CN105868537B - PET-CT Dynamic medicals image intelligence quantified system analysis and analysis method - Google Patents
PET-CT Dynamic medicals image intelligence quantified system analysis and analysis method Download PDFInfo
- Publication number
- CN105868537B CN105868537B CN201610173991.2A CN201610173991A CN105868537B CN 105868537 B CN105868537 B CN 105868537B CN 201610173991 A CN201610173991 A CN 201610173991A CN 105868537 B CN105868537 B CN 105868537B
- Authority
- CN
- China
- Prior art keywords
- medical image
- quantitative analysis
- user
- subsystem
- information
- 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.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 47
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 73
- 230000002285 radioactive effect Effects 0.000 claims abstract description 23
- 210000000056 organ Anatomy 0.000 claims abstract description 18
- 238000000034 method Methods 0.000 claims description 32
- 238000005457 optimization Methods 0.000 claims description 28
- 239000000700 radioactive tracer Substances 0.000 claims description 10
- 238000003745 diagnosis Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 5
- 238000013475 authorization Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000003384 imaging method Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000012879 PET imaging Methods 0.000 claims description 3
- 238000000205 computational method Methods 0.000 claims description 3
- 230000002068 genetic effect Effects 0.000 claims description 3
- 238000011017 operating method Methods 0.000 claims description 3
- 208000004605 Persistent Truncus Arteriosus Diseases 0.000 claims 3
- 208000037258 Truncus arteriosus Diseases 0.000 claims 3
- 229940079593 drug Drugs 0.000 claims 1
- 239000003814 drug Substances 0.000 claims 1
- 230000000149 penetrating effect Effects 0.000 claims 1
- 238000002156 mixing Methods 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 10
- 229910019142 PO4 Inorganic materials 0.000 description 5
- 210000001519 tissue Anatomy 0.000 description 5
- 210000002381 plasma Anatomy 0.000 description 4
- 238000012636 positron electron tomography Methods 0.000 description 4
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 description 3
- 238000002059 diagnostic imaging Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000008103 glucose Substances 0.000 description 3
- 230000004060 metabolic process Effects 0.000 description 3
- NBIIXXVUZAFLBC-UHFFFAOYSA-K phosphate Chemical compound [O-]P([O-])([O-])=O NBIIXXVUZAFLBC-UHFFFAOYSA-K 0.000 description 3
- 239000010452 phosphate Substances 0.000 description 3
- 241000039077 Copula Species 0.000 description 2
- 102000005548 Hexokinase Human genes 0.000 description 2
- 108700040460 Hexokinases Proteins 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000002591 computed tomography Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012905 input function Methods 0.000 description 2
- 230000037323 metabolic rate Effects 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 102000018711 Facilitative Glucose Transport Proteins Human genes 0.000 description 1
- 108091052347 Glucose transporter family Proteins 0.000 description 1
- 210000001015 abdomen Anatomy 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 210000001367 artery Anatomy 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000017531 blood circulation Effects 0.000 description 1
- 210000000170 cell membrane Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000002224 dissection Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 125000002791 glucosyl group Chemical group C1([C@H](O)[C@@H](O)[C@H](O)[C@H](O1)CO)* 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000036039 immunity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000035479 physiological effects, processes and functions Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 239000000376 reactant Substances 0.000 description 1
- 230000004797 therapeutic response Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
Classifications
-
- G06F19/321—
-
- G06F19/34—
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a kind of PET CT Dynamic medicals image intelligence quantified system analysis and analysis methods.Including medical image management subsystem, it is connected with medical service organ information system；Subscriber information management subsystem；Medical image quantitative analysis subsystem, is connected respectively with subscriber information management subsystem and medical image management subsystem；Medical image quantitative analysis subsystem is built-in with PET-CT quantified system analysis；Information exchange module, medical image management subsystem and medical image quantitative analysis subsystem are all connected with information exchange module；Information exchange module is showing quantitative analysis results；User terminal.It can realize that kinetic parameters optimize by mixing intelligent optimizing algorithm in the present invention.An accurately area-of-interest can be calculated by the time radioactive activity curve set being distributed based on dynamic characteristic in the present invention.System of the present invention provides personalized quantitative analysis service by updating and learning the experience of user.
