CN108325020A - A kind of Intravenous Anesthesia multi-parameter index supervisory system of closed - Google Patents
A kind of Intravenous Anesthesia multi-parameter index supervisory system of closed Download PDFInfo
- Publication number
- CN108325020A CN108325020A CN201810193928.4A CN201810193928A CN108325020A CN 108325020 A CN108325020 A CN 108325020A CN 201810193928 A CN201810193928 A CN 201810193928A CN 108325020 A CN108325020 A CN 108325020A
- Authority
- CN
- China
- Prior art keywords
- parameter
- index
- module
- signal
- control host
- 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.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M5/00—Devices for bringing media into the body in a subcutaneous, intra-vascular or intramuscular way; Accessories therefor, e.g. filling or cleaning devices, arm-rests
- A61M5/14—Infusion devices, e.g. infusing by gravity; Blood infusion; Accessories therefor
- A61M5/142—Pressure infusion, e.g. using pumps
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4821—Determining level or depth of anaesthesia
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M5/00—Devices for bringing media into the body in a subcutaneous, intra-vascular or intramuscular way; Accessories therefor, e.g. filling or cleaning devices, arm-rests
- A61M5/14—Infusion devices, e.g. infusing by gravity; Blood infusion; Accessories therefor
- A61M5/168—Means for controlling media flow to the body or for metering media to the body, e.g. drip meters, counters ; Monitoring media flow to the body
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M5/00—Devices for bringing media into the body in a subcutaneous, intra-vascular or intramuscular way; Accessories therefor, e.g. filling or cleaning devices, arm-rests
- A61M5/14—Infusion devices, e.g. infusing by gravity; Blood infusion; Accessories therefor
- A61M5/142—Pressure infusion, e.g. using pumps
- A61M2005/14208—Pressure infusion, e.g. using pumps with a programmable infusion control system, characterised by the infusion program
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M5/00—Devices for bringing media into the body in a subcutaneous, intra-vascular or intramuscular way; Accessories therefor, e.g. filling or cleaning devices, arm-rests
- A61M5/14—Infusion devices, e.g. infusing by gravity; Blood infusion; Accessories therefor
- A61M5/142—Pressure infusion, e.g. using pumps
- A61M2005/14288—Infusion or injection simulation
- A61M2005/14292—Computer-based infusion planning or simulation of spatio-temporal infusate distribution
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
- A61M2230/00—Measuring parameters of the user
- A61M2230/08—Other bio-electrical signals
Abstract
The invention discloses a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed, including parameter signal acquisition module, control host module, injection device module；The parameter signal acquisition module is connected with control host module, parameter signal acquisition module is converted into for acquiring human body raw EEG signal, NIRS signals, EMG signal, and by above-mentioned signal and can be transmitted to control host module with the parameter index of sedation depth monitoring, analgesia depth and degree of flaccid muscles；The control host module is calculated the dose and injection rate that human body needs in real time, and provides monitoring index in real time for parameter identification, model prediction, controller feedback；The control terminal of control host module is connected with injection device module, and by controlling, host module obtains pharmaceutical quantities to the injection device module and then injection speed is carried out and injected to human body.The present invention uses the closed-loop system of multi-parameter index monitoring, solves the uncertainty of traditional parameters, has many advantages, such as real-time, and precision is high.
Description
Technical field
The present invention relates to anesthesia control field, the multi-parameter index supervisory system of closed of especially a kind of Intravenous Anesthesia.
Background technology
According to statistics, China have every year about 9 million peoples/time general anesthesia operation, it is contemplated that can reach to amount for surgical in 2018
1.55 hundred million people/time.Clinically some important operations, such as cardiac operation, neurosurgery and patient with brain trauma
Monitoring, need fine administration of anaesthesia, just can guarantee safety in patient's art and postoperative.These individuals to Anesthesia medicine
Change, more stringent requirements are proposed for Precise control.
