CN109864745A - A kind of novel risk of stroke appraisal procedure and system - Google Patents
A kind of novel risk of stroke appraisal procedure and system Download PDFInfo
- Publication number
- CN109864745A CN109864745A CN201910016245.6A CN201910016245A CN109864745A CN 109864745 A CN109864745 A CN 109864745A CN 201910016245 A CN201910016245 A CN 201910016245A CN 109864745 A CN109864745 A CN 109864745A
- Authority
- CN
- China
- Prior art keywords
- parameter
- assessment
- stroke
- information
- coupling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The present invention provides a kind of novel risk of stroke appraisal procedures, acquire target organism information, goal-based assessment parameter is generated after carrying out data processing to it, and goal-based assessment parameter is compared with criterion evaluation parameter, assessment report is generated according to comparison result and desired physiological information.Also provide a kind of novel risk of stroke assessment system: acquisition mould group acquires target organism information;Comparison module carries out data processing to target organism information and generates goal-based assessment parameter, and it is compared with criterion evaluation parameter;Assessment result generates mould group, and comparison result and desired physiological information are generated assessment report.Advantage is to establish the risk of stroke assessment models based on Bayesian network, introduce the target organism information acquired in real time, and it is analyzed with criterion evaluation parameter, the objectives physiologic information provided in conjunction with user generates corresponding assessment report, it helps user and doctor to carry out risk of stroke monitoring, improves the accuracy of risk of stroke assessment.
Description
Technical field
The present invention establishes the risk of stroke prediction model based on Bayesian network, and it is old to introduce near-infrared spectrum technique assessment
The brain oxygen signal of year crowd assesses the risk of stroke of user in conjunction with every biological information of user, belongs to artificial intelligence, occupies
The interleaving techniques field of family endowment and health medical treatment, more particularly to a kind of novel risk of stroke appraisal procedure and system.
Background technique
Aging of population is the great livelihood issues that China faces, and aging bring maximum public health problem is chronic
The problem of non-communicable diseases.Cerebral apoplexy is elderly population common clinical and frequently-occurring disease.Headstroke occurs suddenly, is to cause always
The major reason of year disability.It is counted, about 13,000,000 people of China's Mainland patients with cerebral apoplexy according to " Chinese cardiovascular disease report 2017 ", often
2,000,000 people of Nian Xinfa patient, wherein 2/3 disables, because of more than dead 160 ten thousand people of apoplexy.Cerebral Haemorrhage Invasion Rate, illness rate, the death rate
Respectively 2,46/,100,000 man-years, 1114.8/10 ten thousand man-years, 114.8/10 ten thousand man-years.Currently, family endowment has become main support
Old mode.For the elderly population of family endowment, stroke risk how is assessed, apoplexy probability is fundamentally reduced, actively carried out
Health-care and rehabilitation training, realize safe in ground aging, are the significant challenges currently faced.How as early as possible accurate
Assessing risk of stroke becomes urgent problem to be solved.
Summary of the invention
The present invention provides a kind of novel risk of stroke appraisal procedure and system, to realize in early stage accurate evaluation brain
Stroke risk.
To achieve the goals above, technical solution of the present invention provides a kind of novel risk of stroke appraisal procedure, packet
It includes: acquisition target organism information, wherein biological information includes desired physiological information and target brain blood oxygen signal.To target organism
Information generates goal-based assessment parameter after carrying out data processing, and goal-based assessment parameter is compared with criterion evaluation parameter, obtains
To comparison result.Assessment report is generated according to comparison result and desired physiological information.
As a preferred embodiment of the above technical solution, preferably, carrying out wavelet transformation analysis to target brain blood oxygen signal, frequency is obtained
Field result also obtains the average phase information in default wave band.Respectively to the average phase in frequency-domain result and the predetermined period
Position information carries out data processing, obtains the goal-based assessment parameter.
As a preferred embodiment of the above technical solution, preferably, carrying out wavelet amplitude analysis to frequency-domain result, wavelet amplitude is obtained
Mean value.To in default wave band average phase information carry out effective connectivity analysis, obtain the default wave band stiffness of coupling and
Couple direction.
As a preferred embodiment of the above technical solution, preferably, if wavelet amplitude mean value is greater than in the criterion evaluation parameter
Standard wavelet amplitude mean value is then exception.It is different if stiffness of coupling is beyond the standard stiffness of coupling in criterion evaluation parameter
Often.If it is different from the standard coupling direction in criterion evaluation parameter to couple direction, for exception.
As a preferred embodiment of the above technical solution, preferably, establishing Bayesian network according to desired physiological information and comparison result
Network model.The parameter in Bayesian network model is learnt by maximum likelihood algorithm, generates each node in network model
Node probability, the assessment report is generated according to node probability.
Technical solution of the present invention also provides a kind of novel risk of stroke assessment system, comprising: acquisition mould group, for adopting
Collect target organism information, wherein the biological information includes desired physiological information and target brain blood oxygen signal.Comparison module is used
Goal-based assessment parameter is generated after carrying out data processing to the target organism information of the acquisition mould group acquisition, and will be described
Goal-based assessment parameter is compared with criterion evaluation parameter, obtains comparison result.Assessment result generates mould group, for according to
Comparison module generates assessment report by comparing the obtained comparison result and the desired physiological information.
