CN108245155A - A kind of stroke types identification system of Multi-channel microwave aerial array - Google Patents
A kind of stroke types identification system of Multi-channel microwave aerial array Download PDFInfo
- Publication number
- CN108245155A CN108245155A CN201711100184.9A CN201711100184A CN108245155A CN 108245155 A CN108245155 A CN 108245155A CN 201711100184 A CN201711100184 A CN 201711100184A CN 108245155 A CN108245155 A CN 108245155A
- Authority
- CN
- China
- Prior art keywords
- microwave
- channel
- multiplex switch
- signal
- stroke types
- 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
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
- A61B5/0507—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves using microwaves or terahertz waves
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
- A61B5/0015—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by features of the telemetry system
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4058—Detecting, measuring or recording for evaluating the nervous system for evaluating the central nervous system
- A61B5/4064—Evaluating the brain
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
- A61B5/6803—Head-worn items, e.g. helmets, masks, headphones or goggles
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6813—Specially adapted to be attached to a specific body part
- A61B5/6814—Head
Abstract
The invention discloses a kind of stroke types identification systems of Multi-channel microwave aerial array.Identification system mainly includes perimeter antenna array, signal occurs and processing module, microwave multi-path multiplex switch and PXI hosts.It is as follows using the key step of identification system：1) identification system is connected.2) signal is occurred to measure correction with processing module.3) multiplexer switch handover measurement channel is controlled, and exports trigger signal.4) 16 groups of all Measurement channels are traversed, occur to measure with processing module using signal and record the microwave transmission parameter magnitudes and phase of each channel.5) dimension-reduction treatment is carried out to initial data using principal component analysis.6) after dimensionality reduction, classifier training is carried out to the data after dimensionality reduction using support vector machines, obtains trained stroke types grader.8) trained stroke types grader is utilized, stroke types discriminating is carried out to object to be measured.
Description
Technical field
The present invention relates to biomedical technical field of medical equipment, the brain soldier of specifically a kind of Multi-channel microwave aerial array
Middle type identification system.
Background technology
Cerebral apoplexy is divided into hemorrhagic apoplexy and Ischemic Stroke, but the clinical treatment means mutual exclusion of two parapoplexies, therefore controls
Palsy type diagnostic before treatment is essential.According to European clinical instructional criterion, 4.5 hours after Ischemic Stroke are molten to receive
The gold period of thrombus treatment is solved, and the sooner the better.But since communications and transportation has been missed when a large amount of patients arrive at hospital due to
Thrombolysis gold period so that only less than 10% ischemic patients with stroke receive thromboembolism treatment.Therefore, development is a kind of just
Take, be quick, the instrument easy to operate that can realize palsy type diagnostic before institute it is significant.
The mobile CT equipment developed at present is used for into the cerebral apoplexy diagnosis before houses of prostitutes or actresses in the Jin and Yuan Dynasties, but contains the doctor of mobile CT equipment
It treats vehicle and has used advanced equipment, cost is higher, but also requirement has well-trained medical staff to accompany.Meanwhile it contains
The medical vehicle of mobile CT equipment is very high to traffic and communications facility requirement, and the application range of the medical vehicle may be limited only to relatively send out
The area and densely populated areas reached.Therefore, a miniature portable, price it is relatively low and be easier to operate to stroke types diagnosis
Equipment still has great importance for some sparsely populated or more poor areas.
Human body different tissues have different dielectric characteristics, therefore can will be applied to palsy based on Microwave Detection Principle
Before patient institute on type diagnostic, and can realize it is quick, non-contact, portable and easy to operate, so as to which traditional Image-aided be replaced to examine
The shortcomings that breaking and overcoming it time-consuming, a wide range of mobile patient strives for of short duration and valuable treatment time after morbidity for patient.But
It is to be currently based on the stroke types discrimination method of microwave to be mainly based upon the detection method of single pair antenna, the palsy disease of extraction
People's encephalic physiological pathology change information is limited, easily causes diagnostic error, so as to influence the treatment of patients with stroke.
Invention content
Present invention aim to address problems of the prior art.
To realize the present invention purpose and the technical solution adopted is that such, a kind of brain soldier of Multi-channel microwave aerial array
Middle type identification system mainly occurs and processing module, microwave multi-path multiplex switch and PXI including perimeter antenna array, signal
Host.