Description
Technical field
The present invention relates to medical image service system, more particularly to a kind of PET-CT Dynamic medicals image intelligence quantitative analysis
System and analysis method.
Background technology
Dynamic image is namely based on the data acquisition system of voxel, carries out voxel grade modeling required time expense very significantly, and by
Influence of noise is big.The dynamic characteristic of PET dynamic images is studied, the quantitative target with physiologic meaning can be obtained.In recent years
Come, in order to improve the exactness and accuracy of diagnostic imaging, international image educational circles proposes image quantitative analysis method.Image is determined
Amount analysis enumerates the imaging means such as dissection, function, molecular image, by accurate image analysis, accurately reflect anatomy and
The relevant parameter of physiology.Image quantitative analysis method can be used for assessment therapeutic response, judge clinical prognosis etc., so as to serve
Patient's diagnosis and treatment and clinical research.Basically, this objective quantification method from image is better than traditional figure
As analysis, its level independent of observer can be avoided what observer in itself and between different observers judged lesion
Difference.
But existing medical image quantitative analysis method, specific aim is weaker, lacks the information system to different medical service organization
System is attached, and lacks the function being compared with patient's history's information, similar patient's medical information, intelligent weaker.Meanwhile
Workload is larger, and time cost is higher, and in the optimum choice of method, outcome evaluation analysis and historical information management etc.
The requirement of medical and technical staff and clinician cannot be met, and should not be promoted.
Invention content
The purpose of the present invention is to provide a kind of PET-CT Dynamic medicals image intelligence quantified system analysis and analysis method,
Medical and technical staff and clinician is helped more conveniently and efficiently to carry out medical imaging diagnosis.
Technical scheme is as follows：
A kind of PET-CT Dynamic medicals image intelligence quantified system analysis, including following component part：
Medical image management subsystem is connected with medical service organ information system, to obtain authorization message；Doctor
Medical service organ information system can be entered after authorization by learning image information management subsystem, and medical image therein is remembered
It records and stores quantitative analysis results and the medical information for authorizing patient is recorded；
Subscriber information management subsystem, for recording the personal identification of system user, and use habit to user and
The personal settings of quantitative analysis method are managed；Use habit include the analysis request of image, system performance, operation content,
Time sets and operating procedure；Quantitative analysis method personal settings include the use of permission, priority and password setting；
Medical image quantitative analysis subsystem, respectively with subscriber information management subsystem and medical image management subsystem
System is connected；Medical image quantitative analysis subsystem is built-in with PET-CT quantified system analysis, is quantified to provide PET-CT
Analysis method；Medical image quantitative analysis subsystem accesses content, access time, behaviour by user management subsystem to user
It is recorded as step, and is recorded to registration and using the user identity of software；Medical image quantitative analysis subsystem root
The Historical medical of the medical image, medical information and the same patient that are there is provided according to medical image management subsystem analyze knot
Fruit carries out quantitative analysis to the diagnosis of patient；And quantitative analysis results are transmitted to information exchange module；Medical image quantitatively divides
Analysis subsystem further includes Function Extension add module；
Information exchange module, medical image management subsystem and medical image quantitative analysis subsystem are all handed over information
Mutual module is connected；Information exchange module to show quantitative analysis results and the comparison situation with historical analysis result, and
Early warning threshold values is compareed, carries out early warning prompting；
User terminal；User terminal is connect by network with information exchange module；User waits to locate by user terminal uploads
The PET-CT presentation contents of reason select processing to be analyzed according to the mandate of medical service organ information system from the system
Image information；User terminal further includes method choice interface, and user is according to quantitative analysis function selection analysis method；User is whole
End can also be to user feedback analysis result.
