, only needs using medicine dynamic image data to be compensated as input, i.e., exportable
Sequence image after respiration motion compensation, for further clinician's diagnostic application.The method has easy to implement, nothing
The advantages of needing manual intervention, can reduce breathing bring adverse effect, improve the validity of respiration motion compensation.
The present invention adopts the following technical scheme: a kind of medicine dynamic image respiratory movement based on sparse subspace clustering is mended
Compensation method includes the following steps:
Step S1 acquires medicine dynamic image, pre-processes to dynamic image;
In the step, noise reduction process mainly is carried out to medicine dynamic image to the pretreatment operation of dynamic image.For example,
Using speckle noise anisotropy parameter algorithm (Speckle Reducing Anisotropic Diffusion, SRAD) to super
Sound contrastographic picture carries out image noise reduction, using non-local mean algorithm (Non-Local Means, NLM) to enhanced CT Dynamic Graph
Picture or Contrast-enhanced MRI dynamic image carry out image noise reduction.
Step S2 selects sparse subspace representation method, and determines that the sparse subspace representation method is corresponding and punish
Penalize item constraint method;
In the step, common sparse subspace representation method can be rarefaction representation, and low-rank representation and ridge regression indicate,
Its corresponding punishment item constraint is respectively l1 norm constraint, nuclear norm constraint and F norm constraint.Indicate dilute in step S3 with Z
Dredging subspace indicates coefficient matrix, then punishes that item constraint is written as f (Z), the constraint of corresponding rarefaction representation coefficient matrix is successively
For ‖ Z ‖1, ‖ Z ‖*With ‖ Z ‖F, by constraining penalty term, may make rarefaction representation matrix to meet different characteristics, with expression
The relevant information of data itself.Such as increase nuclear norm constraint, can mining data space structural information.In selection generally with number
It is foundation according to the information of expression and the complexity of calculating.To make full use of dynamic image spatial structural form and time series to believe
Breath indicates in selection ridge regression, on the basis of carrying out F norm constraint to penalty term, is further added by Laplce's regular terms
Constraint;
Laplce's regular terms constraint representation are as follows:
Wherein, ziFor the corresponding expression coefficient of the i-th frame image, zjFor the corresponding expression coefficient of jth frame image, n is Dynamic Graph
As totalframes, w is weight coefficient, expression formula are as follows:
Wherein, d is Size of Neighborhood coefficient, for adjusting the size of neighborhood.
Step S3, the sparse subspace representation method selected according to the step S2 and the penalty term constrained procedure determined,
Based on an optimization object function, solves and obtain the sparse subspace expression coefficient matrix of the medicine dynamic image, it is described sparse
Subspace indicates that coefficient matrix is used to indicate the relationship in medicine dynamic image between frame and frame;
The optimization object function indicates are as follows:
Wherein, α and β is preset balance parameters, for controlling the ratio of penalty term.
The step S3 further uses alternating direction multipliers method (Alternating Direction Method of
Multipliers, ADMM) optimization object function is solved, the sparse subspace for obtaining the medicine dynamic image indicates system
Matrix number Z.
Step S4 constructs measuring similarity matrix G according to the sparse coefficient matrix Z that the step S3 is obtained;It is described similar
Spend metric matrix G make are as follows:
Step S5, according to the measuring similarity matrix G that the step S4 is constructed, using Spectral Clustering to described similar
Degree metric matrix G is cut, and the cluster result of the medicine dynamic image is obtained;
In the step, the measuring similarity matrix is cut using normalization blanking method commonly used in the prior art
It cuts, the number which clusters as needed, medicine dynamic image corresponding to the measuring similarity matrix is cut into not
Same cluster.
Step S6 selects maximum kind image for respiration motion compensation image according to the cluster result that the step S5 is obtained;
The maximum kind image is the image class that amount of images is most in the cluster result.
The above method proposed by the present invention can be widely used in the respiration motion compensation problem of medicine dynamic image.?