Further, perimeter antenna array, signal occur to pass through coaxial cable with processing module and microwave multi-path multiplex switch
Connection.
The perimeter antenna array mainly includes 8 microwave paster antennas, acrylic annulus and 16 groups of Measurement channels.
8 microwave paster antennas are uniformly fixed on the acrylic annulus.
8 microwave paster antennas microwave for transmitting and receiving.
The signal occurs mainly to include sender unit and vector network analyzer with processing module.
The sender unit is used for transmiting signal source.
The sender unit has external trigger input port.
The vector network analyzer receives the external trigger signal that the microwave multi-path multiplex switch is sent.The vector net
Network analyzer measures 16 groups of Measurement channels.
The microwave multi-path multiplex switch has trigger output end mouth.
The trigger output end mouth transmits what the sender unit generated by connecting the external trigger input port
Signal source.
The microwave multi-path multiplex switch is used to switch 16 groups of Measurement channels.
The microwave multi-path multiplex switch often switches after one group of Measurement channel group outside vector network analyzer transmission
Trigger signal.
The PXI hosts mainly include local disk.
The on off operating mode of microwave multi-path multiplex switch described in the PXI host computer controls.
The local disk preserve the microwave transmission parameter that the vector network analyzer measurement obtains amplitude data and
Phase data.
A kind of stroke types identification system using the Multi-channel microwave aerial array is come the stroke types that differentiate
Experiment, mainly include the following steps that：
1) the stroke types identification system of connecting multi-channel microwave antenna array, and system and device is initialized.
2) detected object is placed in the center of the acrylic annulus.
3) signal is occurred to measure and correct with processing module.It is utilized on the vector network analyzer
VBA writes the program for measuring and preserving microwave scattering parameters.Described program is started.It is external tactile that described program enters waiting
Signaling states.
4) the microwave multi-path multiplex switch is inserted into PXI hosts.
5) start identification system, 8 microwave paster antennas transmitting microwave is allowed to pass through detected object.
6) it is connected using the PXI host computer controls two microwave multi-path multiplex switch.The microwave multi-path connected
Multiplex switch traverses multiple Measurement channels of timesharing.The microwave multi-path multiplex switch connected is to the vector network analyzer
Send external trigger signal.
7) signal is generated receives the trigger signal with processing module.The vector network analyzer is to current channel
Microwave transmission parameter amplitude and phase measure.The data that measurement obtains are preserved to the local disk.
8) step 6 and step 7 are repeated, controls the break-make of the microwave multi-path multiplex switch, so as to 16 groups of Measurement channels
It switches over.16 groups of Measurement channels are traversed, the microwave transmission for being measured and being recorded each channel using the vector network analyzer is joined
Number amplitude and phase.
9) using Principal Component Analysis to the microwave transmission parameter magnitudes of 16 groups of Measurement channels and the initial data of phase into
Row dimension-reduction treatment.
10) after dimensionality reduction, classifier training is carried out to the data after dimensionality reduction by training sample using support vector machines, so as to
Obtain trained stroke types grader.
11) trained stroke types grader is utilized, to being located at the detected object of the acrylic circle ring center
Carry out stroke types discriminating.
The solution have the advantages that unquestionable.The present invention builds multichannel brain palsy by establishing aerial array
Type identification system.
When differentiating using the present invention to stroke types, non-contact detection can be realized, so as to avoid to tested
The patient of survey causes wound.Meanwhile hardware system of the invention mainly generates processing module by hood type antenna and signal and forms,
It is at low cost, it is small, it is easy to carry, it can be used in patients with stroke onset site or ambulance, so as to shorten morbidity to disease
Time loss during people's emergency treatment.
When the present invention is detected stroke types using ring shape microwave aerial array and multi-channel wide frequency band, antenna array
Row measuring system can obtain more patients with stroke encephalic pathological informations, and the present invention is smaller, sensitive by the interference of external environment
Degree is high.It is excessive that the present invention for multi-channel system leads to the problem of data volume, and dimensionality reduction has been carried out to original multi-channel measurement data
Processing reduces redundancy, improves detection performance, achievees the purpose that accurately to differentiate patient's stroke types.
Description of the drawings
Fig. 1 is the block diagram of system；
Fig. 2 is system module figure；
Fig. 3 is Cerebral Hemorrhage of Rabbit datagram；
Fig. 4 is rabbit cerebral ischemia datagram.