As a preferred embodiment of the above technical solution, preferably, comparison module, comprising: wavelet transformation analysis unit, for adopting
The target brain blood oxygen signal collected in the target organism information of mould group acquisition carries out wavelet transformation analysis, obtains frequency-domain result,
Also obtain the average phase information in default wave band.Goal-based assessment parameter acquiring unit, for the wavelet transformation analysis list
The average phase information in frequency-domain result and predetermined period that member is obtained by wavelet transformation analysis carries out data processing, obtains mesh
Mark assessment parameter.
As a preferred embodiment of the above technical solution, preferably, goal-based assessment parameter acquiring unit, comprising: wavelet amplitude analysis
Module, for carrying out wavelet amplitude point by the frequency-domain result that wavelet transformation analysis obtains to wavelet transformation analysis unit
Analysis, obtains wavelet amplitude mean value.Effective connectivity analysis module, for being obtained to wavelet transformation analysis unit by wavelet transformation analysis
Average phase information in the default wave band arrived carries out effective connectivity analysis, obtains stiffness of coupling and the coupling side of default wave band
To.
As a preferred embodiment of the above technical solution, preferably, comparison module, further includes: comparator, for wavelet amplitude point
The Standard wavelet amplitude mean value in the wavelet amplitude mean value and the criterion evaluation parameter that analysis module obtains is compared, if
The wavelet amplitude mean value is greater than the Standard wavelet amplitude mean value in the criterion evaluation parameter, then is exception.It is also used to effect
Standard stiffness of coupling in the stiffness of coupling and criterion evaluation parameter of the default wave band for answering linking parsing module analysis to obtain carries out
Compare, if stiffness of coupling exceeds the standard stiffness of coupling in the criterion evaluation parameter, for exception.It is further used for pairing effect
Compared in standard coupling direction in the coupling direction and criterion evaluation parameter of the default wave band that linking parsing module analysis obtains
Compared with if coupling direction is different from the standard coupling direction in the criterion evaluation parameter, for exception.
As a preferred embodiment of the above technical solution, preferably, assessment result generates mould group, comprising: model foundation unit is used for
Bayesian network model is established according to the comparison result that the desired physiological information of acquisition mould group acquisition and comparator generate.It comments
Report generation unit is estimated, in the Bayesian network model for establishing by maximum likelihood algorithm to model foundation unit
Parameter is learnt, and the node probability of each node in network model is generated, and generates the assessment report according to the node probability.
Technical solution of the present invention provides a kind of novel risk of stroke appraisal procedure, acquires target organism information, right
It generates goal-based assessment parameter after carrying out data processing, and goal-based assessment parameter is compared with criterion evaluation parameter, according to
Comparison result and desired physiological information generate assessment report.A kind of risk of stroke assessment system of type is also provided, mould group is acquired,
Acquire target organism information.Comparison module, to target organism information carry out data processing generate goal-based assessment parameter, and by its with
Criterion evaluation parameter is compared.Assessment result generates mould group, for comparison result and desired physiological information to be generated assessment report
It accuses.It is an advantage of the invention that establishing the risk of stroke assessment models based on Bayesian network, it is raw to introduce the target acquired in real time
Object information, and it is analyzed with criterion evaluation parameter, the objectives physiologic information generation provided in conjunction with user is accordingly commented
Estimate report, helps user and doctor to carry out risk of stroke monitoring, improve the accuracy of risk of stroke assessment.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to make one simply to introduce, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 provides the flow diagram one of novel risk of stroke appraisal procedure for the embodiment of the present invention.
Fig. 2 provides the flow diagram two of novel risk of stroke appraisal procedure for the embodiment of the present invention.
Fig. 3 is the schematic diagram for the light source probe template that the embodiment of the present invention uses.
Fig. 4 is the structural schematic diagram one of novel risk of stroke assessment system provided by the invention.
Fig. 5 is the structural schematic diagram of comparison module in Fig. 4.
Fig. 6 is the structural schematic diagram of goal-based assessment parameter acquiring unit in Fig. 5.
Fig. 7 is the structural schematic diagram that assessment result generates mould group in Fig. 4.
Fig. 8 is the structural schematic diagram two of novel risk of stroke assessment system provided by the invention.
Fig. 9 is the risk of stroke prediction model based on Bayesian network of technical solution of the present invention building.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, the technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Fig. 1 provides the flow diagram of novel risk of stroke appraisal procedure for the embodiment of the present invention, as shown in Figure 1,
Novel risk of stroke appraisal procedure provided in this embodiment, comprising:
Step 101, acquisition target organism information.
Wherein, biological information includes desired physiological information and target brain blood oxygen signal.
Step 102 generates goal-based assessment parameter after carrying out data processing to target organism information, and by goal-based assessment parameter
It is compared with criterion evaluation parameter, obtains comparison result.