It is well known that anesthesia depth monitoring can use three kinds of indexs in art, i.e., calm, analgesia and of flaccid muscles.For
It is most about calm study on monitoring for anesthesia clinically, there are more moneys to be based on brain electricity generation one in the market single
Index estimates sedation depth, for example, Bispectral Index, Nacrotrend indexes, entropy index, small echo index etc..In analgesia side
Face is difficult the single analgesia level of extraction from EEG since analgesic is different from downern to cerebration receptor.It closes
In analgesic, monitoring has heart rate variability, mean arterial blood pressure etc..The monitoring of flesh pine is relatively easy, can have by myoelectricity component
Flesh pine degree in effect monitoring art, and the interaction without drug between muscle relaxant and calmness, analgestic.Meanwhile downern is (such as
Propofol) between analgesic (such as Remifentanil) there is synergistic effect, this can cause fixed PKPD models estimation to generate mistake
Difference.
Although many now concerning the research in terms of Anesthesia Monitoring and closed-loop control, most of system existing at present
System is only to consider index calm or that analgesia is single, does not also account for the synergistic effect between antalgic and sedative drug.Therefore
There is presently no an anesthesia monitoring systems can provide the multi-parameters indexs such as calmness, analgesia in anesthesia simultaneously.
Invention content
Present invention aims at a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed is provided, can acquire simultaneously EEG and
NIRS signals extract sedation and analgesia index of correlation from EEG and NIRS, based on Propofol-Remifentanil combined anesthesia, consider
The synergistic effect of drug combination realizes corresponding closed-loop control to parameters such as analgesia, calmness respectively.
To achieve the above object, following technical scheme is used：System of the present invention include parameter signal acquisition module,
Control host module, injection device module；The parameter signal acquisition module is connected with control host module, parameter signal acquisition
Module for acquiring human body raw EEG signal, NIRS signals, EMG signal, and convert above-mentioned signal to can monitor it is calm deep
The parameter index of degree, analgesia depth and degree of flaccid muscles is transmitted to control host module；The control host module is used for parameter
The dose and injection rate that human body needs in real time is calculated in identification, model prediction, controller feedback, and provides monitoring in real time and refer to
Mark；The control terminal of control host module is connected with injection device module, and the injection device module is obtained by controlling host module
It takes pharmaceutical quantities and injection speed then to carry out to inject human body.
Further, the parameter signal acquisition module is made of EEG-NIRS signal acquiring systems, has EEG signal mould
Block, NIRS signaling modules, electromyography signal module；The EEG signal module includes the acquisition and calm index to original signal
Extraction；The NIRS signaling modules include the extraction of the acquisition and index of easing pain to original NIRS signals；The electromyography signal mould
Block includes to the acquisition of original signal and the extraction of flesh pine index.The combined monitoring of three kinds of signals and its corresponding index is
The basis of system multi parameters control.
Further, the sedation depth parameter LIA is by tetra- kinds of indexs of PLZC, SFS, MPF, BetaRatio
Characteristic component obtains after the study of integration, feature extraction and deep neural network.It solves the inaccurate of single index and passes
The uncertainty and delay problem for parameter of uniting；
Further, the analgesia depth parameter NAF be by original NIRS signals carry out multi-scale wavelet decomposition and
The primary rule transformation of Bill-youth, and by calculating the area under indicatrix A-LL, obtain analgesia depth parameter NAF；It is calculated
Pain/analgesic reaction in index NAF energy accurate evaluation operations；
NAF=[(α * LLmin)2+α*β]1/2/12.8
Wherein α and β is the constant variables of analgesia depth parameter NAF, is worth the constant for 0-100；LLmin is characterized curve A-
The minimum value of areas of the LL under adjacent four child windows.
Further, the flesh pine parameter RPSD is that the size for the ratio for accounting for gross energy according to myoelectricity component judges patient
Degree of flaccid muscles, that is, after extracting frequency range (>=40Hz) where electromyography signal, pass through the relative power spectrum for calculating myoelectricity component
Density assesses patient's degree of flaccid muscles.