In figure：Perimeter antenna array, signal occur and processing module, microwave multi-path multiplex switch, PXI hosts, 8 microwaves
Paster antenna, acrylic annulus, sender unit, vector network analyzer and local disk.
Specific embodiment
With reference to embodiment, the invention will be further described, but should not be construed the above-mentioned subject area of the present invention only
It is limited to following embodiments.Without departing from the idea case in the present invention described above, according to ordinary skill knowledge and used
With means, various replacements and change are made, should all be included within the scope of the present invention.
Embodiment 1：
A kind of stroke types identification system of Multi-channel microwave aerial array mainly includes perimeter antenna array 1, signal
Occur and processing module 2, microwave multi-path multiplex switch 3 and PXI hosts 4.
Further, perimeter antenna array 1, signal occur to pass through coaxial electrical with processing module 2 and microwave multi-path multiplex switch 3
Cable connects.
The perimeter antenna array 1 is mainly led to including 8 microwave paster antennas 101, acrylic annulus 102 and 16 groups of measurements
Road.
Further, single microwave paster antenna 101 is that a plain conductor is coiled into definite shape, such as it is round, rectangular, three
It is angular etc., using conductor both ends as the structure of output terminal.
8 microwave paster antennas 101 are uniformly fixed on the acrylic annulus 102.
8 microwave paster antennas 101 microwave for transmitting and receiving.
The signal occurs mainly to include sender unit 201 and vector network analyzer 202 with processing module 2.
The sender unit 201 is used for transmiting signal source.
The sender unit 201 has external trigger input port.
Further, the sender unit 201 is in amplitude response, the frequency for measuring various telecommunication systems or telecommunication apparatus
When characteristic, transmission characteristic and other electrical parameters, for emitting the signal source of test or driving source.
The vector network analyzer 202 receives the external trigger signal that the microwave multi-path multiplex switch 3 is sent.The arrow
Amount Network Analyzer 202 measures 16 groups of Measurement channels.
Further, the vector network analyzer 202 can measure the various parameters of one port network or two-port network
Amplitude, and phase data can be measured.The test of phase fluctuation parameter is the electronic delay using the vector network analyzer 202
(ElectricalDelay) function is realized.
The vector network analyzer 202 can show test data with Smith chart.
The microwave multi-path multiplex switch 3 has trigger output end mouth.
The trigger output end mouth is generated by connecting the external trigger input port transmission sender unit 201
Signal source.
The microwave multi-path multiplex switch 3 is used to switch 16 groups of Measurement channels.
The microwave multi-path multiplex switch 3 often switches one group of Measurement channel group, and the vector network analyzer 202 is sent out backward
Send external trigger signal.
The PXI hosts 4 mainly include local disk 401.
Further, (the PCIextensionsforInstrumentation, towards the PCI of instrument system of PXI hosts 4
Extension) it is a kind of measurement based on PC machine issued by NI companies and automation platform.The PXI hosts 4 combine PCI
The electric bus characteristic and CompactPCI of (PeripheralComponentInterconnection- peripheral component interconnections)
The characteristic of the robustness of (compact PCI), modularization and Eurocard mechanical encapsulations, so as to develop into be suitable for testing, measure with
Mechanical, the electrical and software specifications of data acquisition applications.
The purpose for working out PXI specifications be in order to by the ratio of performance to price advantage of Desktop PC and pci bus towards instrument field
Necessary extension ideally combine, form a kind of virtual instrument test platform of mainstream.This becomes the PXI hosts 4
It measures and the high-performance of automated system, inexpensive carrying platform.
The PXI hosts 4 control the on off operating mode of the microwave multi-path multiplex switch 3.
The local disk 401 preserves the vector network analyzer 202 and measures the obtained amplitude of microwave transmission parameter
Data and phase data.
Embodiment 2：
As depicted in figs. 1 and 2, a kind of stroke types identification system using the Multi-channel microwave aerial array reflects
The experiment of other stroke types, mainly includes the following steps that：
1) the stroke types identification system of connecting multi-channel microwave antenna array, and system and device is initialized.
2) detected object is placed in the center of the acrylic annulus 102.
3) signal is occurred to measure and correct with processing module 2.It is sharp on the vector network analyzer 202
The program for measuring and preserving microwave scattering parameters is write with VBA.Described program is started.It is external that described program enters waiting
Trigger signal state.