In this method, medicine dynamic image is pre-processed first, to reduce influence of the picture noise to compensation effect;Then sufficiently
Using dynamic image spatial structural form and time serial message, solve to obtain the medicine dynamic in a manner of punishing item constraint
The sparse subspace of image indicates coefficient matrix, and the sparse subspace indicates that coefficient matrix illustrates frame in medicine dynamic image
Relationship between frame;Indicate that coefficient matrix constructs measuring similarity matrix then according to thin subspace;Recycle spectral clustering side
The cluster result of method acquisition medicine dynamic image;Finally, using maximum image class result as final respiration motion compensation image.
The present invention has the advantage that: 1, providing a kind of medicine dynamic image respiratory movement based on sparse subspace clustering
Compensation method;
2, the method is easy to implement, is not necessarily to manual intervention;
3, the method is only needed using medicine dynamic image data to be compensated as input, i.e., exportable through respiration motion compensation
Sequence image afterwards improves the validity of respiration motion compensation, reduces breathing bring adverse effect, compensated sequence chart
As for further clinician's diagnostic application.
Detailed description of the invention
Fig. 1 is that the present invention is based on the processes that the medicine of sparse subspace clustering dynamic enhances image respiration motion compensation method
Figure.
Fig. 2 is the ultrasonic contrast image for not carrying out respiration motion compensation.
Fig. 3 is the ultrasonic contrast image respiration motion compensation result of the method for the present invention.
Specific embodiment
The preferred embodiment of the present invention is described in detail with reference to the accompanying drawing, so that advantages and features of the invention energy
It is easier to be readily appreciated by one skilled in the art, so as to make a clearer definition of the protection scope of the present invention.
As shown in Figure 1, the present invention is based on the medicine of sparse subspace clustering dynamic enhancing image respiration motion compensation methods
The following steps are included:
Step S1 acquires ultrasonic contrast dynamic image, pre-processes to dynamic image;
In the step, noise reduction process mainly is carried out to medicine dynamic image to the pretreatment operation of dynamic image.Using
Speckle noise anisotropy parameter algorithm (Speckle Reducing Anisotropic Diffusion, SRAD) makes ultrasound
Shadow image carries out image noise reduction.
Step S2 selects sparse subspace representation method, and determines that the sparse subspace representation method is corresponding and punish
Penalize item constraint method;
In the step, common sparse subspace representation method can be rarefaction representation, and low-rank representation and ridge regression indicate,
Its corresponding punishment item constraint is respectively l1 norm constraint, nuclear norm constraint and F norm constraint.Indicate dilute in step S3 with Z
Dredging subspace indicates coefficient matrix, then punishes that item constraint is written as f (Z), the constraint of corresponding rarefaction representation coefficient matrix is successively
For ‖ Z ‖1, ‖ Z ‖*With ‖ Z ‖F, by constraining penalty term, may make rarefaction representation matrix to meet different characteristics, with expression
The relevant information of data itself.Such as increase nuclear norm constraint, can mining data space structural information.In selection generally with number
It is foundation according to the information of expression and the complexity of calculating.In this example, to make full use of dynamic image spatial structural form
And time serial message, it is indicated in selection ridge regression, on the basis of carrying out F norm constraint to penalty term, is further added by Laplce
Regular termsConstraint;
Laplce's regular terms constraint representation are as follows:
Wherein, ziFor the corresponding expression coefficient of the i-th frame image, zjFor the corresponding expression coefficient of jth frame image, n is Dynamic Graph
As totalframes, w is weight coefficient, expression formula are as follows:
Wherein, d is Size of Neighborhood coefficient, for adjusting the size of neighborhood.Size of Neighborhood coefficient value is 2 in this example,
The weight coefficient matrix is then are as follows:
Step S3, the sparse subspace representation method selected according to the step S2 and the penalty term constrained procedure determined,
Based on an optimization object function, solves and obtain the sparse subspace expression coefficient matrix of the medicine dynamic image, it is described sparse
Subspace indicates that coefficient matrix is used to indicate the relationship in medicine dynamic image between frame and frame;
The optimization object function indicates are as follows:
Wherein, α and β is preset balance parameters, for controlling the ratio of penalty term.α and β value is respectively in this example
0.01 and 15.