A kind of quantitative analysis side realized using PET-CT Dynamic medicals image intelligence quantified system analysis as described above
Method includes the following steps：
Step 1, user believe by the pending PET-CT dynamic images of user terminal uploads or by medical image
Breath management subsystem via medical service organ information system mandate, is accessed and is treated in specified medical service organ information system
The PET-CT dynamic images of processing；
Step 2, user delineate the quantitative analysis target area in PET-CT dynamic images by user terminal manually, note
For region of interest ROI (Region of Interest)；
Step 3, medical image quantitative analysis subsystem are distributed according to the dynamic characteristic of ROI region, are calculated and are generated the area
The when m- radioactive activity curve set TACs (Time-Activity Curve, TAC) in domain；
Step 4, medical image quantitative analysis subsystem are lived using m- radioactivity during the completion pair of mixed type intelligent optimization method
Spend the tracer dynamics model parameter estimation of curve set；
Step 5, medical image quantitative analysis subsystem specify the quantitative target of output according to user, to estimates of parameters into
Row calculates and returns to analysis result to user by information exchange module；
Step 6, medical image quantitative analysis subsystem further include warning information library, retrieval warning information library, if this is fixed
Figureofmerit is more than the early warning threshold values of the index in warning information library, then generates early warning information and sent by information exchange module
User；
Step 7, medical image quantitative analysis subsystem pass through the record data of user management subsystem, update user's use
Custom and system setting.
Its further technical solution is：In the step 3, when m- radioactive activity curve set computational methods be：
Step 31, user delineate quantitative analysis target area on medical image, obtain area-of-interest；
In step 32, medical image in the region of interest, m- radioactivity work when one corresponding of each pixel
Write music line yi(i=1,2 ..., n), curve yiLength be ni, PET imaging times point
Curve yiWith imaging time point xiThere are following p rank multinomial regression relations：
yi=Xiβ+εi,εi~N (0, σ2I)
In above formula, β is the regression coefficient vector of a rank of (p+1) × 1, εiIt is additional gaussian noise item, XiIt is Fan get Meng
Regression matrix；Then yiConditional probability density function obey N (yi|Xiβ,σ2I normal distribution)；
Step 33, by when m- radioactive activity curve yi(i=1,2 ..., n) it is divided into M different classes；It introduces and closes
Copula{ m ∈ M }, then yiConditional probability density can be represented with a Finite mixture model, be based on polynomial regression
The curve cluster of mixed model；Mean value calculation is carried out by the curve to M Clustering, obtain area-of-interest when
M- radioactive activity curve set.
Its further technical solution is：In the step 4, mixed type intelligent optimization method is to construct artificial immunity net
The Stochastic Optimization Algorithms of network intelligent optimization algorithm or genetic Optimization Algorithm, it is nested with based on gradient optimization algorithm to make
With progress global optimization；Local optimum is carried out by the optimization algorithm declined based on gradient, improves search speed.
The method have the benefit that：
1st, the quantitative analysis method in the present invention is intelligent：Can kinetic model be realized by mixing intelligent optimizing algorithm
Parameter optimization.
2nd, the present invention carries out secondary splitting to the region of interest ROI that user delineates by hand.By being based on dynamic characteristic
The when m- radioactive activity curve set of distribution calculates, and obtains an accurately area-of-interest.
3rd, system of the present invention provides personalized quantitative analysis service, tool by updating and learning the experience of user
Body：
3.1st, the present invention can be authorized from the medical service organ information system of connection, intelligent Matching image and related doctor
Treat information.If successful match, medical image is uploaded again without user, system, which can be read, comes from medical service organ information system
The mandate medical image and relevant information of system.Image uplink time and flow are not only saved, but also enhances the safety of system；
3.2nd, the present invention can optimize quantified system analysis setting, such as recommend optimal quantitative analysis method according to evaluation criterion；
3.3rd, the present invention, which can compare, provides historical analysis as a result, providing warning information.As certain quantitative target is exceeded or variation
It is very big, prompt disease variation risk；
3.4th, the present invention can be according to user's use habit optimization system usage experience, and in subscriber information management subsystem
Memory retention.
Description of the drawings
Fig. 1 is the system construction drawing of the present invention.
Fig. 2 is medical image quantitative analysis subsystem information exchange principle figure.