4) the microwave multi-path multiplex switch 3 is inserted into PXI hosts.
5) start identification system, 8 microwave paster antennas, 101 antenna transmitting microwave is allowed to pass through detected object.
6) two microwave multi-path multiplex switch 3 is controlled to connect using the PXI hosts 4.The microwave connected is more
Road multiplex switch 3 traverses multiple Measurement channels of timesharing.
Further, palsy, which differentiates, needs quick detection, is not required to detect in real time, and is to avoid different sense channels mutual
Interference, therefore multi-channel detection is realized in a manner of timesharing.
The microwave multi-path multiplex switch 3 connected sends external trigger signal to the vector network analyzer.
Further, the regular hour that is delayed is needed during the 3 handover measurement channel of microwave multi-path multiplex switch, waits for vector
Network Analyzer completes the measurement of the channel and data preserve.
7) signal is generated receives the trigger signal with processing module 2.The vector network analyzer 202 is to current
The amplitude and phase of the microwave transmission parameter of channel measure.The data that measurement obtains are preserved to the local disk 401.
8) step 6 and step 7 are repeated, controls the break-make of the microwave multi-path multiplex switch 3, so as to 16 groups of Measurement channels
It switches over.16 groups of Measurement channels are traversed, measure and record the microwave transmission of each channel using the vector network analyzer 202
Parameter magnitudes and phase.
9) using Principal Component Analysis to the microwave transmission parameter magnitudes of 16 groups of Measurement channels and the initial data of phase into
Row dimension-reduction treatment.
The key step for analyzing the principal component of data is as follows：
9.1) initial data of microwave transmission parameter magnitudes and phase is acquired, determines situational variables.
R sample is acquired out, each sample has p variable.The r sample and the p variable form a r × p
Type matrix.
9.2) initial data is standardized, to eliminate influence of the dimension to data processing.
Standardization formula：yuv=xuv-xu/su (1)
In formula, yuvFor the variate-value after standardization.xuvFor real variable value.xuFor arithmetic mean of instantaneous value.suFor standard deviation.u
Row subscript for matrix-vector.V is the row subscript of matrix-vector.
Treated, and data matrix is：
9.3) the normal orthogonal feature vector of characteristic root and response is calculated.
Correlation matrix Z is：
In formula, zuvFor former variable zuWith zvRelated coefficient.U is the row subscript of matrix-vector.V is under the row of matrix-vector
Mark.
zuv=zvu (4)
In formula, zuvFor former variable zuWith zvRelated coefficient.U is the row subscript of matrix-vector.V is under the row of matrix-vector
Mark.yduAnd ydvFor the variate-value after standardization.For yduArithmetic mean of instantaneous value.For ydvArithmetic mean of instantaneous value.R is sample
Number.Integer d be arbitrary sample, 1≤d≤r.
| λ e-Z |=0 (6)
In formula, λ is the characteristic value of matrix Z.E is unit matrix.Z is the matrix that formula (3) represents.
According to formula (6), the p characteristic value of matrix Z can be obtained, the variance contribution size of each principal component is suitable by characteristic root
Sequence is successively decreased arrangement.
Utilize each eigenvalue λjSolve equation group Zb=λjB corresponds to the characteristic variable b of each characteristic valuej。
Wherein, Z is the matrix that formula (3) represents.B is characterized variable.λjFor each characteristic value.
9.4) principal component contributor rate and contribution rate of accumulative total are calculated.
P original variable y1,y2,...,ypPartitioning of total variance variable gs independent for p1,g2,...,gpVariance
The sum of.
J-th principal component yJVariance contribution ratio be：
In formula, λJEach characteristic value for matrix Z.J be principal component number, J=1,2 ..., p.P is independent variable number.
First principal component contribution rate is maximum, i.e. g1The ability of comprehensive original variable is most strong.g2,g3,...,gpIntegration capability
Successively decrease successively.
If only take m principal component therein, then the contribution rate of accumulative total of this m principal component is：
In formula, λJIt is each characteristic value of matrix Z.J be principal component number, J=1,2 ..., p.P is original variable number.
λrIt is all characteristic values of matrix Z.Integer r be arbitrary sample, 1≤r≤m.M is principal component number.
9.5) new generalized variable is obtained.
