CN109087284A - A kind of cardiovascular cannula Image-aided detection device and detection method - Google Patents

A kind of cardiovascular cannula Image-aided detection device and detection method Download PDF

Info

Publication number
CN109087284A
CN109087284A CN201810753415.4A CN201810753415A CN109087284A CN 109087284 A CN109087284 A CN 109087284A CN 201810753415 A CN201810753415 A CN 201810753415A CN 109087284 A CN109087284 A CN 109087284A
Authority
CN
China
Prior art keywords
data
image
module
vector
cardiovascular
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
Application number
CN201810753415.4A
Other languages
Chinese (zh)
Inventor
李川
刘卫金
邓以川
梁开运
何云
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing Cardiovascular Hospital Hong Wah Zhonglian Co Ltd
Original Assignee
Chongqing Cardiovascular Hospital Hong Wah Zhonglian Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chongqing Cardiovascular Hospital Hong Wah Zhonglian Co Ltd filed Critical Chongqing Cardiovascular Hospital Hong Wah Zhonglian Co Ltd
Priority to CN201810753415.4A priority Critical patent/CN109087284A/en
Publication of CN109087284A publication Critical patent/CN109087284A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30048—Heart; Cardiac
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular

Abstract

The invention belongs to the field of medical instrument technology, disclosing a kind of cardiovascular cannula Image-aided detection device and detection method, cardiovascular cannula Image-aided detection device includes: image capture module, heart rate detection module, respiratory rate detection module, central control module, categorization module, data management module, data memory module, display module.The present invention establishes new optimization CNN model by categorization module, then boundary graph and segmentation figure are overlapped and constitute complete organization chart, compensating for output image resolution ratio reduces bring image information loss, and institutional framework reduction degree is high, and image display effect is more preferable；Different Data quality assessment models is established using purpose according to the data characteristics and data of different phase by data management module simultaneously, ensure to carry out the quality of data accurate and effective assessment, overcomes the shortcomings that lacking effective Data quality assessment model in the prior art.