Further, the control host module is excellent for model, parameter identification, controller, index comprising monitoring interface, medicine
Change and database；The monitoring interface includes to have the function of to be manually entered and parameter index monitoring function；The medicine is for mould
Type is the patient model for the synergistic effect for considering that drug is used in mixed way；The parameter identification is the model based on least square method
Parameter identification；The controller is Model Predictive Control and the closed loop controller based on extension prediction adaptive algorithm；It is described
Index optimization be pass through feature extraction and unsupervised formula deep neural network study carry out parameter optimization；The database is
Patient data's information is established in storage.
Further, the injection device module is made of the calmness of Parallel Design, analgesia, muscle relaxants syringe, is led to
Cross with control host module establish RS232 serial communication protocols, by control host module calculate calmness, analgesia, flesh pine three
Kind drug injection velocity information passes to syringe and carries out drug injection to patient.
Further, the patient model for the synergistic effect that the consideration drug is used in mixed way, for Propofol and Rui Fen
The synergistic effect that too Buddhist nun's drug combination generates, on the basis of traditional PKPD models, two kinds of drug concentration-response relations can be by
The relationship standardized below is indicated：
Wherein, DOA (t) is anesthesia calming effects；T is time (s) variable；θ is drug concentration effect parameter；E0For index
Initial constant (0-100)；Emax(θ) is the maximum possible drug effect under θ；Uprop(t) it is the concentration effect of Propofol；URem
(t) it is the concentration effect of Remifentanil；Uprop(t)+URem(t) it is hybrid medicine effect；U50(θ) is at θ 50% maximum effect
Answer constant；γ (θ) is maximum possible drug effect of the concentration-response relation at θ.
Further, the parameter identification, that is, patient model identification module is to be recognized using least square method, is distinguished
The patient model of knowledge is as follows：
a4y(k-4)+a3y(k-3)+a2y(k-2)+a1y(k-1)+a0Y (k)=b1u(k-1)+b2u(k-2)+b3u(k-3)
Wherein, the infusion velocity of the input of u representative models, i.e. Propofol and Remifentanil；K is in infusion velocity u matrix
Variable, round numbers；Y represents Anesthesia Monitoring index, i.e., calm or analgesia；a4, a3, a2, a1, a0, b1, b2, b3For the model of identification
Parameter；
Further, the index optimization be using various visual angles learning method, to index feature carry out double optimization and
Confirm；Include the following steps：
(1) it extracts the different characteristic of anesthesia and establishes the feature pool of all features；The feature of different age group is regarded as more
A visual angle；
(2) use the distribution probability that meets accident of the multilayer neural network index feature label that training obtains for the first time as anesthesia
State feature, combine the age characteristics of patient and retraining carried out to multilayer Holy Bible network；
(3) by establishing the incidence relation of different characteristic and state, most suitable character representation life under each state is found
At state.
Compared with prior art, the invention has the advantages that：
1, the Anesthesia Monitoring for realizing multi-parameter index provides status monitoring basis for precise controlling anesthesia.
2, by the EEG signal of homemade EEG-NIRS system acquisitions, exist to PLZC, MPF, SFS, BetaRatio tetra-
The index of time domain frequency domain carries out feature extraction and optimization, establishes the independent calm parameter LIA of system, solves existing index not
Certainty.
3, according to homemade EEG-NIRS systems, multi-resolution decomposition is carried out to original near infrared signal and feature extraction obtains
The independent analgesia parameter NAF (NIRS-Area-Feature) of system, obtained analgesia parameter pass through clinical trial, can be more
Accurately assessing pain/analgesia stimulates degree.
4, consider the synergistic effect between calm and analgesic, establish hybrid medicine patient model, it can be to the not phase
Fast reaction is made in the operation stimulation of prestige；The anesthesia control system for being manually entered and automatically controlling simultaneously combination mitigates anesthesia significantly
The live load of teacher promotes operation safety, can will be put into more energy in prior operation decision.
Description of the drawings
Fig. 1 is the module frame chart of present system.
Fig. 2 is that the multiple-input and multiple-output of present system anaesthetizes closed loop controlling structure figure.
Fig. 3 is that the multi-parameter narcosis of present system optimizes schematic diagram.