That is,
In formula, p represents Independent Vector number.y1、y2、y3......ypFor the variate-value after standardization.M is principal component
Number.L11、L21、L31......LmpFor principal component scores coefficient.
10) after dimensionality reduction, classifier training is carried out to the data after dimensionality reduction by training sample using support vector machines, so as to
Obtain trained stroke types grader.
Training step is as follows：
Choose training sample set (xi,yi) and vector xa∈Rn, wherein i=1,2...l.N is sample dimension.RnFor data sky
Between.Y ∈ { -1 ,+1 } are category labels.Y is by the training sample set (xi,yi) it is divided into two classes.
Hyperplane equation is：
ω x+g=0 (10)
In formula, ω is weight vector.G is the biasing of formula (10).
The constraints of formula (10) is as follows：
ζi≥0
In formula, ζiFor arbitrary slack variable.
At this point, maximum spacing hyperplane is Generalized optimal Optimal Separating Hyperplane.
The constraints of formula (10) becomes：
yi[(ω·xi)+g]≥1-ζi (11)
It is solved using support vector machines, the initial shape of corresponding optimization problem is：
In formula, w is supporting vector.wTTransposition for w.G is biasing.ρ is the threshold value of supporting vector.ζ is slack variable.v
For the limits of error.ζiFor arbitrary slack variable.L is the sum of slack variable.Serial numbers of the i for slack variable, 1≤i≤l.
The constraints of formula (12) is：
In formula, w is supporting vector.wTTransposition for w.ζiFor arbitrary slack variable.Serial numbers of the i for slack variable, 1≤i≤
l.ρ is the threshold value of supporting vector.φ(xi) for the vector after Nonlinear Mapping.
Its dual form is
In formula, w is supporting vector.α is the dual form of w.αTTransposition for α.Q is l × l rank positive semidefinite matrixs.
The constraints of formula (14) is：
In formula, eTTransposition for unit matrix.αiFor the weighted value corresponding to each sample.V is the limits of error.L becomes for relaxation
The sum of amount.α is the dual form of w.W is supporting vector.
Qij≡yiyjK(xi,xj)≡φ(xi)Tφ(xi) (16)
In formula, K (xi,xj) it is kernel function.xiAnd xjIt is two vectors after Nonlinear Mapping in feature space.1≤i≤
l。φ(xi) for the vector after Nonlinear Mapping.φ(xi)TFor φ (xi) transposition.
Discriminant function is：
In formula, K (xi, x) and it is kernel function.1≤i≤l.ρ is the parameter vector of Optimal Separating Hyperplane.xiAnd xjIt is non-linear reflect
Penetrate two vectors in rear feature space.1≤i≤l.αiFor the weighted value corresponding to each sample.
11) trained stroke types grader is utilized, to being located at detected pair of 102 center of acrylic annulus
As carrying out stroke types discriminating.
Embodiment 3：
Using a kind of stroke types identification system of Multi-channel microwave aerial array to Rabbits with Cerebral Ischemia and cerebral hemorrhage man
Rabbit is differentiated.Specific implementation step is as follows：
1) DATA REASONING
1.1) rabbits with cerebral hemorrhage detects.
1.1.1) prepare 5 rabbit, and processing early period is carried out to 5 rabbit, it is main to include anesthesia, establish acute brain
Bleeding Rabbit Model etc..Handle well 5 rabbit are placed on measuring table, and a data are measured to 5 rabbit, will be surveyed
The data measured are calculated as experimental data during 0ML.
1.1.2 after) setting syringe pump parameter, start syringe pump, set interval as 1min, shot 1ML (streams
Speed).
1.1.3 the start button of syringe pump) is pressed, the data of record at this time after the completion of injection.The microwave multi-path is opened
It closes 3 and shares 16 groups of Measurement channels, control the microwave multi-path switch 3 using program, open one group of Measurement channel every time, be calculated as
Experimental data during 1ML
1.1.4 step 1.1.3) is repeated, completes data acquisition when 2ML, 3ML and 4ML.
1.2) Rabbits with Cerebral Ischemia detects.
1.2.1 after) rabbit ischemia model is prepared, measuring table is positioned over, does not fasten tight knot binding at this time.Automatic prison
Ranging sequence triggering variable connector 16 groups of channels are opened successively, are opened one group every time, are recorded data at this time and preservation, will at this time
Data are considered as experimental data during ischemic 0min.Automatic monitoring program enters standby mode, fastens ligature at once, is considered as ischemic
Start.