Description

A kind of cardiovascular cannula Image-aided detection device and detection method
Technical field
The invention belongs to the field of medical instrument technology more particularly to a kind of cardiovascular cannula Image-aided detection device and inspections Survey method.
Background technique
Currently, the prior art commonly used in the trade is such that
Heart is a hollow flesh sexual organ, positioned at the middle part in thoracic cavity, is divided into left and right two chambers by an interval, each Chamber is divided into ventricle two parts of superposed atrium and lower part again.Atrium is collected into heart blood, and ventricular ejection goes out the heart.Ventricle Inlet and outlet have a valve, guarantee blood one-way flow.Human body is under different physiological status, the metabolism of each organ-tissue It is horizontal different, it is also different to the needs of blood flow.Cardiovascular activity can change the heart and arrange blood in the case where the nerve and body fluid of body are reconciled Amount and peripheral resistance, coordinate the Blood flow distribution between each organ-tissue, to meet each organ-tissue to the needs of blood flow.So And it is existing when angiogenesis lesion, intravascular then can include different type tissue, and the detection and classification to these tissues are also It relies primarily on manually, it is very time-consuming；Effective assessment models are lacked to the quality of data of cardiovascular remote supervision system simultaneously； The Data Quality Control Techniques of different phase are done things in his own way, and are not integrated effectively.
In conclusion problem of the existing technology is:
It is existing when angiogenesis lesion, it is intravascular then can include different type tissue, to these tissue detection and Classification also relies primarily on manually, very time-consuming；The quality of data shortage of cardiovascular remote supervision system is effectively commented simultaneously Estimate model.
Conventional color image contour extraction method is when extracting object boundary vulnerable to the interference of initial profile point and convergence rate Slowly, big so as to cause the color image profile noise of extraction, influence the effect of image segmentation.
Existing shape similarity often has the least mean-square error and geometry of probability statistics algorithm, characteristic value with recognition methods The Weighted Average Algorithm etc. of external appearance characteristic necessary condition.Although achieving certain efficiency, there is also some shortcomings: algorithm The matching of realization process and visual discrimination is not intuitive；Algorithm is complicated, causes data processing amount big, and operating cost is high；Algorithm Evenness analysis causes the variation of important geometrical characteristic in figure to the influence of overall similarity, and Stability and veracity is caused to be deposited In certain deviation.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of cardiovascular cannula Image-aided detection device and inspections Survey method.
The invention is realized in this way a kind of cardiovascular cannula Image-aided detection method, the cardiovascular cannula image Aided detection method includes:
Acquire user's angiocardiogram data, wherein calculate user with Sobel operator or based on Color Space Clustering method The predicted value of angiocardiogram objective contour:
Image object profile is considered as to the collection { dl being made of N number of unit line segmenti}1,2 ..., N, for i=1,2 ..., N；
In C0In find and dliCorresponding position, according to C0The tangent line of middle corresponding position is as dliSampled reference value, it is raw At primary collection；Constantly assemble to known optimum solution direction according to state transition model guidance particle, avoids standard grain The method degenerated in sub- filtering realizes particle state transfer, and calculates the corresponding profile point set of each particle；According to building Vertical observation model calculates particle weights；The parameter dl obtained with the weighted average calculation current iteration of particle collectionj (i)=(kj (i), bj (i))；
IfWherein, take ε=0.5, then To using the weighted average of particle collection as dliThe estimation of parameter；To the surpriseization part in the image removal figure after estimation；It establishes The mathematical model of two figures establishes eigenmatrix corresponding with figure by the complete Vector Groups of description figure, calculates adjacent two The angle on side；Calculate the minimum distance between two figures；Enhancement processing is carried out to calculated result；To painstaking effort tube edge of different shapes Shape contour images carry out similarity detection；
Pass through the fractional lower-order ambiguity function for the digital modulation signals x (t) that heart rate detection module embedsDetect user's heart rate data information；Wherein, when τ is Prolong offset, f is Doppler frequency shift, 0 < a, b < α/2, x*(t) conjugation for indicating x (t), when x (t) is real signal, x (t)< p > =| x (t) |< p >sgn(x(t))；When x (t) is time multiplexed signal, [x (t)]< p >=| x (t) |p-1x*(t)；
Organization type classification is carried out to the angiocardiogram of acquisition；
The Cardiovascular data quality of acquisition is managed；The image and detection data of acquisition are stored；
Angiocardiogram and detection data content to acquisition are shown.
Further, state transition model avoids the method degenerated in standard particle filtering from including:
It is located at and calculates i-th of unit line segment l of image outlineiParameterWhen, particle collection is when iteration to s walks
Wherein Θs (i)For straight line parameter set, Ws (i)For particle weights, Es (i)It is calculated for Snake energy function model Profile validity estimate, be the foundation of granular Weights Computing；Based on conditions above, local optimum particle are as follows:
Global optimum's particle are as follows:
PSO state transition model is as follows:
Wherein rk, rk1, rk2, rb, rb1, rb2Equal Normal Distribution, and
Further, the side length of the mathematical model polygon of foundation and adjacent angle are by one vector S of construction counterclockwise1It indicates more Side shape:
S1=(l1,α1,l2, α2…lN-1,αN-1,lN,αN)；
S1There are mapping relations one by one with the polygon, indicates unrelated with corner initial order；
Complete Vector Groups have 2N vector S counterclockwise1、S2……S2N-1、S2NHave with polygon and reflects one by one Relationship is penetrated, a complete Vector Groups of the polygon is constituted, is expressed as follows:
S1=(l1,α1,l2, α2…lN-1,αN-1,lN,αN)；
S2=(α1,l2, α2…lN-1,αN-1,lN,αN,l1)；
……
S2N-1=(lN,αN,l1,α1,l2, α2…lN-1,αN-1)；
S2N=(αN,l1,α1,l2, α2…lN-1,αN-1,lN)；
With matrix SEIt indicates complete vector, and defines SEFor the eigenmatrix of the polygon, SEIt is expressed as follows:
The Euclidean distance of most like vector and maximum phase and coefficient are specific in acquisition source figure and targeted graphical eigenmatrix Include:
Firstly, establishing the eigenmatrix P of source figure P and targeted graphical Q respectively counterclockwiseEAnd QE:
PE=[P1 T P2 T … P2N-1 T P2N T]；
QE=[Q1 T Q2 T … Q2N-1 T Q2N T]；
Euclidean distance formula d (x, y) and included angle cosine formula sim (x, y) are as follows:
With d (x, y) and it is the basis sim (x, y), redefines two matrix Ds and S, make:
Find out the minimum value in D and S；
Eu is enabled respectivelye=min { Dij, 1≤i≤j=2N；Sime=max { Sij, 1≤i≤j=2N；
Then the eigenmatrix of needle directional structure vectorical structure figure P and Q, the above-mentioned calculation method of repetition find out two features in order again Minimum value Eu in matrix between most complete vectorcAnd Simc；
Finally enable Eu=min { Eue, Euc}；
Sim=min { Sime, Simc}；
Eu and Sim be two figure of P, Q correspond to most like vector Euclidean distance and it is maximum mutually and coefficient；
The enhancement of calculated result is handled and includes:
Initial vector is carried out once to repeatedly deformation, on the basis of with adjacent corner sequence structure initial vector, then The geometrical characteristic for adding figure, using the adjacent corner of order of addition than as new initial vector；Initial vector is carried out Multiple nonlinear processing is once arrived, carries out evolution processing using by initial vector；
Multiple similarity calculation is carried out to deformed initial vector, finally by weighted average value, with Euclidean distance Eu It is as follows with the evaluation formula mutually with coefficient S im:
N is the number of vector deformation, k in above formulaiFor weight coefficient, EuiAnd SimiVector is European after deforming for i-th Distance, Eu (P, Q) are the evaluation of Euclidean distance, n=4, kiTake 0.25.
Further, image classification method includes:
Step 1 obtains multiple marked IVOCT images；
Step 2 establishes IVOCT image pattern collection, and the IVOCT image pattern collection is divided into training sample set and test specimens This collection；
Step 3, building convolutional neural networks structure；
Step 4 is trained the convolutional neural networks using the training sample set, to obtain CNN model；
The test sample collection is inputted the CNN model by step 5, obtains the corresponding organization type figure of different tissues；
The step 2 includes:
The IVOCT image marked to each Zhang Suoshu carries out the transformation of diversified forms respectively, after obtaining multiple transformation Image, each in multiple described transformed images is set as a sample；Wherein, the transformation packet of the diversified forms It includes and cuts out, translates, overturning, rotating, deforming and one of gray-value variation or combination；
The multiple sample is set as the IVOCT image pattern collection.
Further, Cardiovascular data qualitative data management method includes:
(1) integrality of the QRS complex for the ECG data that data acquisition phase is extracted, data storage and management stage Data integrality, consistency, accuracy and timeliness and the data in data process&analysis stage complicated classification degree Quality testing is carried out, judges whether the quality of data complies with standard；
(2) when data acquire one or several in rank, data storage and management stage and data process&analysis stage Quality of data when not being inconsistent standardization, take corresponding control measure so that each stage data fit standard；For quality Data up to standard provide the suggestion of most suitable disaggregated model and data prediction.
Another object of the present invention is to provide a kind of calculating for realizing the cardiovascular cannula Image-aided detection method Machine program.
Another object of the present invention is to provide a kind of information for realizing the cardiovascular cannula Image-aided detection method Data processing terminal.
Another object of the present invention is to provide a kind of computer readable storage mediums, including instruction, when it is in computer When upper operation, so that computer executes the cardiovascular cannula Image-aided detection method.
Another object of the present invention is to provide a kind of painstaking effort for realizing the cardiovascular cannula Image-aided detection method Cannula Image-aided detection device, the cardiovascular cannula Image-aided detection device include:
Image capture module is connect with central control module, for acquiring user's angiocardiogram data by CT equipment；
Heart rate detection module is connect with central control module, for detecting user's heart rate data information；
Respiratory rate detection module, connect with central control module, for detecting user's respiratory rate data information；
Central control module, with image capture module, heart rate detection module, respiratory rate detection module, categorization module, number According to management module, data memory module, display module connection, worked normally for controlling modules；
Categorization module is connect with central control module, for carrying out organization type classification to the angiocardiogram of acquisition；
Data management module is connect with central control module, for being managed to the Cardiovascular data quality of acquisition；
Data memory module is connect with central control module, for acquisition image and detection data store；
Display module is connect with central control module, for showing the angiocardiogram and detection data content of acquisition.
Another object of the present invention is to provide a kind of hearts equipped with the cardiovascular cannula Image-aided detection device Vascular catheterization image detection equipment.
Advantages of the present invention and good effect are as follows:
The present invention establishes new optimization CNN model by categorization module, exports angiocarpy IVOCT by the first output end The structure of different types of tissue in image, second output terminal export the boundary profile of different types of tissue in angiocarpy, make The boundary and profile for obtaining each tissue are shown respectively, are then overlapped boundary graph and segmentation figure and are constituted complete knot of tissue Composition, compensating for output image resolution ratio reduces bring image information loss, and institutional framework reduction degree is high, image display effect More preferably；Different data matter is established using purpose according to the data characteristics and data of different phase by data management module simultaneously Measure assessment models, it is ensured that carry out accurate and effective assessment to the quality of data, overcome and lack effective data matter in the prior art The shortcomings that measuring assessment models.
It interferes and restrains vulnerable to initial profile point when extracting object boundary for conventional color image contour extraction method Speed is slow, big so as to cause the color image profile noise of extraction, the effect of image segmentation has been influenced, in consideration of it, of the invention Propose the color image contours extract algorithm based on particle filter.Firstly, providing the prediction of image outline and establishing two herein Dimension space, to make full use of image information；Then, the state transition model based on PSO optimization method is constructed, which promotees It is close to known optimum state into particle, the distribution of particle is improved, convergence rate is accelerated；Preferable quantitative description mesh Mark the effect of contours extract.