Specifically, carrying out wavelet transformation analysis to target brain blood oxygen signal, frequency-domain result is obtained, is also obtained in default wave band
Average phase information.Data processing is carried out to the average phase information in the frequency-domain result and predetermined period respectively, is obtained
The goal-based assessment parameter.Wherein, wavelet amplitude analysis is carried out to frequency-domain result, obtains wavelet amplitude mean value.Also to default wave
Average phase information in section carries out effective connectivity analysis, obtains stiffness of coupling and the coupling direction of default wave band.
Further, if wavelet amplitude mean value is greater than the Standard wavelet amplitude mean value in criterion evaluation parameter, for exception.
If stiffness of coupling is beyond the standard stiffness of coupling in criterion evaluation parameter, for exception.If coupling direction and criterion evaluation parameter
In standard coupling direction it is different, then be exception.
Step 103 generates assessment report according to the comparison result and the desired physiological information.
Bayesian network model is established according to the desired physiological information and the comparison result;
The parameter in the Bayesian network model is learnt by maximum likelihood algorithm, generates the network model
In each node node probability, the assessment report is generated according to the node probability.
Technical solution that now present invention be described in more detail, presetting wave band is brain blood oxygen Section III section, Section IV section, such as Fig. 2 institute
Show:
Step 201, function near infrared spectrometer, light source probe template acquire brain blood oxygen signal.
Specifically, when it is 10-20 minutes a length of, gathered person pass through movement/immobile terminal (computer, mobile phone, tablet computer
Deng) risk of stroke assessment system into cloud computing server provides corresponding physiologic information (including age, gender, height, body
Weight, blood pressure, amount of exercise, sleep state etc.).Wherein, Fig. 3 is the schematic diagram for the light source probe template that the embodiment of the present invention uses,
This light source probe template is based on 10-10 system, adjacent according to the light source and probe of template arrangement function near infrared spectroscopy instrument
Light source and probe composition acquisition channel.
Step 202 carries out wavelet transformation analysis to brain blood oxygen signal.
Specifically, pre-set scaling sequence, so that the frequency f of wavelet transformation and relative 0.005-2Hz frequency
Rate is corresponding, and the transformational relation of wavelet transformation mesoscale s and frequency f meet:
Wherein, fc is wavelet center frequency, usually takes 1Hz;δ t is the sampling period;
Carry out continuous complex wavelet conversion process, formula are as follows:
Wherein, W (s, t) is the wavelet coefficient obtained after converting, is obtained by formula (2);T is time parameter;G (u) table
Show surveyed blood oxygen signal time series.Wherein, formula (1) obtains frequency-domain result, in formula (2), is made using morlet small echo
For morther wavelet, so the wavelet coefficient after wavelet transformation is plural number, to obtain following phase informations.
Wavelet transformation frequency-domain result of each channel blood oxygen signal on frequency domain is obtained as a result, and in Section III section, Section IV section
Average phase information in 2 specific default wave bands in time domain.
Step 203 carries out amplitude analysis to wavelet transformation frequency-domain result, obtains wavelet amplitude mean value.
Wavelet transform result on frequency domain is input in wavelet amplitude analysis module, passes through the trapezoidal integration side on frequency domain
Method obtains each channel blood oxygen signal in Section III section, the wavelet amplitude mean value of 2 specific bands of Section IV section.
Step 204, to average phase information carry out effective connectivity analysis, obtain Section III section, Section IV section stiffness of coupling and
Couple direction.
Average phase information input in 2 resulting Section III section, Section IV section specific bands in time domain is divided to effective connectivity
It analyses in module, establishes phase vibrator model of each blood oxygen signal in 2 Section III section, Section IV section specific bands:
φk(t)=ωk(t)+qk(φ1,φ2,…,φN,t)+ξk(t) (3)
Wherein, k=1 ..., N, N are the oscillator number in whole network, take N=2 in the present embodiment;ω describes phase increasing
Long rate, the intrinsic frequency of referred to as a certain vibrator model;ξ is white Gaussian noise;Q is coupling function, for describing between oscillator
Couple state.
Dynamic Bayesian inference is according to the time series χ={ x measuredl≡x(tl), it goes to speculate and entirely acquired
Coefficient of coup matrix c in journey between two time serieses(k)And noise matrix D.Calculation formula are as follows:
Wherein, M={ c(k),D};T1=lh, l=1 ..., L, h are the sampling interval, and L is total sampling number.pprior(M) it is
Priori probability density, pχ(M | χ) it is posterior probability density, physiology meaning is phase oscillator under the action of this coupled relation
Reach the maximum probability of stable state.To seek pχThe maximum value of (M | χ) can be by phase vibrator model weight when the sampling interval, h was sufficiently small
Structure are as follows:
Wherein,Specifically, when due to being in actually calculating
Between sequence, rather than ideal continuous signal, it is to preferably carry out subsequent programming that phase vibrator model, which is reconstructed,
It calculates, to generate more accurate assessment report.