Fig. 3 is three chamber four parameter model figures of FDG metabolism.
Specific embodiment
PET-CT (positron emission tomography-computed tomography) refers to that positive electron is sent out
Penetrate tomography-x-ray computer tomography instrument.
The present invention provides a kind of PET-CT medical images intelligence quantified system analysis and analysis method, is used to help medical technologies people
Member and clinician more conveniently and efficiently carry out medical imaging diagnosis.
Fig. 1 is the system construction drawing of the present invention.System of the present invention includes：
(1), medical image management subsystem is connected with medical service organ information system, and letter is authorized to obtain
Breath.Medical service organ includes the mechanisms such as hospital, area medical center, community medical service center.Medical image management
Subsystem to：
(1a), to the medical image into the system is authorized to be recorded and stores quantitative analysis results.
(1b), medical information is recorded.
(2), subscriber information management subsystem is used for：
(2a), the personal identification for recording system user, such as medical and technical staff, clinician；
(2b), the use habit of user is managed, including analyze the request of image, system performance, operation content, when
Between setting, operating procedure etc. improve user experience；
(2c), quantitative analysis method personal settings are managed, include the use of permission, priority, password setting etc..
(3), medical image quantitative analysis subsystem, respectively with subscriber information management subsystem and medical image management
Subsystem is all connected.Fig. 2 is medical image quantitative analysis subsystem information exchange principle figure.In medical image quantitative analysis
In system：
(3a), PET-CT quantified system analysis is built-in with, to provide PET-CT quantitative analysis methods；Medical image is determined
Amount analyzing subsystem also provides extension addition function, quantitative analysis method is contributed to enrich constantly perfect；
(3b), it is connect by carrying out communication with user management subsystem, accesses user content, access time, operation step
Suddenly it is recorded, and is recorded to registration and using the user identity of software, according to the use habit of user, personal identification etc.
Historical information carries out quantitative analysis；
(3c), the medical image provided according to medical image management subsystem, the historical analysis knot of same patient
Quantitative analysis results are transmitted to information exchange module by fruit etc..
(4), information exchange module, medical image management subsystem and medical image quantitative analysis subsystem all with letter
Breath interactive module is connected；Information exchange module carries out information exchange with user terminal：Information exchange module is used for：
(4a), display quantitative analysis results；
(4b), display are the same as the comparison situation of historical analysis result；
(4c), critical value display exception item progress early warning prompting is given for exception item.
(5), user terminal:User terminal is connect by network with information exchange module, and network can be wide area network, local
Net, Metropolitan Area Network (MAN) etc..User terminal can be used for realizing following functions：
(5a), user refer to medical and technical staff and clinician by user terminal；
(5b), user are by the pending PET-CT dynamic images content of user terminal uploads or according to medical services
The mandate of organizational information system accesses some pending PET-CT dynamic shadow in specified medical service organ information system
Picture；
(5c), user terminal providing method selection interface；User selects quantitative analysis method according to image modality, and user is whole
Hold the analysis result after user feedback analysis.
The invention also includes use determining for PET-CT Dynamic medicals image intelligence quantified system analysis realization as described above
Analysis method includes the following steps：
Step 1, user believe by the pending PET-CT dynamic images of user terminal uploads or by medical image
Breath management subsystem via medical service organ information system mandate, is accessed and is treated in specified medical service organ information system
The PET-CT dynamic images of processing；
Step 2, user are delineated the quantitative analysis target area in PET-CT dynamic images, are remembered manually by user terminal
For region of interest ROI (Region of Interest)；
Step 3, medical image quantitative analysis subsystem are distributed according to the dynamic characteristic of region of interest ROI, calculate life
Into the radioactive concentration curve set in the region；
Step 4, the tracer dynamics using m- radioactive activity curve set during the completion pair of mixed type intelligent optimization method
Model parameter estimation；
Step 5, the quantitative target that output is specified according to user calculate estimates of parameters and pass through information exchange module and return
Analysis result is returned to user；
Step 6, medical image quantitative analysis subsystem further include warning information library, retrieval warning information library, if this is fixed
Figureofmerit is more than the early warning threshold values of the index in warning information library, then generates early warning information and send user；
Step 7, medical image quantitative analysis subsystem pass through the record data of user management subsystem, update user's use
Custom and system setting.