By simulation result it is found that its background is all more complicated in the color image of experiment, and come from experimental result It sees, predicts that the profile of color image is very important, if the profile predicted is stable and accurate, to subsequent profile It extracts and image segmentation is very helpful.The present invention effectively alleviates slow by the interference of initial profile point and convergence rate The problems such as, the image outline especially under Low SNR extracts, and the image object profile extracted is also highly satisfactory, Better than general contours extract algorithm.
The present invention improves machine to the visual discrimination effect of shape similarity, especially to manually being not easy to differentiate high similarity The difficult point of figure has very great help；Test pattern effect has stronger stability and reliability；Detection time is short, and operation is efficient, Implementation result is at low cost.The present invention only inquires the side of figure, reduces data processing amount.The present invention passes through constructing graphic Eigenmatrix, suitable decision criteria is chosen, and multiple enhancement nonlinear transformation is carried out to eigenmatrix element, with majority Value, multi-standard weighted average establish Measurement of Similarity, reached algorithm efficiently and have stronger stability.
The fractional lower-order ambiguity function for the digital modulation signals x (t) that the present invention is embedded by heart rate detection moduleDetect user's heart rate data letter Breath；It can get accurate data, provide foundation for post-processing.
Detailed description of the invention
Fig. 1 is cardiovascular cannula Image-aided structure of the detecting device block diagram provided in an embodiment of the present invention.
In figure: 1, image capture module；2, heart rate detection module；3, respiratory rate detection module；4, central control module； 5, categorization module；6, data management module；7, data memory module；8, display module.
Fig. 2 is categorization module image classification method flow chart provided in an embodiment of the present invention.
Specific embodiment
In order to further understand the content, features and effects of the present invention, the following examples are hereby given, and cooperate attached drawing Detailed description are as follows.
As shown in Figure 1, cardiovascular cannula Image-aided detection device provided in an embodiment of the present invention, comprising: Image Acquisition Module 1, heart rate detection module 2, respiratory rate detection module 3, central control module 4, categorization module 5, data management module 6, Data memory module 7, display module 8.
Image capture module 1 is connect with central control module 4, for acquiring user's angiocardiogram number by CT equipment According to；
Heart rate detection module 2 is connect with central control module 4, for detecting user's heart rate data information；
Respiratory rate detection module 3 is connect with central control module 4, for detecting user's respiratory rate data information；
Central control module 4, with image capture module 1, heart rate detection module 2, respiratory rate detection module 3, classification mould Block 5, data management module 6, data memory module 7, display module 8 connect, and work normally for controlling modules；
Categorization module 5 is connect with central control module 4, for carrying out organization type classification to the angiocardiogram of acquisition；
Data management module 6 is connect with central control module 4, for being managed to the Cardiovascular data quality of acquisition；
Data memory module 7 is connect with central control module 4, for acquisition image and detection data store；
Display module 8 is connect with central control module 4, for showing the angiocardiogram and detection data content of acquisition.
Such as Fig. 2, categorization module image classification method provided in an embodiment of the present invention is as follows:
S101, multiple marked IVOCT images are obtained；
S102, IVOCT image pattern collection is established, the IVOCT image pattern collection is divided into training sample set and test specimens This collection；
S103, building convolutional neural networks structure；
S104, the convolutional neural networks are trained using the training sample set, to obtain CNN model；
S105, the test sample collection is inputted into the CNN model, obtains the corresponding organization type figure of different tissues.
In step S102,
The IVOCT image marked to each Zhang Suoshu carries out the transformation of diversified forms respectively, after obtaining multiple transformation Image, each in multiple described transformed images is set as a sample；Wherein, the transformation packet of the diversified forms It includes and cuts out, translates, overturning, rotating, deforming and one of gray-value variation or combination；
The multiple sample is set as the IVOCT image pattern collection.
6 management method of data management module provided by the invention is as follows:
(1) integrality of the QRS complex for the ECG data that data acquisition phase is extracted, data storage and management stage Data integrality, consistency, accuracy and timeliness and the data in data process&analysis stage complicated classification degree Quality testing is carried out, judges whether the quality of data complies with standard；
(2) when data acquire one or several in rank, data storage and management stage and data process&analysis stage Quality of data when not being inconsistent standardization, take corresponding control measure so that each stage data fit standard；For quality Data up to standard provide the suggestion of most suitable disaggregated model and data prediction.
When the invention works, user's angiocardiogram data are acquired by image capture module 1；Pass through heart rate detection module 2 detection user's heart rate data information；User's respiratory rate data information is detected by respiratory rate detection module 3；Center control The angiocardiogram that module 4 dispatches 5 pairs of categorization module acquisitions carries out organization type classification；It is acquired by 6 Duis of data management module Cardiovascular data quality be managed；Then, the image and detection data acquired by 7 Duis of data memory module is deposited Storage；Finally, passing through the angiocardiogram and detection data content of the display acquisition of display module 8.
Below with reference to concrete analysis, the invention will be further described.
Cardiovascular cannula Image-aided detection method provided in an embodiment of the present invention, the cardiovascular cannula Image-aided inspection Survey method includes:
Acquire user's angiocardiogram data, wherein calculate user with Sobel operator or based on Color Space Clustering method The predicted value of angiocardiogram objective contour:
Image object profile is considered as to the collection { dl being made of N number of unit line segmenti}1,2 ..., N, for i=1,2 ..., N；
In C0In find and dliCorresponding position, according to C0The tangent line of middle corresponding position is as dliSampled reference value, it is raw At primary collection；Constantly assemble to known optimum solution direction according to state transition model guidance particle, avoids standard grain The method degenerated in sub- filtering realizes particle state transfer, and calculates the corresponding profile point set of each particle；According to building Vertical observation model calculates particle weights；The parameter dl obtained with the weighted average calculation current iteration of particle collectionj (i)=(kj (i), bj (i))；
If
Wherein, ε=0.5 is taken, then is obtained using the weighted average of particle collection as dliThe estimation of parameter；Image after estimation is gone Except the surpriseization part in figure；The mathematical model for establishing two figures is established corresponding with figure by the complete Vector Groups of description figure Eigenmatrix, calculate the angle on adjacent both sides；Calculate the minimum distance between two figures；Calculated result is carried out at enhancement Reason；Similarity detection is carried out to painstaking effort tube edge shape contour images of different shapes；
Pass through the fractional lower-order ambiguity function for the digital modulation signals x (t) that heart rate detection module embedsDetect user's heart rate data information；Wherein, τ is time delay Offset, f is Doppler frequency shift, 0 < a, b < α/2, x*(t) conjugation for indicating x (t), when x (t) is real signal, x (t)< p >=| x(t)|< p >sgn(x(t))；When x (t) is time multiplexed signal, [x (t)]< p >=| x (t) |p-1x*(t)；
Organization type classification is carried out to the angiocardiogram of acquisition；
The Cardiovascular data quality of acquisition is managed；The image and detection data of acquisition are stored；
Angiocardiogram and detection data content to acquisition are shown.
State transition model avoids the method degenerated in standard particle filtering from including:
It is located at and calculates i-th of unit line segment l of image outlineiParameterWhen, particle collection is when iteration to s walks
Wherein Θs (i)For straight line parameter set, Ws (i)For particle weights, Es (i)It is calculated for Snake energy function model Profile validity estimate, be the foundation of granular Weights Computing；Based on conditions above, local optimum particle are as follows:
Global optimum's particle are as follows:
PSO state transition model is as follows:
Wherein rk, rk1, rk2, rb, rb1, rb2Equal Normal Distribution, and
Further, the side length of the mathematical model polygon of foundation and adjacent angle are by one vector S of construction counterclockwise1It indicates more Side shape:
S1=(l1,α1,l2, α2…lN-1,αN-1,lN,αN)；
S1There are mapping relations one by one with the polygon, indicates unrelated with corner initial order；
Complete Vector Groups have 2N vector S counterclockwise1、S2……S2N-1、S2NHave with polygon and reflects one by one Relationship is penetrated, a complete Vector Groups of the polygon is constituted, is expressed as follows:
S1=(l1,α1,l2, α2…lN-1,αN-1,lN,αN)；
S2=(α1,l2, α2…lN-1,αN-1,lN,αN,l1)；
……
S2N-1=(lN,αN,l1,α1,l2, α2…lN-1,αN-1)；
S2N=(αN,l1,α1,l2, α2…lN-1,αN-1,lN)；
With matrix SEIt indicates complete vector, and defines SEFor the eigenmatrix of the polygon, SEIt is expressed as follows:
The Euclidean distance of most like vector and maximum phase and coefficient are specific in acquisition source figure and targeted graphical eigenmatrix Include:
Firstly, establishing the eigenmatrix P of source figure P and targeted graphical Q respectively counterclockwiseEAnd QE:
PE=[P1 T P2 T … P2N-1 T P2N T]；
QE=[Q1 T Q2 T … Q2N-1 T Q2N T]；
Euclidean distance formula d (x, y) and included angle cosine formula sim (x, y) are as follows:
With d (x, y) and it is the basis sim (x, y), redefines two matrix Ds and S, make:
Find out the minimum value in D and S；
Eu is enabled respectivelye=min { Dij, 1≤i≤j=2N；Sime=max { Sij, 1≤i≤j=2N；
Then the eigenmatrix of needle directional structure vectorical structure figure P and Q, the above-mentioned calculation method of repetition find out two features in order again Minimum value Eu in matrix between most complete vectorcAnd Simc；
Finally enable Eu=min { Eue, Euc}；
Sim=min { Sime, Simc}；
Eu and Sim be two figure of P, Q correspond to most like vector Euclidean distance and it is maximum mutually and coefficient；
The enhancement of calculated result is handled and includes:
Initial vector is carried out once to repeatedly deformation, on the basis of with adjacent corner sequence structure initial vector, then The geometrical characteristic for adding figure, using the adjacent corner of order of addition than as new initial vector；Initial vector is carried out Multiple nonlinear processing is once arrived, carries out evolution processing using by initial vector；
Multiple similarity calculation is carried out to deformed initial vector, finally by weighted average value, with Euclidean distance Eu It is as follows with the evaluation formula mutually with coefficient S im:
N is the number of vector deformation, k in above formulaiFor weight coefficient, EuiAnd SimiVector is European after deforming for i-th Distance, Eu (P, Q) are the evaluation of Euclidean distance, n=4, kiTake 0.25.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real It is existing.When using entirely or partly realizing in the form of a computer program product, the computer program product include one or Multiple computer instructions.When loading on computers or executing the computer program instructions, entirely or partly generate according to Process described in the embodiment of the present invention or function.The computer can be general purpose computer, special purpose computer, computer network Network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or from one Computer readable storage medium is transmitted to another computer readable storage medium, for example, the computer instruction can be from one A web-site, computer, server or data center pass through wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL) Or wireless (such as infrared, wireless, microwave etc.) mode is carried out to another web-site, computer, server or data center Transmission).The computer-readable storage medium can be any usable medium or include one that computer can access The data storage devices such as a or multiple usable mediums integrated server, data center.The usable medium can be magnetic Jie Matter, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk Solid State Disk (SSD)) etc..
The above is only the preferred embodiments of the present invention, and is not intended to limit the present invention in any form, Any simple modification made to the above embodiment according to the technical essence of the invention, equivalent variations and modification, belong to In the range of technical solution of the present invention.