Z in formula (5)nIt is noise function time series integral:
The then negative log-likelihood function S=-ln ζ (χ | M) of the probability density ζ (χ | M) of original signal is defined as:
By calculating the stationary point coordinate of negative log-likelihood function, the coupling under a certain state between phase vibrator model can be obtained
Collaboration number and noise matrix;Obtained Bayesian inference result is returned in negative log-likelihood function and continues recursive calculation, is referred to
Is there are not significant changes in the coefficient of coup matrix led, then shows that phase vibrator model reaches stable state, using at this time
Coupling function can solve the effective connectivity parameter between phase oscillator.Recursive procedure is as follows such as formula (8):
C=I-1r (11)
Wherein, formula (9) I is density matrix, and formula (10) r is temporary variable matrix, and formula (11) c is coefficient of coup square
Battle array.The termination condition of the above cyclic process is that recursive algorithm is not generating adjustment effect formula (12) to coefficient of coup matrix, i.e.,
∑(cprior-cpost)2/cpost 2< ε (12)
ε is minimum constant;Provide coefficient of coup matrix cprior=0, thus Iprior=0.
By the recurrence as a result, stiffness of coupling is defined as oscillator φkWith oscillator φiBetween coefficient of coup matrix Euclid
Norm:Coupling direction is used to distinguish oscillator φkWith oscillator φiBetween coupled relation, is defined as:
Wherein, D<i,k>∈ [- 1,1];If the D that formula (13) obtains<i,k>Coupling direction for canonical is φkIt is directed toward φi, it is negative
Then φiIt is directed toward φk。
More than, by Dynamic Bayesian inference estimation method, obtain per channel blood oxygen signal two-by-two in Section III section, Section IV
The stiffness of coupling of 2 specific bands of section and coupling direction.
Wavelet amplitude mean value, stiffness of coupling and coupling direction are compared with criterion evaluation parameter step 205 respectively,
Obtain comparison result.
Wherein, criterion evaluation parameter is the relevant parameter by being detected to healthy population.
Specifically, to the Standard wavelet amplitude in the wavelet amplitude in each channel of Section III wave band and criterion evaluation parameter into
Row compares, and the number for wavelet amplitude exception occur is denoted as a1 by the recording exceptional if beyond this range.Specifically, above-mentioned wave band
There is dry passage in if, above-mentioned relatively processing is carried out to the wavelet amplitude in each channel, is recorded if there is exception, to obtain
A1, further, following " abnormal number " are identical with this, and are repeated no more.
The stiffness of coupling of Section III wave band is compared with the standard stiffness of coupling range in criterion evaluation parameter, if super
Standard stiffness of coupling range then recording exceptional out, is denoted as b1 for the number for stiffness of coupling exception occur.
The coupling direction of Section III wave band is compared with the standard coupling direction in criterion evaluation parameter, as long as direction
It is different, then it is assumed that it is abnormal, the number for coupling direction exception occur is denoted as c1.
A1, b1 and c1 are summed, obtain S1, S1 is the total abnormal number of Section III wave band, and obtains S1 and specifically correspond to
Threat probability values P1;Specifically, the corresponding specific P1 value of S1 is obtained by corresponding look-up table.
Further, to the Standard wavelet amplitude in the wavelet amplitude in each channel of Section IV wave band and criterion evaluation parameter into
Row compares, and the number for wavelet amplitude exception occur is denoted as a2 by the recording exceptional if beyond this range.
The stiffness of coupling of Section IV wave band is compared with the standard stiffness of coupling range in criterion evaluation parameter, if exceeding
Standard stiffness of coupling range then recording exceptional, is denoted as b2 for the number for stiffness of coupling exception occur.
The coupling direction of Section IV wave band is compared with the standard coupling direction in criterion evaluation parameter, as long as direction is not
Together, then it is assumed that it is abnormal, the number for coupling direction exception occur is denoted as c2.
A2, b2 and c2 are summed, S2 is obtained, and obtains S2 specifically corresponding threat probability values P2;The corresponding tool of S2
The P2 value of body is obtained by corresponding look-up table.
The reason of selecting above-mentioned Section III wave band and Section IV wave band is that blood oxygen signal is in 2 Section III section, Section IV section certain waves
The feature of section respectively corresponds its specific physiological sources.Section III section: 0.052-0.145Hz, from muscle-derived activity;Section IV section,
0.021-0.052Hz, from neural sexuality, associated parameter data and blood oxygen signal of the blood oxygen signal in muscle-derived activity
Associated parameter data and cerebral apoplexy in neural sexuality have direct relation.
Step 206 arranges the physiological data in desired physiological information.
Wherein, the physiological data for including in desired physiological information includes but is not limited to: age, gender, height, weight, blood
Pressure, amount of exercise, sleep state.The important journey of possible related influence factor usually occurs according to each physiological data and cerebral apoplexy
Degree is ranked up.Usual significance level is clipped to low level by advanced are as follows: age, gender, apoplexy medical history, weight, blood pressure, movement
Amount sleep state, drinks, smokes, educational level, height.The above sequence is obtained according to big data, in practical arranging order process
In, the specific sequence sequencing of above-mentioned physiological data changes according to the difference of big data result.
Step 207 establishes Bayesian network model according to P1, P2 and desired physiological information using tabu search algorithm.