In above-mentioned steps 3, the dynamic characteristic distribution of region ROI, based on FDG tracer dynamics models.Tracer
Kinetic model is the rate that reactant, product and bioprocess are described with one group of mathematical model, and quantitative measurement is in blood plasma
And the time in tissue-Radioactive variation situation is to assess physiological parameter.M- radioactive activity curve when wherein, in blood plasma
As the input function of model, the multiple output function observation of the activity curves of more area-of-interests as model.
Below with common 18F-FDG tracer dynamics model example.18F-FDG full name are the fluoro- 2- deoxidations-D- Portugals of 2-
Grape sugar, can participate in body metabolism with positive charge as common glucose sugar, be the tracer used in PET clinical examinations.
Fig. 3 is three chamber four parameter model figures of FDG metabolism.In figure 3, what the chamber on the left side represented is blood space, with abdomen actively
Arteries and veins TAC is as the input function C of modelB(t)；Intermediate chamber represents the FDG in tissue, is denoted as CE(t)；The chamber table on the right
Show that FDG is phosphorylated later product C in the tissueM(t).In the FDG models of three chambers, four parameter, k1Represent the FDG in blood plasma
Into the rate of tissue, the factors such as value and quantity, the velocity of blood flow of glucose transporter albumen of cell membrane surface are related；k2Generation
FDG in table organization returns to the rate of blood plasma, influence factor and k1It is identical；k3The FDG in histocyte is represented in hexokinase
Under the action of be phosphorylated to the rate of FDG-6-PO4；k4It represents FDG-6-PO4 and FDG is reverted under the action of phosphate
Rate, k3And k4Numerical value and the quantity of hexokinase and phosphate it is related with activity.The phosphate concentration of different tissues
It is not consistent, but concentration is all very low in most of tissues.In addition, there are another parameter f, table in tracer kinetics research
What is shown is the influence coefficient that the radioactive activity in PET image in blood images surrounding tissue.
Step 3 also calculates the when m- radioactive activity curve set TACs of region of interest ROI, and m- radioactivity is lived immediately
It writes music the set of line (TAC, Time-Activity Curve).Medical image is usually made of region of interest and background area, relatively
For background area, region of interest includes important diagnostic message.The quantitative analysis target area that user delineates by hand obtains
One coarse area-of-interest.The TAC curves for delineating ROI acquisition PET images are collection model observations, build mode input
The committed step of function and the pretreatment stage of model parameter estimation are completed from four-dimensional dynamic image to two-dimensional time-radiation
The information transfer process of property activity curve collection.It is big due to manually delineating domain error, it is accordingly required in particular to be distributed based on dynamic characteristic
Dynamic image region automatic Extraction Algorithm help.
When m- radioactive activity curve set computational methods be：
Step 31, user delineate quantitative analysis target area on medical image, obtain area-of-interest；
In step 32, medical image in the region of interest, m- radioactivity work when one corresponding of each pixel
Write music line yi(i=1,2 ..., n), curve yiLength be ni, PET imaging times point
Curve yiWith imaging time point xiThere are following p rank multinomial regression relations：
yi=Xiβ+εi,εi~N (0, σ2I)
In above formula, β is the regression coefficient vector of a rank of (p+1) × 1, εiIt is additional gaussian noise item, XiIt is that model obtains
Cover regression matrix；Then yiConditional probability density function obey N (yi|Xiβ,σ2I normal distribution)；
Step 33, by when m- radioactive activity curve yi(i=1,2 ..., n) it is divided into M different classes；It introduces and closes
Copula{ m ∈ M }, then yiConditional probability density can be represented with a Finite mixture model, when m- radioactive activity it is bent
Line yi(i=1,2 ..., n) it is the curve cluster based on polynomial regression mixed model；Pass through the curve to M Clustering
Statistical calculations are carried out, such as mean value calculation obtains the when m- radioactive activity curve set TACs of area-of-interest.