1.2.2 automatic monitoring program) is set as every 6min automatically to start once, records data at this time.
1.2.3 1.2.2) is repeated, record total time is 1.5 hours to 2 hours.
2) data processing
To each of bleeding and ischemic state, in 16 channels, each Air conduct measurement transmission and reflection parameters (s11,
S12, s21, s22) amplitude and phase, in the range of 300KHz~3GHz sample 1001 frequency point datas, as differentiate classify
Raw measurement data.Then to raw measurement data, dimension-reduction treatment is carried out, the data after dimensionality reduction are obtained, as grader
Input data.
3) experimental result
Cerebral Hemorrhage of Rabbit measurement data is as shown in Figure 3.Rabbit cerebral ischemia measurement data is as shown in Figure 4.Two kinds of comparison is waited to reflect
Other data can be seen that as palsy degree increases, and the trend of rising is presented in the two curve, and the two difference is mainly reflected in
Waveform.Bleeding changes greatly, and peak value is apparent, and ischemic is relatively steady, the peak value not protruded.Identification result is as shown in table 1.
1 Cerebral Hemorrhage of Rabbit of table and cerebral ischemia identification experiment result
More than four groups the results show that either maximum residual quantity or minimum residual quantity, the identification result obtained after dimensionality reduction is accurate
Exactness is very high, has two groups of situations to reach 100%, and the wherein best situation of effect is bleeding 1ml and ischemic 42min, bleeding 1ml and
Ischemic 6min effects are taken second place.
The result of initial data before comparison dimensionality reduction can be seen that identification result and one kept between dimensionality reduction and initial data
Cause property, i.e., the identification result after dimensionality reduction is accurate.Dimensionality reduction not only realizes reduction data volume, also achieves filtering.Therefore, after dimensionality reduction
Obtained data can accurately identify cerebral ischemia palsy and cerebral hemorrhage palsy.
Claims (3)
1. a kind of stroke types identification system of Multi-channel microwave aerial array, it is characterised in that：Mainly include loop aerial
Array (1), signal occur and processing module (2), microwave multi-path multiplex switch (3) and the PXI hosts (4)；
The perimeter antenna array (1) mainly includes 8 microwave paster antennas (101), acrylic annulus (102) and 16 groups of measurements
Channel；
8 microwave paster antennas (101) are uniformly fixed on the acrylic annulus (102)；
8 microwave paster antennas (101) microwave for transmitting and receiving；
The signal occurs mainly to include sender unit (201) and vector network analyzer (202) with processing module (2).
The sender unit (201) is for transmiting signal source；
The sender unit (201) has external trigger input port；
The vector network analyzer (202) receives the external trigger signal that the microwave multi-path multiplex switch (3) sends；The arrow
Amount Network Analyzer (202) measures 16 groups of Measurement channels；
The microwave multi-path multiplex switch (3) has trigger output end mouth；
The trigger output end mouth transmits what the sender unit (201) generated by connecting the external trigger input port
Signal source；
The microwave multi-path multiplex switch (3) is for switching 16 groups of Measurement channels；
The microwave multi-path multiplex switch (3) often switches one group of Measurement channel group vector network analyzer (202) hair backward
Send external trigger signal；
The PXI hosts (4) mainly include local disk (401)；
The PXI hosts (4) control the on off operating mode of the microwave multi-path multiplex switch (3)；
The local disk (401) preserves the vector network analyzer (202) and measures the obtained amplitude of microwave transmission parameter
Data and phase data.
2. a kind of stroke types identification system of Multi-channel microwave aerial array according to claim 1, feature exist
In：Perimeter antenna array (1), signal generation are connected with processing module (2) and microwave multi-path multiplex switch (3) by coaxial cable
It connects.