Specific implementation mode
The present invention will be further described below in conjunction with the accompanying drawings：
As shown in Figure 1, present system includes patient (11), parameter signal acquisition module (12) controls host module
(13), injection device module (14)；
Patient is connected with parameter signal acquisition module, and parameter signal acquisition module is by EEG-NIRS signal acquiring system structures
At, can synchronous acquisition NIRS (121) signal, EEG signal (122), electromyography signal (123), extract relevant calm (141),
Ease pain (142), the index of flesh loose (143), and parameter index information is passed into control host module.The control host mould
Block includes the collaboration effect for having the function of to be manually entered and the monitoring interface (131) of parameter index monitoring, consideration drug are used in mixed way
The patient model (134) answered, the identification of Model Parameters (132) based on least square method, Model Predictive Control and pre- based on extension
Survey the closed loop controller (133) of adaptive algorithm, the index that is learnt by feature extraction and unsupervised formula deep neural network it is excellent
Change (135) and establishes the database (136) of patient data's information.Control host module is connected with injection device module, passes through
RS232 serial communication protocols are established, medicament categories needed for the calculated patient of host module, injection medicament amount and injection will be controlled
Velocity information passes to injection device module.The injection device module is by the calmness of Parallel Design, analgesia, muscle relaxants note
Emitter carries out infusion of drug to patient, realizes automatic anesthesia closed-loop control.
The multi-index parameter includes sedation depth, analgesia depth, degree of flaccid muscles, brain oxygen metabolism information.
The sedation depth parameter is to pass through multi-resolution decomposition and spy in time domain and frequency domain from the EEG signal of acquisition
Sign extraction has obtained arrangement Lempel-Ziv complexities (PLZC), synchronous speed ratio (SFS), middle frequency spectrum (MPF), β ratios
(BetaRatio) four kinds of indexs, after the study of integration, feature extraction and deep neural network, what is obtained can precisely judge
Sedation depth parameter index LIA.
PLZC specific algorithms are described as follows：
(1) for the flag sequence { x (i) of N points:1≤i≤N }, sum is m！It is a, it is reconstructed.Then according to weight
Flag sequence after structure calculates Lempel-Ziv complexity.There are one upper for the subsequence sum of one flag sequence (x (n))
Limit, and it is denoted as L (n)：
L (n)=c (n) [logαc(n)+1] (1)
Wherein it is the number marked in current markers sequence, is equal to our m in arranging labeling process！.
(2) PLZC can be defined as a standardized c (n)：
(3) wherein n indicates the total length of flag sequence.When N is very big, PLZC can be reduced to:
β ratio specific algorithms are as follows：
BetaRatio=log (P30-47Hz/P11-20Hz) (4)
MPF specific algorithms are as follows：
SFS specific algorithms are as follows：
SEF=log (P0.5-47Hz/P40-47Hz) (7)
By to can via sorting techniques such as SVM and random forests by the characteristic component input feature vector matrix of four kinds of signals
To show that the accurate indicatrix for meeting patient's sedation in art, optimization obtain the index of the sedation depth monitoring of system alone
LIA solves the uncertainty and delay problem of the inaccurate and traditional parameters of single index.
The analgesia depth parameter NAF (NIRS-Area-Feature) is from the NIRS signals of acquisition, to original letter
Number carry out multi-scale wavelet decomposition.Approximation component (0-0.196Hz) under scale 8 is extracted into progress wavelet inverse transformation and obtains original
Beginning signal brain function pertinent trends component；The primary rule of Bill-youth is carried out according to the trend component of two signals to convert to obtain HbO2
With the variation of Hb relative concentrations.Based on this drafting characteristic signal curve generate feature L waveforms, measure two L waves between
Time.Utilize more than the interval series of 64 seconds standard vector LL, the calculating of standard value (θ)：
Then, each LL samples divided by standard value (θ)：
LLi'=LLi/θ (9)
Energy distribution calculating is carried out to LL interval series after mean value and normalization, obtains A-LL indicatrixes.By the letter of 64s
Number amount is divided into 4 sections, every section of 16s.By calculating the area value under indicatrix A-LL, the area under each 16s child windows is recorded
S1, S2, S3, S4.In order to obtain a part for total window area, the metric of a 0-100 is calculated, we define LLmin
=min (S1, S2, S3, S4)：
NAF=[(α * LLmin)2+α*β]1/2/12.8 (10)
By experimental analysis and theoretical validation, at α=4.8 and β=1.6, calculated index NAF can accurate evaluation hand
Pain/analgesic reaction in art.After calculating every time, the window of mobile 64s can be measured continuously.The sampling of final argument
Rate depends on the period of window movement.