In above-mentioned steps 4, the tracer power to radioactive concentration curve set is completed using mixed type intelligent optimization method
Learn model parameter estimation.I.e. the parameter Estimation optimization method of tracer dynamics model is mixing intelligent optimizing method, specifically
By constructing Stochastic Optimization Algorithms, such as artificial immune network intelligent optimization algorithm, genetic Optimization Algorithm, with being declined based on gradient
The mixing nesting of optimization algorithm uses, and gives full play to the global optimization ability of Stochastic Optimization Algorithms, while by being based under gradient
The optimization algorithm of drop improves the search speed of local optimum.Later while build time-radioactive activity curve set TACs, make
With Multiple Parameter Estimation Methods, the kinetic parameters of m- radioactive activity curve set TACs during acquisition.
In above-mentioned steps 5, medical image quantitative analysis subsystem specifies the quantitative target of output according to user, to parameter
Estimated value calculates and returns to analysis result to user by information exchange module.In the present embodiment, the output that user specifies is determined
Figureofmerit is glucose metabolic rate parameter, uses Ki (m)It represents.When m- radioactive activity curve set TACs kinetic parameters
For { k1 (m),k2 (m),k3 (m),k4 (m),f(m)；M=1,2 ..., M }, calculate glucose metabolic rate parameter Ki (m)：
Ki (m)=k1 (m)·k3 (m)/(k2 (m)+k3 (m))。
What has been described above is only a preferred embodiment of the present invention, and the present invention is not limited to above example.It is appreciated that this
The other improvements and change that field technology personnel directly export or associate without departing from the spirit and concept in the present invention
Change, be considered as being included within protection scope of the present invention.
Claims (4)
1. a kind of PET-CT Dynamic medicals image intelligence quantified system analysis, which is characterized in that including following component part：
Medical image management subsystem is connected with medical service organ information system, to obtain authorization message；Medicine shadow
As information management subsystem can record simultaneously medical image therein into medical service organ information system after authorization
Store quantitative analysis results；
Subscriber information management subsystem, for recording the personal identification of system user, and use habit to user and quantitative
The personal settings of analysis method are managed；Use habit includes analyzing the request of image, system performance, operation content, time
Setting and operating procedure；Quantitative analysis method personal settings include the use of permission, priority and password setting；
Medical image quantitative analysis subsystem, respectively with subscriber information management subsystem and medical image management subsystem phase
Connection；Medical image quantitative analysis subsystem is built-in with PET-CT quantified system analysis, to provide PET-CT quantitative analyses
Method；Medical image quantitative analysis subsystem accesses user content, access time, operation step by user management subsystem
Suddenly it is recorded, and is recorded to registration and using the user identity of software；Medical image quantitative analysis subsystem is according to doctor
Historical medical's analysis result of medical image, medical information and same patient that image information management subsystem is provided are learned,
Quantitative analysis is carried out to the diagnosis of patient；And quantitative analysis results are transmitted to information exchange module；Medical image quantitative analysis
Subsystem further includes Function Extension add module；
Information exchange module, medical image management subsystem and medical image quantitative analysis subsystem all with information exchange mould
Block is connected；Information exchange module compares to show quantitative analysis results and the comparison situation with historical analysis result
Threshold value of warning carries out early warning prompting；
User terminal；User terminal is connect by network with information exchange module；User is pending by user terminal uploads
PET-CT presentation contents or the shadow for selecting processing to be analyzed from the system according to the mandate of medical service organ information system
As information；User terminal further includes method choice interface, and user is according to quantitative analysis function selection analysis method；User terminal is also
It can be to user feedback analysis result.