3. a kind of stroke types using any one of 1 to 2 claim Multi-channel microwave aerial array differentiate system
The experiment of stroke types for uniting to differentiate, which is characterized in that mainly include the following steps that：
1) the stroke types identification system of connecting multi-channel microwave antenna array, and system and device is initialized；
2) detected object is placed in the center of the acrylic annulus (102)；
3) signal is occurred to measure and correct with processing module (2)；It is sharp on the vector network analyzer (202)
The program for measuring and preserving microwave scattering parameters is write with VBA；Described program is started；It is external that described program enters waiting
Trigger signal state；
4) the microwave multi-path multiplex switch (3) is inserted into PXI hosts；
5) start identification system, 8 microwave paster antennas (101) antenna transmitting microwave is allowed to pass through detected object；
6) two microwave multi-path multiplex switch (3) is controlled to connect using the PXI hosts (4)；The microwave connected is more
Road multiplex switch (3) traverses multiple Measurement channels of timesharing；The microwave multi-path multiplex switch (3) connected is to the vector net
Network analyzer sends external trigger signal；
7) signal is generated receives the trigger signal with processing module (2)；The vector network analyzer (202) is to current
The amplitude and phase of the microwave transmission parameter of channel measure；The data that measurement obtains are preserved to the local disk
(401)；
8) step 6 and step 7 are repeated, controls the break-make of the microwave multi-path multiplex switch (3), so as to 16 groups of Measurement channels into
Row switching；16 groups of Measurement channels are traversed, measure and record the microwave transmission of each channel using the vector network analyzer (202)
Parameter magnitudes and phase；
9) the microwave transmission parameter magnitudes of 16 groups of Measurement channels and the initial data of phase are dropped using Principal Component Analysis
Dimension processing；
10) after dimensionality reduction, classifier training is carried out to the data after dimensionality reduction by training sample using support vector machines, so as to obtain
Trained stroke types grader；
11) trained stroke types grader is utilized, to being located at the detected object at acrylic annulus (102) center
Carry out stroke types discriminating.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711100184.9A CN108245155A (en) | 2017-11-09 | 2017-11-09 | A kind of stroke types identification system of Multi-channel microwave aerial array |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711100184.9A CN108245155A (en) | 2017-11-09 | 2017-11-09 | A kind of stroke types identification system of Multi-channel microwave aerial array |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108245155A true CN108245155A (en) | 2018-07-06 |
Family
ID=62721001
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711100184.9A Pending CN108245155A (en) | 2017-11-09 | 2017-11-09 | A kind of stroke types identification system of Multi-channel microwave aerial array |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108245155A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109498010A (en) * | 2018-11-30 | 2019-03-22 | 中国人民解放军陆军军医大学 | Data processing system applied to the diagnosis closed injury of brain of continuous wave spectrum scan |
| CN110393526A (en) * | 2019-08-16 | 2019-11-01 | 北京师范大学 | A kind of high frequency feeble computer signals amplification acquisition system |
| CN111317470A (en) * | 2019-12-27 | 2020-06-23 | 中国人民解放军陆军军医大学 | Portable microwave detection system for clinical pre-hospital diagnosis of cerebral apoplexy type |
| CN112487713A (en) * | 2020-11-26 | 2021-03-12 | 江苏科技大学 | Method for extracting physical size of microstrip antenna based on multilayer PCA Gaussian process |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6454711B1 (en) * | 1999-04-23 | 2002-09-24 | The Regents Of The University Of California | Microwave hemorrhagic stroke detector |
| CN103018738A (en) * | 2011-09-20 | 2013-04-03 | 中国科学院电子学研究所 | Microwave three-dimensional imaging method based on rotary antenna array |
| CN105769190A (en) * | 2016-03-29 | 2016-07-20 | 重庆大学 | Device for detecting cephalophyma and encephaledema |
| CN105832331A (en) * | 2016-03-18 | 2016-08-10 | 中国人民解放军第三军医大学 | Non-contact cerebral hemorrhage detection device based on broadband antenna technology and detection method of detection device |
| CN105866772A (en) * | 2016-04-18 | 2016-08-17 | 浙江大学 | Novel method for positioning metal object in human body based on microwave coherence imaging |
-
2017