Bayesian Network Learning is carried out using tabu search algorithm (Tabu Search, TS), determines that Bayesian network respectively saves
The sequence of point (above-mentioned each physiological data to obtain in stroke impact factor step 206), establishes Bayesian network model, that is, exists
Under the premise of a given data sample set, finds one and match best network structure with training sample set.
Step 208 learns the probability of node each in Bayesian network model, generates assessment report.
Specifically, carrying out parameter learning using probability of the maximum likelihood estimation algorithm to each node, i.e., in sample data base
On plinth, seek the probability distribution of each node of network.Using network topology structure and training sample set and priori knowledge, shellfish is determined
Conditional probability density at this network model node of leaf, is denoted as P (xi)。
The joint probability distribution formula of any stochastic variable combination of Bayesian network model are as follows:
Wherein xparents(i)It is xiDirect precursor node joint father node.The corresponding variable of each node, each
Arrow expresses a kind of conditional probability, and formula (14) is gained probability.
The risk of stroke prediction model based on Bayesian network of foregoing description as shown in connection with fig. 9 can be seen that brain blood
Oxygen III wave band and IV wave band are directed toward in the arrow direction of cerebral apoplexy, are obtained including at least through the above steps 201 to step 205
Wavelet amplitude mean value, stiffness of coupling, coupling direction etc. related data informations.Arrow that can also be different from Fig. 9 is directed toward easily
Know and the physiological data in desired physiological information is arranged described in above-mentioned steps 206.Last above-mentioned data information summarizes
Afterwards through the above steps 207 establish Bayesian network model after generate final assessment report.
The present invention also provides a kind of novel risk of stroke assessment systems, as shown in Figure 4:
Mould group 41 is acquired, for acquiring target organism information, wherein the biological information includes desired physiological information and mesh
Mark brain blood oxygen signal.
Comparison module 42, the target organism information for acquiring to acquisition mould group 41 generate mesh after carrying out data processing
Mark assessment parameter, and the goal-based assessment parameter is compared with criterion evaluation parameter, obtain comparison result.
Assessment result generates mould group 43, for according to comparison module 42 by comparing the obtained comparison result and described
Desired physiological information generates assessment report.
As shown in figure 5, comparison module 42, comprising:
Wavelet transformation analysis unit 51, the target brain blood oxygen in target organism information for acquiring to acquisition mould group 41 are believed
Number wavelet transformation analysis is carried out, obtains frequency-domain result, also obtain the average phase information in default wave band.
Goal-based assessment parameter acquiring unit 52, for what is obtained to wavelet transformation analysis unit 51 by wavelet transformation analysis
Average phase information in frequency-domain result and predetermined period carries out data processing, obtains goal-based assessment parameter.
As shown in fig. 6, goal-based assessment parameter acquiring unit 52, including:
Wavelet amplitude analysis module 521, the frequency for being obtained to wavelet transformation analysis unit 51 by wavelet transformation analysis
Field result carries out wavelet amplitude analysis, obtains wavelet amplitude mean value.
Effective connectivity analysis module 522, for being obtained to the wavelet transformation analysis unit 51 by wavelet transformation analysis
The default wave band in average phase information carry out effective connectivity analysis, obtain the stiffness of coupling and coupling of the default wave band
Close direction.
Further, comparison module, further includes:
Comparator 523, wavelet amplitude mean value and criterion evaluation ginseng for being obtained to wavelet amplitude analysis module 521
Standard wavelet amplitude mean value in number is compared, if wavelet amplitude mean value is greater than the Standard wavelet amplitude in criterion evaluation parameter
Mean value is then exception.Stiffness of coupling and criterion evaluation for the default wave band that pairing effect linking parsing module analysis 522 obtains
Standard stiffness of coupling in parameter is compared, if stiffness of coupling exceeds the standard stiffness of coupling in the criterion evaluation parameter,
It is then abnormal.Coupling direction and the criterion evaluation parameter of obtained default wave band are analyzed for pairing effect linking parsing module 522
In standard coupling direction be compared, if coupling direction in criterion evaluation parameter standard couple direction it is different, be different
Often.
As shown in fig. 7, assessment result generates mould group 43, comprising:
Model foundation unit 71, what desired physiological information and comparator 523 for being acquired according to acquisition mould group 41 generated
Comparison result establishes Bayesian network model.
Assessment report generation unit 72, the pattra leaves for being established by maximum likelihood algorithm to model foundation unit 71
Parameter in this network model is learnt, and the node probability of each node in the network model is generated, general according to the node
Rate generates the assessment report.
Now further the risk of stroke assessment system novel to one kind provided by the invention is described in detail, specifically
, as shown in Figure 8.
Acquisition terminal 81, for acquiring the target brain blood oxygen signal and desired physiological information of tested person.Specifically, acquisition mesh
The terminal of mark brain blood oxygen signal can be function near infrared spectrometer, light source probe template;Acquire the terminal of desired physiological information
It can be movement/immobile terminal, corresponding target can also be obtained automatically from cloud by the identity information of tested person
Physiologic information.
Processor 82, target brain blood oxygen signal and desired physiological information for acquiring to acquisition terminal 81 carry out data point
Analysis handles and generates corresponding assessment report.