2. a kind of quantitative point realized using PET-CT Dynamic medicals image intelligence quantified system analysis as described in claim 1
Analysis method, which is characterized in that include the following steps：
Step 1, user by the pending PET-CT dynamic images of user terminal uploads or pass through medical image information pipe
Subsystem is managed, via medical service organ information system mandate, is accessed pending in specified medical service organ information system
PET-CT dynamic images；
Step 2, user delineate the quantitative analysis target area in PET-CT dynamic images by user terminal manually, are denoted as sense
Interest region ROI；
Step 3, medical image quantitative analysis subsystem are distributed according to the dynamic characteristic of ROI region, are calculated and are generated the region
When m- radioactive activity curve set TACs；
Step 4, medical image quantitative analysis subsystem are bent using m- radioactive activity during the completion pair of mixed type intelligent optimization method
The tracer dynamics model parameter estimation of line collection TACs；
Step 5, medical image quantitative analysis subsystem specify the quantitative target of output according to user, and estimates of parameters is counted
It calculates and analysis result is returned to user by information exchange module；
Step 6, medical image quantitative analysis subsystem further include warning information library, retrieval warning information library, if the quantitative finger
Mark is more than the early warning threshold values of the index in warning information library, then generates early warning information and send use by information exchange module
Family；
Step 7, medical image quantitative analysis subsystem update user's use habit by the record data of user management subsystem
And system setting.
3. quantitative analysis method as claimed in claim 2, it is characterised in that：In the step 3, when m- radioactive activity it is bent
The computational methods of line collection TACs are：
Step 31, user delineate quantitative analysis target area on medical image, obtain area-of-interest；
In step 32, medical image in the region of interest, each pixel m- radioactive activity song when one corresponding
Line yi(i=1,2 ..., n), curve yiLength be ni, PET imaging times point
Curve yiWith imaging time point xiThere are following p rank multinomial regression relations：
yi=Xiβ+εi,εi~N (0, σ2I)
In above formula, β is the regression coefficient vector of a rank of (p+1) × 1, εiIt is additional gaussian noise item, XiIt is that Fan get Meng is returned
Matrix；Then yiConditional probability density function obey N (yi|Xiβ,σ2I normal distribution)；I represents unit matrix；
Step 33, by when m- radioactive activity curve yi(i=1,2 ..., n) it is divided into M different classes；Introduce associations{ m ∈ M }, then yiConditional probability density can be represented with a Finite mixture model, be based on polynomial regression mix
The curve cluster of model；Mean value calculation is carried out by the curve to M Clustering, obtain area-of-interest when m- put
Penetrating property activity curve collection.
4. quantitative analysis method as claimed in claim 2, it is characterised in that：In the step 4, mixed type intelligent optimization method
Artificial immune network intelligent optimization algorithm or the Stochastic Optimization Algorithms of genetic Optimization Algorithm to be constructed, with declining based on gradient
The nested of optimization algorithm uses, and carries out global optimization；Local optimum is carried out by the optimization algorithm declined based on gradient, raising is searched
Suo Sudu.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610173991.2A CN105868537B (en) | 2016-03-24 | 2016-03-24 | PET-CT Dynamic medicals image intelligence quantified system analysis and analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610173991.2A CN105868537B (en) | 2016-03-24 | 2016-03-24 | PET-CT Dynamic medicals image intelligence quantified system analysis and analysis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105868537A CN105868537A (en) | 2016-08-17 |
| CN105868537B true CN105868537B (en) | 2018-06-08 |
Family
ID=56624913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610173991.2A Active CN105868537B (en) | 2016-03-24 | 2016-03-24 | PET-CT Dynamic medicals image intelligence quantified system analysis and analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105868537B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107544736A (en) * | 2017-05-31 | 2018-01-05 | 上海联影医疗科技有限公司 | Medical image processing method and equipment |