The flesh pine parameter, the present invention carry out the monitoring of flesh pine degree in art by relative power spectral density (RPSD),
The size that the ratio of gross energy is accounted for according to myoelectricity component, to judge the degree of flaccid muscles of patient.According to the oscillation frequency of EEG signal
Rate is clinically mainly segmented into five seed bandwidth：Delta waves (0-4Hz), theta waves (4-8Hz), alpha waves (8-
13Hz), beta waves (13-30Hz) and gamma waves (30-47Hz).The wide power spectral density of these subbands (PSD) uses pwelch
Method calculate.The specific calculation of RSPD is as follows：
P () represents energy, and RPSD () represents relative power spectral density.Here f1And f2Respectively represent low frequency and height
Frequency.P (Isosorbide-5-Nitrae 7) represents the energy for including five seed bandwidth (delta waves, theta waves, alpha waves, gamma waves) from 1Hz-47Hz
Amount.According to the above method, after frequency range (>=40Hz) where extracting electromyography signal, by the relative power spectrum for calculating myoelectricity component
Density (RPSD), you can accurate evaluation patient's degree of flaccid muscles, and the phase interaction without drug between muscle relaxant and calmness, analgestic
With.By clinical trial, the method is accurate and feasible to the judgement of flesh pine.
The hybrid medicine patient model of the system, the collaboration generated for Propofol and Remifentanil drug combination are imitated
It answers, on the basis of traditional PKPD models, two kinds of drug concentration-response relations can carry out table by the relationship of following standardization
Show：
Wherein Uprop(t)+URem(t) it is hybrid medicine effect；γ (θ) is slope of the concentration-response relation at θ；DOA
(t) it is anesthesia calming effects.
The controller of the control system, as shown in Fig. 2, from the angle of practicability and reliability, in the present invention
Using Model Predictive Control and based on extension prediction adaptive control algorithm.For sedation/analgesia closed-loop control, obtaining reliably
Identification model after, control strategy is as follows：
(1) model prediction is carried out.In current sample time, based on identification model, it is expected sedation/analgesia value, applied
Outputting and inputting in the following period is predicted in input and acquired output.
(2) optimum control is carried out, makes the following output close to desired output.
(3) rolling optimization.At current time, only with first input for most having control sequence, in next sampling instant,
It repeats the above steps, obtains optimum control input.
(4) optimized by the PREDICTIVE CONTROL being repeated, it is expected that obtaining optimal control results.
Control input signal u can be expressed as：U (s)=K1 (s) r (s)+K2 (s) y (s).
For the closed loop controller of flesh pine, common muscle inhibition agent is atracurium, in this regard, using a succinct dimension
It receives model, single-input single-output control is showed to flesh pine nut.
The parameter identification of the system, that is, patient model recognizes module, and the present invention uses the clinical number of each patient oneself
According to, by recognize module, the personal physiological information of patient is integrated into model, obtains respective model parameter.Using minimum two
Multiplication is recognized, and the patient model of identification is a Fourth Order Differential Equations：
The wherein input of u representative models, i.e., the infusion velocity of two kinds drugs；Y represents Anesthesia Monitoring index, i.e., calm or town
Bitterly；a4, a3, a2, a1, a0, b1, b2, b3For the model parameter of identification.