Specifically, processor includes: data management module 821, assessment models module 822, real time analysis module 823, report
Accuse generation module 824.
Data management module 821, target brain blood oxygen signal and desired physiological for acquiring to acquisition acquisition terminal 81 are believed
Breath is managed.Including but not limited to, record, data downloading and upload, basis importance relevant to cerebral apoplexy are to target
Physiologic information is ranked up.
Assessment models module 822, for using Bayesian network assessment models according to user's items physiologic information (including year
Age, gender, height, weight, blood pressure, amount of exercise, sleep state etc.) and brain blood oxygen signal data etc. carry out risk of stroke
Assessment.
Real time analysis module 823, for carrying out risk of stroke Factor minute to user according to risk of stroke assessment result
Analysis assessment.
Report generation module 824, for the assessment result and reality according to data management module 821, assessment models module 822
When analysis module 823 analysis result generate and the risk of stroke analysis result information that is presented to the user.
The present invention provides a kind of novel risk of stroke appraisal procedures, acquire target organism information, count to it
According to generation goal-based assessment parameter after processing, and goal-based assessment parameter is compared with criterion evaluation parameter, according to comparison result
Assessment report is generated with desired physiological information.Also provide a kind of risk of stroke assessment system of type: acquisition mould group acquisition target
Biological information;Comparison module carries out data processing to target organism information and generates goal-based assessment parameter, and by itself and criterion evaluation
Parameter is compared;Assessment result generates mould group, and comparison result and desired physiological information are generated assessment report.Advantage is to establish
Risk of stroke assessment models based on Bayesian network introduce the target organism information acquired in real time, and it are commented with standard
Estimate parameter to be analyzed, the objectives physiologic information provided in conjunction with user generates corresponding assessment report, helps user and doctor
Risk of stroke monitoring is carried out, the accuracy of risk of stroke assessment is improved.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations;To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement;And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (10)
1. a kind of novel risk of stroke appraisal procedure characterized by comprising
Acquire target organism information, wherein the biological information includes desired physiological information and target brain blood oxygen signal;
Goal-based assessment parameter is generated after carrying out data processing to the target organism information, and by the goal-based assessment parameter and is marked
Quasi- assessment parameter is compared, and obtains comparison result;
Assessment report is generated according to the comparison result and the desired physiological information.
2. novel risk of stroke appraisal procedure according to claim 1, which is characterized in that described raw to the target
Object information generates goal-based assessment parameter after carrying out data processing, and the goal-based assessment parameter is compared with criterion evaluation parameter
It is right, obtain comparison result, comprising:
Wavelet transformation analysis is carried out to the target brain blood oxygen signal, obtains frequency-domain result, also obtains being averaged in default wave band
Phase information;
Data processing is carried out to the average phase information in the frequency-domain result and the predetermined period respectively, obtains the target
Assess parameter.
3. novel risk of stroke appraisal procedure according to claim 2, which is characterized in that described respectively to the frequency
Average phase information in field result and the default wave band carries out data processing, obtains the goal-based assessment parameter, comprising:
Wavelet amplitude analysis is carried out to the frequency-domain result, obtains wavelet amplitude mean value;
Effective connectivity analysis is carried out to the average phase information in the default wave band, obtains the stiffness of coupling of the default wave band
With coupling direction.
4. any novel risk of stroke appraisal procedure according to claim 3, which is characterized in that described and by institute
It states goal-based assessment parameter to be compared with criterion evaluation parameter, obtains comparison result, comprising:
If the wavelet amplitude mean value is greater than the Standard wavelet amplitude mean value in the criterion evaluation parameter, for exception;
If the stiffness of coupling is beyond the standard stiffness of coupling in the criterion evaluation parameter, for exception;
If the coupling direction is different from the standard coupling direction in the criterion evaluation parameter, for exception.
5. novel risk of stroke appraisal procedure according to claim 4, which is characterized in that described according to the comparison
As a result assessment report is generated with the desired physiological information, comprising:
Bayesian network model is established according to the desired physiological information and the comparison result;
The parameter in the Bayesian network model is learnt by maximum likelihood algorithm, is generated each in the network model
The node probability of node generates the assessment report according to the node probability.
6. a kind of novel risk of stroke assessment system characterized by comprising
Mould group is acquired, for acquiring target organism information, wherein the biological information includes desired physiological information and target brain blood
Oxygen signal;
Comparison module is commented for generation target after carrying out data processing to the target organism information of the acquisition mould group acquisition
Estimate parameter, and the goal-based assessment parameter is compared with criterion evaluation parameter, obtains comparison result;
Assessment result generates mould group, compares the obtained comparison result and the target for passing through according to the comparison module
Physiologic information generates assessment report.
7. novel risk of stroke assessment system according to claim 6, which is characterized in that the comparison module, packet
It includes:
Wavelet transformation analysis unit, for the target brain blood in the target organism information to the acquisition mould group acquisition
Oxygen signal carries out wavelet transformation analysis, obtains frequency-domain result, also obtains the average phase information in default wave band;
Goal-based assessment parameter acquiring unit, described in being obtained to the wavelet transformation analysis unit by wavelet transformation analysis
Average phase information in frequency-domain result and the predetermined period carries out data processing, obtains the goal-based assessment parameter.