| CN109686433B (en) * | 2018-12-28 | 2022-12-20 | 上海联影医疗科技股份有限公司 | Medical image equipment setting method, device, equipment and non-transient storage medium |
| CN110796180B (en) * | 2019-10-12 | 2022-06-07 | 吉林大学 | Model training system and method based on artificial intelligence |
| CN112052885B (en) * | 2020-08-20 | 2023-07-18 | 沈阳智核医疗科技有限公司 | Image processing method, device, equipment and PET-CT system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101203747A (en) * | 2004-01-21 | 2008-06-18 | 医软信息科技有限公司 | Method and system for intelligent qualitative and quantitative analysis of digital radiography softcopy reading |
| CN104939852A (en) * | 2015-06-17 | 2015-09-30 | 西安电子科技大学 | Fluorescence three-dimensional imaging tumor available receptor density quantitative method based on PET guide irradiation |
-
2016
- 2016-03-24 CN CN201610173991.2A patent/CN105868537B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101203747A (en) * | 2004-01-21 | 2008-06-18 | 医软信息科技有限公司 | Method and system for intelligent qualitative and quantitative analysis of digital radiography softcopy reading |
| CN104939852A (en) * | 2015-06-17 | 2015-09-30 | 西安电子科技大学 | Fluorescence three-dimensional imaging tumor available receptor density quantitative method based on PET guide irradiation |
Non-Patent Citations (2)
| Title |
|---|
| PET/CT的原理及临床应用;王家富，等.;《亚太传统医药》;20100930;第6卷(第9期);第171-172页 * |
| 新型PET/CT分子影像动力学建模技术在肺癌诊断中的应用;刘丽，等.;《中国数字医学》;20151231;第10卷(第4期);第5-8页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105868537A (en) | 2016-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11869202B2 (en) | Method and system for processing multi-modality image | |
| US10282588B2 (en) | Image-based tumor phenotyping with machine learning from synthetic data | |
| RU2684173C2 (en) | System and method for automatic treatment planning | |
| CN110364238B (en) | Machine learning based contrast management | |
| US20210151187A1 (en) | Data-Driven Estimation of Predictive Digital Twin Models from Medical Data | |
| US7606405B2 (en) | Dynamic tumor diagnostic and treatment system | |
| JP2022025095A (en) | System and method for translation of medical imaging using machine learning | |
| Xue et al. | Direct estimation of regional wall thicknesses via residual recurrent neural network | |
| US20100142774A1 (en) | method, a system, and an apparatus for using and processing multidimensional data | |
| WO2015163089A1 (en) | Medical image information system, medical image information processing method, and program | |
| US20140341449A1 (en) | Computer system and method for atlas-based consensual and consistent contouring of medical images | |
| CN105868537B (en) | PET-CT Dynamic medicals image intelligence quantified system analysis and analysis method | |
| US8423596B2 (en) | Methods of multivariate data cluster separation and visualization | |
| Mohammed et al. | Novel Crow Swarm Optimization Algorithm and Selection Approach for Optimal Deep Learning COVID‐19 Diagnostic Model | |
| Mazzanti et al. | Imaging, health record, and artificial intelligence: hype or hope? | |
| EP3570288A1 (en) | Method for obtaining at least one feature of interest | |
| Fei et al. | An overview of healthcare data analytics with applications to the COVID-19 pandemic | |
| WO2006119340A2 (en) | Dynamic tumor diagnostic and treatment system | |
| CN117116485A (en) | Data processing method, device, computing equipment and storage medium | |
| Lachinov et al. | Learning spatio-temporal model of disease progression with NeuralODEs from longitudinal volumetric data | |
| Wang et al. | ICA-Unet: An improved U-net network for brown adipose tissue segmentation | |
| JPWO2019208130A1 (en) | Medical document creation support devices, methods and programs, trained models, and learning devices, methods and programs | |
| Ngo et al. | Federated Fuzzy Clustering for Decentralized Incomplete Longitudinal Behavioral Data | |
| EP3667674A1 (en) | Method and system for evaluating images of different patients, computer program and electronically readable storage medium | |
| CN117238484B (en) | Intelligent diagnosis and treatment device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20160817 Assignee: JIANGSU SINOGRAM MEDICAL TECHNOLOGY Co.,Ltd. Assignor: AFFILIATED HOSPITAL OF JIANGNAN University Contract record no.: X2023980054637 Denomination of invention: PET-CT dynamic medical imaging intelligent quantitative analysis system and analysis method Granted publication date: 20180608 License type: Common License Record date: 20231228 |