The step S3 further uses alternating direction multipliers method (Alternating Direction Method of
Multipliers, ADMM) optimization object function is solved, the sparse subspace for obtaining the medicine dynamic image indicates system
Matrix number Z.
Step S4 constructs measuring similarity matrix G according to the sparse coefficient matrix Z that the step S3 is obtained;
The measuring similarity matrix G make are as follows:
Step S5, according to the measuring similarity matrix G that the step S4 is constructed, using Spectral Clustering to described similar
Degree metric matrix G is cut, and the cluster result of the medicine dynamic image is obtained;
In the step, the measuring similarity matrix is cut using normalization blanking method commonly used in the prior art
It cuts, the number which clusters as needed, medicine dynamic image corresponding to the measuring similarity matrix is cut into not
Same cluster.In this example, clusters number is set as 20.
Step S6 selects maximum kind image for respiration motion compensation image according to the cluster result that the step S5 is obtained;
The maximum kind image is the image class that amount of images is most in the cluster result.In this example, medical ultrasonic contrastographic picture
Compensation before and after Comparative result it is as shown in Figure 2 and Figure 3.
Not limited to this, any change or replacement expected without creative work should all be covered in guarantor of the invention
Within the scope of shield.Therefore, protection scope of the present invention should be determined by the scope of protection defined in the claims.
Claims (5)
1. a kind of medicine dynamic image respiration motion compensation method based on sparse subspace clustering, which is characterized in that including such as
Lower step:
Step S1 acquires medicine dynamic image, preprocess method is determined, to reduce influence of the picture noise to compensation effect;
Step S2 selects sparse subspace representation method, and determines the corresponding penalty term of the sparse subspace representation method
Constrained procedure;
Step S3 is asked according to sparse subspace representation method and identified penalty term constrained procedure that the step S2 is selected
The sparse subspace that solution obtains the medicine dynamic image indicates coefficient matrix, and the sparse subspace indicates that coefficient matrix is used for
Indicate the relationship in medicine dynamic image between frame and frame;
Step S4, the sparse subspace obtained according to the step S3 indicate that coefficient matrix is handled, construct measuring similarity
Matrix;
Step S5 obtains medicine dynamic image using Spectral Clustering according to the measuring similarity matrix that the step S4 is obtained
Cluster result;
Step S6 obtains the cluster result of medicine dynamic image according to the step S5, selects final respiration motion compensation image.
2. the medicine dynamic image respiration motion compensation method according to claim 1 based on sparse subspace clustering,
It is characterized in that, in the step (1), the pretreatment operation of dynamic image mainly carries out at noise reduction medicine dynamic image
Reason.
3. the medicine dynamic image respiration motion compensation method according to claim 1 based on sparse subspace clustering,
It is characterized in that, sparse subspace representation method described in the step (2) is rarefaction representation or low-rank representation or ridge regression table
Show.
4. the medicine dynamic image respiration motion compensation method according to claim 1 based on sparse subspace clustering,
It is characterized in that, solves to obtain the sparse subspace of the medicine dynamic image in a manner of punishing item constraint in the step (2)
Indicate coefficient matrix, the sparse subspace indicates that coefficient matrix illustrates the relationship in medicine dynamic image between frame and frame.