8. novel risk of stroke assessment system according to claim 7, which is characterized in that the goal-based assessment parameter
Acquiring unit, comprising:
Wavelet amplitude analysis module, the frequency domain for being obtained to the wavelet transformation analysis unit by wavelet transformation analysis
As a result wavelet amplitude analysis is carried out, wavelet amplitude mean value is obtained;
Effective connectivity analysis module, it is described default for being obtained to the wavelet transformation analysis unit by wavelet transformation analysis
Average phase information in wave band carries out effective connectivity analysis, obtains stiffness of coupling and the coupling direction of the default wave band.
9. any novel risk of stroke assessment system according to claim 8, which is characterized in that the comparison mould
Group, further includes:
Comparator, the wavelet amplitude mean value and the criterion evaluation parameter for being obtained to the wavelet amplitude analysis module
In Standard wavelet amplitude mean value be compared, if the wavelet amplitude mean value be greater than the criterion evaluation parameter in standard it is small
Wave amplitude mean value is then exception;
Stiffness of coupling and the criterion evaluation for the default wave band analyzed the effective connectivity analysis module
Standard stiffness of coupling in parameter is compared, if the stiffness of coupling is strong beyond the standard coupling in the criterion evaluation parameter
Degree, then be exception;
The coupling direction of the default wave band for analyzing the effective connectivity analysis module and the criterion evaluation
Standard coupling direction in parameter is compared, if the coupling direction couples direction with the standard in the criterion evaluation parameter
Difference is then exception.
10. novel risk of stroke assessment system according to claim 9, which is characterized in that the assessment result is raw
At mould group, comprising:
Model foundation unit, the comparison generated for the desired physiological information and comparator according to acquisition mould group acquisition
As a result Bayesian network model is established;
Assessment report generation unit, the Bayesian network for being established by maximum likelihood algorithm to the model foundation unit
Parameter in network model is learnt, and the node probability of each node in the network model is generated, raw according to the node probability
At the assessment report.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910016245.6A CN109864745A (en) | 2019-01-08 | 2019-01-08 | A kind of novel risk of stroke appraisal procedure and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910016245.6A CN109864745A (en) | 2019-01-08 | 2019-01-08 | A kind of novel risk of stroke appraisal procedure and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109864745A true CN109864745A (en) | 2019-06-11 |
Family
ID=66917532
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910016245.6A Pending CN109864745A (en) | 2019-01-08 | 2019-01-08 | A kind of novel risk of stroke appraisal procedure and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109864745A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110974172A (en) * | 2019-12-13 | 2020-04-10 | 北京理工大学 | Real-time physiological parameter measuring system |
CN111481193A (en) * | 2020-04-17 | 2020-08-04 | 国家康复辅具研究中心 | Fall risk assessment and early warning method and system |
CN111631731A (en) * | 2020-05-09 | 2020-09-08 | 国家康复辅具研究中心 | Near-infrared brain function and touch force/motion information fusion assessment method and system |
CN111968744A (en) * | 2020-10-22 | 2020-11-20 | 深圳大学 | Bayesian optimization-based parameter optimization method for stroke and chronic disease model |
CN114999658A (en) * | 2022-08-04 | 2022-09-02 | 四川大学华西医院 | System and method for digital intervention on ICU (posterior ICU) syndrome |
CN116230212A (en) * | 2023-04-04 | 2023-06-06 | 曜立科技(北京)有限公司 | Diagnosis decision system for postoperative cerebral apoplexy review based on data processing |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110257545A1 (en) * | 2010-04-20 | 2011-10-20 | Suri Jasjit S | Imaging based symptomatic classification and cardiovascular stroke risk score estimation |
CN104510451A (en) * | 2013-09-29 | 2015-04-15 | 海思康利(北京)新技术有限公司 | Non-load monitoring system |
CN105246399A (en) * | 2013-06-26 | 2016-01-13 | 英特尔公司 | Detection of a leading stroke risk indicator |
US20170140115A1 (en) * | 2017-01-24 | 2017-05-18 | Liwei Ma | Intelligent stroke risk prediction and monitoring system |
CN106901749A (en) * | 2017-01-06 | 2017-06-30 | 苏州大学 | A kind of extracting method of cortex hemoglobin information representation locomitivity parameter |
CN107273652A (en) * | 2017-03-10 | 2017-10-20 | 马立伟 | Intelligent risk of stroke monitoring system |
CN108685577A (en) * | 2018-06-12 | 2018-10-23 | 国家康复辅具研究中心 | A kind of brain function rehabilitation efficacy apparatus for evaluating and method |
-
2019
- 2019-01-08 CN CN201910016245.6A patent/CN109864745A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110257545A1 (en) * | 2010-04-20 | 2011-10-20 | Suri Jasjit S | Imaging based symptomatic classification and cardiovascular stroke risk score estimation |