Claims (10)

1. a kind of cardiovascular cannula Image-aided detection method, which is characterized in that cardiovascular cannula Image-aided detection side Method includes:
Acquire user's angiocardiogram data, wherein calculate user's painstaking effort with Sobel operator or based on Color Space Clustering method The predicted value of pipe image object profile:
；
Image object profile is considered as to the collection { dl being made of N number of unit line segmenti}1,2 ..., N, for i=1,2 ..., N；
In C0In find and dliCorresponding position, according to C0The tangent line of middle corresponding position is as dliSampled reference value, generate just Beginning particle collection；Constantly assemble to known optimum solution direction according to state transition model guidance particle, standard particle is avoided to filter The method degenerated during wave realizes particle state transfer, and calculates the corresponding profile point set of each particle；According to foundation Observation model calculates particle weights；The parameter dl obtained with the weighted average calculation current iteration of particle collectionj (i)=(kj (i), bj (i))；
If
Wherein, ε=0.5 is taken, then is obtained using the weighted average of particle collection as dliThe estimation of parameter；To the image removal after estimation Surpriseization part in figure；The mathematical model for establishing two figures is established corresponding with figure by the complete Vector Groups of description figure Eigenmatrix calculates the angle on adjacent both sides；Calculate the minimum distance between two figures；Calculated result is carried out at enhancement Reason；Similarity detection is carried out to painstaking effort tube edge shape contour images of different shapes；
Pass through the fractional lower-order ambiguity function for the digital modulation signals x (t) that heart rate detection module embeds
Detect user's heart rate data information；Wherein, τ is delay skew, and f is Doppler frequency shift, 0 < a, b < α/2, x*(t) x is indicated (t) conjugation, when x (t) is real signal, x (t)< p >=| x (t) |< p >sgn(x(t))；When x (t) is time multiplexed signal, [x (t)]< p >= |x(t)|p-1x*(t)；
Organization type classification is carried out to the angiocardiogram of acquisition；
The Cardiovascular data quality of acquisition is managed；The image and detection data of acquisition are stored；
Angiocardiogram and detection data content to acquisition are shown.
2. cardiovascular cannula Image-aided detection method as described in claim 1, which is characterized in that state transition model avoids marking The method degenerated in quasi particle filtering includes:
It is located at and calculates i-th of unit line segment l of image outlineiParameterWhen, particle collection is when iteration to s walks
Wherein Θs (i)For straight line parameter set, Ws (i)For particle weights, Es (i)The wheel being calculated for Snake energy function model Wide validity is estimated, and is the foundation of granular Weights Computing；Based on conditions above, local optimum particle are as follows:
Global optimum's particle are as follows:
PSO state transition model is as follows:
Wherein rk, rk1, rk2, rb, rb1, rb2Equal Normal Distribution, and
3. cardiovascular cannula Image-aided detection method as described in claim 1, which is characterized in that the mathematical model of foundation is used more The side length of side shape and adjacent angle are by one vector S of construction counterclockwise1Indicate polygon:
S1=(l1,α1,l2, α2…lN-1,αN-1,lN,αN)；
S1There are mapping relations one by one with the polygon, indicates unrelated with corner initial order；
Complete Vector Groups have 2N vector S counterclockwise1、S2……S2N-1、S2NThere is mapping one by one to close with polygon System, constitutes a complete Vector Groups of the polygon, is expressed as follows:
S1=(l1,α1,l2, α2…lN-1,αN-1,lN,αN)；
S2=(α1,l2, α2…lN-1,αN-1,lN,αN,l1)；
……
S2N-1=(lN,αN,l1,α1,l2, α2…lN-1,αN-1)；
S2N=(αN,l1,α1,l2, α2…lN-1,αN-1,lN)；
With matrix SEIt indicates complete vector, and defines SEFor the eigenmatrix of the polygon, SEIt is expressed as follows:
The Euclidean distance of most like vector and maximum phase and coefficient specifically include in acquisition source figure and targeted graphical eigenmatrix:
Firstly, establishing the eigenmatrix P of source figure P and targeted graphical Q respectively counterclockwiseEAnd QE:
PE=[P1 T P2 T … P2N-1 T P2N T]；
QE=[Q1 T Q2 T … Q2N-1 T Q2N T]；
Euclidean distance formula d (x, y) and included angle cosine formula sim (x, y) are as follows:
With d (x, y) and it is the basis sim (x, y), redefines two matrix Ds and S, make:
Find out the minimum value in D and S；
Eu is enabled respectivelye=min { Dij, 1≤i≤j=2N；Sime=max { Sij, 1≤i≤j=2N；
Then the eigenmatrix of needle directional structure vectorical structure figure P and Q, the above-mentioned calculation method of repetition find out two eigenmatrixes in order again In minimum value Eu between most complete vectorcAnd Simc；
Finally enable Eu=min { Eue, Euc}；
Sim=min { Sime, Simc}；
Eu and Sim be two figure of P, Q correspond to most like vector Euclidean distance and it is maximum mutually and coefficient；
The enhancement of calculated result is handled and includes:
Initial vector once on the basis of with adjacent corner sequence structure initial vector, then add to repeatedly deformation The geometrical characteristic of figure, using the adjacent corner of order of addition than as new initial vector；Initial vector is carried out primary To multiple nonlinear processing, evolution processing is carried out using by initial vector；
Multiple similarity calculation is carried out to deformed initial vector, finally by weighted average value, with Euclidean distance Eu and phase It is as follows with the evaluation formula of coefficient S im:
N is the number of vector deformation, k in above formulaiFor weight coefficient, EuiAnd SimiThe Euclidean distance of vector after being deformed for i-th, Eu (P, Q) is the evaluation of Euclidean distance, n=4, kiTake 0.25.
4. cardiovascular cannula Image-aided detection method as described in claim 1, which is characterized in that image classification method includes:
Step 1 obtains multiple marked IVOCT images；
Step 2 establishes IVOCT image pattern collection, and the IVOCT image pattern collection is divided into training sample set and test sample Collection；
Step 3, building convolutional neural networks structure；
Step 4 is trained the convolutional neural networks using the training sample set, to obtain CNN model；
The test sample collection is inputted the CNN model by step 5, obtains the corresponding organization type figure of different tissues；
The step 2 includes:
The IVOCT image marked to each Zhang Suoshu carries out the transformation of diversified forms respectively, to obtain multiple transformed figures Each in multiple described transformed images is set as a sample by picture；Wherein, the transformation of the diversified forms includes cutting One of sanction, translation, overturning, rotation, deformation and gray-value variation combine；
The multiple sample is set as the IVOCT image pattern collection.
5. cardiovascular cannula Image-aided detection method as described in claim 1, which is characterized in that Cardiovascular data qualitative data Management method includes:
(1) integrality of the QRS complex for the ECG data that data acquisition phase is extracted, the number in data storage and management stage According to integrality, consistency, accuracy and timeliness and the data in data process&analysis stage complicated classification degree carry out Quality testing, judges whether the quality of data complies with standard；
(2) when data acquire one or several the number in rank, data storage and management stage and data process&analysis stage When not being inconsistent standardization according to quality, take corresponding control measure so that each stage data fit standard；For requisite quality Data, the suggestion of most suitable disaggregated model and data prediction is provided.
6. a kind of computer program for realizing cardiovascular cannula Image-aided detection method described in Claims 1 to 5 any one.
7. a kind of realize at the information data of cardiovascular cannula Image-aided detection method described in Claims 1 to 5 any one Manage terminal.
8. a kind of computer readable storage medium, including instruction, when run on a computer, so that computer is executed as weighed Benefit requires cardiovascular cannula Image-aided detection method described in 1-5 any one.
9. a kind of cardiovascular cannula Image-aided detection for realizing cardiovascular cannula Image-aided detection method described in claim 1 Device, which is characterized in that the cardiovascular cannula Image-aided detection device includes:
Image capture module is connect with central control module, for acquiring user's angiocardiogram data by CT equipment；
Heart rate detection module is connect with central control module, for detecting user's heart rate data information；
Respiratory rate detection module, connect with central control module, for detecting user's respiratory rate data information；
Central control module, with image capture module, heart rate detection module, respiratory rate detection module, categorization module, data pipe Module, data memory module, display module connection are managed, is worked normally for controlling modules；
Categorization module is connect with central control module, for carrying out organization type classification to the angiocardiogram of acquisition；
Data management module is connect with central control module, for being managed to the Cardiovascular data quality of acquisition；
Data memory module is connect with central control module, for acquisition image and detection data store；
Display module is connect with central control module, for showing the angiocardiogram and detection data content of acquisition.
10. a kind of cardiovascular cannula Image detection equipped with cardiovascular cannula Image-aided detection device described in claim 9 Equipment.
CN201810753415.4A 2018-07-10 2018-07-10 A kind of cardiovascular cannula Image-aided detection device and detection method Pending CN109087284A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810753415.4A CN109087284A (en) 2018-07-10 2018-07-10 A kind of cardiovascular cannula Image-aided detection device and detection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810753415.4A CN109087284A (en) 2018-07-10 2018-07-10 A kind of cardiovascular cannula Image-aided detection device and detection method