- 2017-11-09 CN CN201711100184.9A patent/CN108245155A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6454711B1 (en) * | 1999-04-23 | 2002-09-24 | The Regents Of The University Of California | Microwave hemorrhagic stroke detector |
| CN103018738A (en) * | 2011-09-20 | 2013-04-03 | 中国科学院电子学研究所 | Microwave three-dimensional imaging method based on rotary antenna array |
| CN105832331A (en) * | 2016-03-18 | 2016-08-10 | 中国人民解放军第三军医大学 | Non-contact cerebral hemorrhage detection device based on broadband antenna technology and detection method of detection device |
| CN105769190A (en) * | 2016-03-29 | 2016-07-20 | 重庆大学 | Device for detecting cephalophyma and encephaledema |
| CN105866772A (en) * | 2016-04-18 | 2016-08-17 | 浙江大学 | Novel method for positioning metal object in human body based on microwave coherence imaging |
Non-Patent Citations (3)
| Title |
|---|
| B.J.MOHAMMED等: "Circular Antenna Array for Brain Imaging Systems", 《PROCEEDINGS OF THE 2012 IEEE INTERNATIONAL SYMPOSIUM ON ANTENNAS AND PROPAGATION》 * |
| MAREK E. BIALKOWSKI等: "UWB Microwave Imaging System Including Circular Array Antenna", 《18-TH INTERNATIONAL CONFERENCE ON MICROWAVES,RADAR AND WIRELESS COMMUNICATIONS》 * |
| 席恋: "基于微波实验平台的脑中风分类判断研究", 《中国优秀硕士论文全文数据库 医药卫生科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109498010A (en) * | 2018-11-30 | 2019-03-22 | 中国人民解放军陆军军医大学 | Data processing system applied to the diagnosis closed injury of brain of continuous wave spectrum scan |
| CN110393526A (en) * | 2019-08-16 | 2019-11-01 | 北京师范大学 | A kind of high frequency feeble computer signals amplification acquisition system |
| CN111317470A (en) * | 2019-12-27 | 2020-06-23 | 中国人民解放军陆军军医大学 | Portable microwave detection system for clinical pre-hospital diagnosis of cerebral apoplexy type |
| CN112487713A (en) * | 2020-11-26 | 2021-03-12 | 江苏科技大学 | Method for extracting physical size of microstrip antenna based on multilayer PCA Gaussian process |
| CN112487713B (en) * | 2020-11-26 | 2022-05-17 | 江苏科技大学 | Method for extracting physical size of microstrip antenna based on multilayer PCA Gaussian process |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108245155A (en) | A kind of stroke types identification system of Multi-channel microwave aerial array | |
| Virtue et al. | Better than real: Complex-valued neural nets for MRI fingerprinting | |
| Maity et al. | Bio-physical modeling, characterization, and optimization of electro-quasistatic human body communication | |
| CN107220572A (en) | The sensitivity of radio frequency reader and the method for testing of Packet Error Ratio, equipment and system | |
| US10674374B2 (en) | Portable spectrum recording and playback apparatus and associated site model | |
| WO2011126860A2 (en) | Medical diagnosis using biometric sensor protocols | |
| US20130006570A1 (en) | Unbound oscilloscope probe systems-using rf and or optical test point links - provides operational access and mobility | |
| CN106456110A (en) | Ultrasonic diagnostic device and system | |
| Abbosh | Microwave systems for head imaging: Challenges and recent developments | |
| EP3823522B1 (en) | Mobile ultrawideband radar for monitoring thoracic fluid levels and cardio-respiratory function | |
| CN113017598B (en) | Radio frequency energy deposition prediction and monitoring method, device, equipment and medium | |
| US20140066728A1 (en) | System for Measuring Skin Hydration | |
| CN109039486A (en) | Multi-user's multiple-input, multiple-output equipment communication performance test macro, method and device | |
| CN107276693A (en) | Terminal radio frequency head end test method, equipment and system | |
| CN110522436A (en) | Pulse signal acquisition system | |
| CN111601539A (en) | Method for automatically diagnosing disease of subject and system for implementing method | |
| CN105738738B (en) | A kind of radio frequency testing device and system | |
| KR20140019519A (en) | Interface system for remote diagnosis auxiliary duvice using wpan | |
| Wu et al. | A single-pair antenna microwave medical detection system based on unsupervised feature learning | |
| CN206946493U (en) | The test equipment and system of a kind of radio frequency reader | |
| KR20210081860A (en) | Seafood Inspection Management Data Process Device | |
| CN109223124A (en) | One kind puncturing dual boot control system based on computer orthopaedics | |
| CN110456976A (en) | Checklist processing method and processing device, storage medium and electronic device | |
| KR102602786B1 (en) | Transceiver of wire communication signal and control method of communication equipment connected with transceiver of wire communication signal | |
| Blumrosen et al. | New wearable body sensor for continuous diagnosis of internal tissue bleeding |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180706 |