CN105246399A (en) * | 2013-06-26 | 2016-01-13 | 英特尔公司 | Detection of a leading stroke risk indicator |
CN104510451A (en) * | 2013-09-29 | 2015-04-15 | 海思康利(北京)新技术有限公司 | Non-load monitoring system |
CN106901749A (en) * | 2017-01-06 | 2017-06-30 | 苏州大学 | A kind of extracting method of cortex hemoglobin information representation locomitivity parameter |
US20170140115A1 (en) * | 2017-01-24 | 2017-05-18 | Liwei Ma | Intelligent stroke risk prediction and monitoring system |
CN107273652A (en) * | 2017-03-10 | 2017-10-20 | 马立伟 | Intelligent risk of stroke monitoring system |
CN108685577A (en) * | 2018-06-12 | 2018-10-23 | 国家康复辅具研究中心 | A kind of brain function rehabilitation efficacy apparatus for evaluating and method |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110974172A (en) * | 2019-12-13 | 2020-04-10 | 北京理工大学 | Real-time physiological parameter measuring system |
CN111481193A (en) * | 2020-04-17 | 2020-08-04 | 国家康复辅具研究中心 | Fall risk assessment and early warning method and system |
CN111631731A (en) * | 2020-05-09 | 2020-09-08 | 国家康复辅具研究中心 | Near-infrared brain function and touch force/motion information fusion assessment method and system |
CN111631731B (en) * | 2020-05-09 | 2022-05-13 | 国家康复辅具研究中心 | Near-infrared brain function and touch force/motion information fusion assessment method and system |
CN111968744A (en) * | 2020-10-22 | 2020-11-20 | 深圳大学 | Bayesian optimization-based parameter optimization method for stroke and chronic disease model |
CN111968744B (en) * | 2020-10-22 | 2021-02-19 | 深圳大学 | Bayesian optimization-based parameter optimization method for stroke and chronic disease model |
CN114999658A (en) * | 2022-08-04 | 2022-09-02 | 四川大学华西医院 | System and method for digital intervention on ICU (posterior ICU) syndrome |
CN114999658B (en) * | 2022-08-04 | 2022-11-11 | 四川大学华西医院 | System and method for digital intervention on ICU (posterior ICU) syndrome |
CN116230212A (en) * | 2023-04-04 | 2023-06-06 | 曜立科技(北京)有限公司 | Diagnosis decision system for postoperative cerebral apoplexy review based on data processing |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109864745A (en) | A kind of novel risk of stroke appraisal procedure and system | |
Chowdhury et al. | Estimating blood pressure from the photoplethysmogram signal and demographic features using machine learning techniques | |
KR102220229B1 (en) | Improving biometric performance in the presence of noise | |
US11517212B2 (en) | Electrocardiogram information dynamic monitoring method and dynamic monitoring system | |
Gow et al. | Multiscale entropy analysis of center-of-pressure dynamics in human postural control: methodological considerations | |
Bhagyalakshmi et al. | GB-SVNN: Genetic BAT assisted support vector neural network for arrhythmia classification using ECG signals | |
Burgos et al. | Real-time detection of apneas on a PDA | |
CN114376564B (en) | Sleep staging method, system, device and medium based on ballistocardiogram signals | |
CN110623678A (en) | Blood glucose measuring device, data processing method thereof and storage medium | |
Roy et al. | On-device reliability assessment and prediction of missing photoplethysmographic data using deep neural networks | |
Yang et al. | Blood pressure estimation with complexity features from electrocardiogram and photoplethysmogram signals | |
Al-Saud et al. | Wireless body area sensor networks signal processing and communication framework: survey on sensing, communication technologies, delivery and feedback | |
Xing et al. | An artifact-resistant feature SKNAER for quantifying the burst of skin sympathetic nerve activity signal | |
Laurino et al. | Moving auto-correlation window approach for heart rate estimation in ballistocardiography extracted by mattress-integrated accelerometers | |
Ali et al. | A novel technique to diagnose sleep apnea in suspected patients using their ECG data | |
EP3419502B1 (en) | Stress detection based on sympathovagal balance | |
Sunkaria et al. | Wavelet based R-peak detection for heart rate variability studies | |
Zhang et al. | Non-invasive blood glucose detection using NIR based on GA and SVR | |
Obi | An Overview of Wearable Photoplethysmographic Sensors and Various Algorithms for Tracking of Heart Rates | |
JP2024500593A (en) | Secure identification methods and systems | |
Yang et al. | Interval-based LDA algorithm for electrocardiograms for individual verification | |
Li et al. | Adaptive detection of Ahead-sEMG based on short-time energy of local-detail difference and recognition in advance of upper-limb movements | |
Shuzan et al. | A novel non-invasive estimation of respiration rate from photoplethysmograph signal using machine learning model | |
Mouney et al. | A new blood pressure estimation approach using ppg sensors: Subject specific evaluation over a long-term period | |
Rawat et al. | Study of Phylogenetic for Computational Analysis of Sleep Apnea Syndrome for Patient (Healthcare & Treatment) Using Machine Learning (Robot Vision) |
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: 20190611 |
|
WD01 | Invention patent application deemed withdrawn after publication |