Publications (1)

Publication Number Publication Date
CN109087284A true CN109087284A (en) 2018-12-25

Family

ID=64837518

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810753415.4A Pending CN109087284A (en) 2018-07-10 2018-07-10 A kind of cardiovascular cannula Image-aided detection device and detection method

Country Status (1)

Country Link
CN (1) CN109087284A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110664388A (en) * 2019-09-05 2020-01-10 广州市蜗牛互动科技有限公司 Heart rate detection method and device, storage medium and equipment
CN114496209A (en) * 2022-02-18 2022-05-13 青岛市中心血站 Blood donation intelligent decision method and system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202211686U (en) * 2011-05-13 2012-05-09 张政波 Cardiovascular function detection device
CN104462744A (en) * 2014-10-09 2015-03-25 广东工业大学 Data quality control method suitable for cardiovascular remote monitoring system
CN105354866A (en) * 2015-10-21 2016-02-24 郑州航空工业管理学院 Polygon contour similarity detection method
CN205072834U (en) * 2015-10-13 2016-03-09 杨士伟 Cardiovascular intubate image auxiliary device
CN106067004A (en) * 2016-05-30 2016-11-02 西安电子科技大学 The recognition methods of digital modulation signals under a kind of impulsive noise
CN107993229A (en) * 2017-12-15 2018-05-04 西安中科微光影像技术有限公司 A kind of tissue classification procedure and device based on cardiovascular IVOCT images

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202211686U (en) * 2011-05-13 2012-05-09 张政波 Cardiovascular function detection device
CN104462744A (en) * 2014-10-09 2015-03-25 广东工业大学 Data quality control method suitable for cardiovascular remote monitoring system
CN205072834U (en) * 2015-10-13 2016-03-09 杨士伟 Cardiovascular intubate image auxiliary device
CN105354866A (en) * 2015-10-21 2016-02-24 郑州航空工业管理学院 Polygon contour similarity detection method
CN106067004A (en) * 2016-05-30 2016-11-02 西安电子科技大学 The recognition methods of digital modulation signals under a kind of impulsive noise
CN107993229A (en) * 2017-12-15 2018-05-04 西安中科微光影像技术有限公司 A kind of tissue classification procedure and device based on cardiovascular IVOCT images

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
伊力哈木·亚尔买买提: "基于粒子滤波的彩色图像轮廓提取算法研究", 《计算机仿真》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110664388A (en) * 2019-09-05 2020-01-10 广州市蜗牛互动科技有限公司 Heart rate detection method and device, storage medium and equipment
CN110664388B (en) * 2019-09-05 2022-03-25 广州市蜗牛互动科技有限公司 Heart rate detection method and device, storage medium and equipment
CN114496209A (en) * 2022-02-18 2022-05-13 青岛市中心血站 Blood donation intelligent decision method and system
CN114496209B (en) * 2022-02-18 2022-09-27 青岛市中心血站 Intelligent decision-making method and system for blood donation

Similar Documents

Publication Publication Date Title
Chen et al. JAS-GAN: generative adversarial network based joint atrium and scar segmentations on unbalanced atrial targets
Lynch et al. Automatic segmentation of the left ventricle cavity and myocardium in MRI data
Sliman et al. Myocardial borders segmentation from cine MR images using bidirectional coupled parametric deformable models
CN110448319B (en) Blood flow velocity calculation method based on contrast image and coronary artery
CN110071913A (en) A kind of time series method for detecting abnormality based on unsupervised learning
Qin et al. Automatic segmentation of right ventricular ultrasound images using sparse matrix transform and a level set
Mazaheri et al. Echocardiography image segmentation: A survey
Xie et al. Anisotropic path searching for automatic neuron reconstruction
CN109087284A (en) A kind of cardiovascular cannula Image-aided detection device and detection method
CN101028187B (en) System and method for image based physiological monitoring of cardiovascular function
CN104143047A (en) Automatic tissue calibration method for IVUS gray-scale image
CN109767061A (en) A kind of appraisal procedure and device of electric energy meter crash rate
CN111297399A (en) Fetal heart positioning and fetal heart rate extraction method based on ultrasonic video
El Berbari et al. An automated myocardial segmentation in cardiac MRI
Qian et al. Segmentation of myocardium from cardiac MR images using a novel dynamic programming based segmentation method
CN108932973A (en) A kind of kidney case digitlization information management system and management method, terminal
EP2902969B1 (en) Image processing device, image processing method, and image processing program
Sharma et al. A novel solution of using deep learning for left ventricle detection: enhanced feature extraction
CN111325202A (en) Ultrasonic Doppler-based color spectrogram frequency spectrum blood flow detection method
CN107730537A (en) The Faint target detection and tracking of density filtering are assumed based on case particle probabilities
CN112258476A (en) Echocardiography myocardial abnormal motion mode analysis method, system and storage medium
El‐Rewaidy et al. Segmentation of the right ventricle in MRI images using a dual active shape model
CN116130090A (en) Ejection fraction measuring method and device, electronic device, and storage medium
Lalande et al. Evaluation of cardiac structure segmentation in cine magnetic resonance imaging
CN115349851A (en) Cardiac function diagnosis method based on atrioventricular plane pump model

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: 20181225

WD01 Invention patent application deemed withdrawn after publication