5. the medicine dynamic image respiration motion compensation method according to claim 1 based on sparse subspace clustering,
It is characterized in that, in the step (5), the measuring similarity matrix is cut using normalization blanking method, the algorithm root
According to the number that needs cluster, medicine dynamic image corresponding to the measuring similarity matrix is cut into different clusters.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811112741.3A CN109325925A (en) | 2018-09-25 | 2018-09-25 | A kind of medicine dynamic image respiration motion compensation method based on sparse subspace clustering |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811112741.3A CN109325925A (en) | 2018-09-25 | 2018-09-25 | A kind of medicine dynamic image respiration motion compensation method based on sparse subspace clustering |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109325925A true CN109325925A (en) | 2019-02-12 |
Family
ID=65265854
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811112741.3A Pending CN109325925A (en) | 2018-09-25 | 2018-09-25 | A kind of medicine dynamic image respiration motion compensation method based on sparse subspace clustering |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109325925A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117854139A (en) * | 2024-03-07 | 2024-04-09 | 中国人民解放军总医院第三医学中心 | Open angle glaucoma recognition method, medium and system based on sparse selection |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108460412A (en) * | 2018-02-11 | 2018-08-28 | 北京盛安同力科技开发有限公司 | A kind of image classification method based on subspace joint sparse low-rank Structure learning |
-
2018
- 2018-09-25 CN CN201811112741.3A patent/CN109325925A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108460412A (en) * | 2018-02-11 | 2018-08-28 | 北京盛安同力科技开发有限公司 | A kind of image classification method based on subspace joint sparse low-rank Structure learning |
Non-Patent Citations (2)
Title |
---|
ELHAMIFAR E, VIDAL R: "《Sparse subspace clustering: Algorithm, theory, and applications》", 《IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE》 * |
KAIZHI WU,ET AL: "《Respiratory compensation in contrast enhanced ultrasound using image clustering》", 《PROC. SPIE 10580, MEDICAL IMAGING 2018: ULTRASONIC IMAGING AND TOMOGRAPHY》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117854139A (en) * | 2024-03-07 | 2024-04-09 | 中国人民解放军总医院第三医学中心 | Open angle glaucoma recognition method, medium and system based on sparse selection |
CN117854139B (en) * | 2024-03-07 | 2024-05-28 | 中国人民解放军总医院第三医学中心 | Open angle glaucoma recognition method, medium and system based on sparse selection |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10699410B2 (en) | Automatic change detection in medical images | |
US10304198B2 (en) | Automatic medical image retrieval | |
US11051715B2 (en) | Image processing apparatus, image processing method, and recording medium recording same | |
US10803354B2 (en) | Cross-modality image synthesis | |
CN110232691A (en) | A kind of dividing method of multi-modal CT images | |
WO2011114733A1 (en) | Medical image conversion device, method, and program | |
Kamiya et al. | Automated segmentation of psoas major muscle in X-ray CT images by use of a shape model: preliminary study | |
JP5415245B2 (en) | MEDICAL IMAGE DISPLAY DEVICE, METHOD, AND PROGRAM | |
JP2005078176A (en) | Non-rigid body registration method between a plurality of images | |
US11715212B2 (en) | Heatmap and atlas | |
KR20190137283A (en) | Method for producing medical image and device for producing medical image | |
WO2022165812A1 (en) | System for synthesizing real-time image by using optical body surface motion signals | |
CN115830016B (en) | Medical image registration model training method and equipment | |
Dai et al. | Deep learning‐based motion tracking using ultrasound images | |
Fischer et al. | An MR-based model for cardio-respiratory motion compensation of overlays in X-ray fluoroscopy | |
Sharma et al. | Heart disease prediction using convolutional neural network | |
US20100030572A1 (en) | Temporal registration of medical data | |
Kamiya et al. | Automated segmentation of recuts abdominis muscle using shape model in X-ray CT images | |
Zhao et al. | Residual-atrous attention network for lumbosacral plexus segmentation with MR image | |
US20120306862A1 (en) | Image processing device, method and program | |
CN109325925A (en) | A kind of medicine dynamic image respiration motion compensation method based on sparse subspace clustering | |
Tobon-Gomez et al. | A multimodal database for the 1 st cardiac motion analysis challenge | |
KR20120000729A (en) | Position tracking method for vascular treatment micro robot through registration between x ray image and ct 3d model | |
CN109767410A (en) | A kind of lung CT and MRI image blending algorithm | |
Zakkaroff et al. | Patient-specific coronary blood supply territories for quantitative perfusion analysis |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190212 |
|
WD01 | Invention patent application deemed withdrawn after publication |