The parameter optimization of the system is that digitized index is mapped to the different conditions of anesthesia, for controller
Control.As shown in figure 3, using various visual angles learning method, double optimization and confirmation are carried out to index feature.It is described in detail below：
(1) multi-source data based on acquisition extracts the different characteristic of anesthesia and establishes the feature pool of all features.It will be different
The feature of age bracket regards multiple visual angles as.
(2) based on multi-modal iterative calculation, using the multilayer neural network index feature label that training obtains for the first time
State feature of the distribution probability as anesthesia of meeting accident, combine the age characteristics of patient and retraining carried out to multilayer Holy Bible network.
(3) Status Flag based on various visual angles finds each state by establishing the incidence relation of different characteristic and state
Under most suitable character representation generate state.
Embodiment described above is only that the preferred embodiment of the present invention is described, not to the model of the present invention
It encloses and is defined, under the premise of not departing from design spirit of the present invention, technical side of the those of ordinary skill in the art to the present invention
The various modifications and improvement that case is made should all be fallen into the protection domain of claims of the present invention determination.
Claims (10)
1. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed, it is characterised in that：The system comprises parameter signal acquisitions
Module, control host module, injection device module；The parameter signal acquisition module is connected with control host module, parameter letter
Number acquisition module for acquiring human body raw EEG signal, NIRS signals, EMG signal converts and can monitor above-mentioned signal to
The parameter index of sedation depth, analgesia depth and degree of flaccid muscles is transmitted to control host module；The control host module is used
The dose and injection rate that human body needs in real time is calculated, and provides in real time in parameter identification, model prediction, controller feedback
Monitoring index；The control terminal of control host module is connected with injection device module, and the injection device module is by controlling host
Module obtains pharmaceutical quantities and then injection speed is carried out and injected to human body.
2. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed according to claim 1, it is characterised in that：The ginseng
Number signal acquisition module is made of EEG-NIRS signal acquiring systems, has EEG signal module, NIRS signaling modules, myoelectricity letter
Number module；The EEG signal module includes the extraction of the acquisition and calm index to original signal；The NIRS signaling modules packet
Extraction containing acquisition and index of easing pain to original NIRS signals；The electromyography signal module include to the acquisition of original signal and
The extraction of flesh pine index.
3. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed according to claim 1, it is characterised in that：The town
Quiet depth parameter LIA is by the characteristic component to tetra- kinds of indexs of PLZC, SFS, MPF, BetaRatio, and by integrating, feature carries
Take and deep neural network study after obtain.
4. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed according to claim 1, it is characterised in that：The town
Pain depth parameter NAF and is passed through by carrying out multi-scale wavelet decomposition and Bill-youth primary rule transformation to original NIRS signals
The area under indicatrix A-LL is calculated, analgesia depth parameter NAF is obtained；In calculated index NAF energy accurate evaluation operation
Pain/analgesic reaction；
NAF=[(α * LLmin)2+α*β]1/2/12.8 (1)
Wherein α and β is the constant variables of analgesia depth parameter NAF, is worth the constant for 0-100；LLmin is characterized curve A-LL and exists
The minimum value of area under adjacent four child windows.
5. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed according to claim 1, it is characterised in that：The flesh
Loose parameter RPSD is that the size for the ratio for accounting for gross energy according to myoelectricity component judges the degree of flaccid muscles of patient, that is, extracts flesh
After frequency range (>=40Hz) where electric signal, by calculating the relative power spectral density of myoelectricity component, patient's degree of flaccid muscles are assessed.
6. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed according to claim 1, it is characterised in that：The control
Host module processed includes monitoring interface, medicine for model, parameter identification, controller, index optimization and database；Monitoring circle
Bread is containing having the function of to be manually entered and parameter index monitoring function；The medicine is the association for considering drug and being used in mixed way for model
With the patient model of effect；The parameter identification is the identification of Model Parameters based on least square method；The controller is
Model Predictive Control and the closed loop controller that adaptive algorithm is predicted based on extension；The index optimization is to pass through feature extraction
Learn to carry out parameter optimization with unsupervised formula deep neural network；The database is that patient data's information is established in storage.
7. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed according to claim 1, it is characterised in that：The note
Injection device module is made of the calmness of Parallel Design, analgesia, muscle relaxants syringe, by being established with control host module
RS232 serial communication protocols pass calmness, analgesia, flesh three kinds of drug injection velocity informations of pine that control host module calculates
It passs syringe and drug injection is carried out to patient.
8. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed according to claim 6, it is characterised in that：Described
The patient model for considering the synergistic effect that drug is used in mixed way, the collaboration generated for Propofol and Remifentanil drug combination are imitated
It answers, on the basis of traditional PKPD models, two kinds of drug concentration-response relations can carry out table by the relationship of following standardization
Show：
Wherein, DOA (t) is anesthesia calming effects；T is time (s) variable；θ is drug concentration effect parameter；E0It is initial for index
Constant (0-100)；Emax(θ) is the maximum possible drug effect under θ；Uprop(t) it is the concentration effect of Propofol；URem(t) it is
The concentration effect of Remifentanil；Uprop(t)+URem(t) it is hybrid medicine effect；U50(θ) be at θ 50% ceiling effect it is normal
Amount；γ (θ) is maximum possible drug effect of the concentration-response relation at θ.
9. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed according to claim 6, it is characterised in that：Described
Parameter identification, that is, patient model identification module is to be recognized using least square method, and the patient model of identification is as follows：
Wherein, the infusion velocity of the input of u representative models, i.e. Propofol and Remifentanil；K is the change in infusion velocity u matrix
Amount, round numbers；Y represents Anesthesia Monitoring index, i.e., calm or analgesia；a4, a3, a2, a1, a0, b1, b2, b3Join for the model of identification
Number.
10. a kind of Intravenous Anesthesia multi-parameter index supervisory system of closed according to claim 6, it is characterised in that：It is described
Index optimization be that double optimization and confirmation are carried out to index feature using various visual angles learning method；Include the following steps：
(1) it extracts the different characteristic of anesthesia and establishes the feature pool of all features；Regard the feature of different age group as multiple regard
Angle；
(2) shape of the distribution probability as anesthesia that meet accident of the multilayer neural network index feature label that training obtains for the first time is used
State feature, the age characteristics for combining patient carry out retraining to multilayer Holy Bible network；
(3) it by establishing the incidence relation of different characteristic and state, finds most suitable character representation under each state and generates shape
State.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810193928.4A CN108325020B (en) | 2018-03-09 | 2018-03-09 | Vein anesthesia multi-parameter index closed-loop monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810193928.4A CN108325020B (en) | 2018-03-09 | 2018-03-09 | Vein anesthesia multi-parameter index closed-loop monitoring system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108325020A true CN108325020A (en) | 2018-07-27 |
| CN108325020B CN108325020B (en) | 2021-01-08 |
Family
ID=62929051
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810193928.4A Active CN108325020B (en) | 2018-03-09 | 2018-03-09 | Vein anesthesia multi-parameter index closed-loop monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108325020B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109157231A (en) * | 2018-10-24 | 2019-01-08 | 阿呆科技（北京）有限公司 | Portable multi-channel Depression trend assessment system based on emotional distress task |
| CN109887599A (en) * | 2019-02-25 | 2019-06-14 | 中国计量大学 | A kind of tcm prescription curative effect deduction method neural network based |
| CN112190233A (en) * | 2020-10-20 | 2021-01-08 | 黄淮学院 | Remote analysis management system for observing postoperative pain at home of patient |
| CN112870480A (en) * | 2021-01-22 | 2021-06-01 | 燕山大学 | Fuzzy control remote control anesthetic drug infusion pump system |
| CN113440690A (en) * | 2021-08-03 | 2021-09-28 | 复旦大学 | Intelligent quantitative drug administration electric needle injection device based on electromyographic signal feedback |
| CN113648484A (en) * | 2021-09-06 | 2021-11-16 | 北京航空航天大学 | Portable target-controlled infusion intravenous anesthesia device and target-controlled infusion control method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102793534A (en) * | 2012-07-11 | 2012-11-28 | 广西威利方舟科技有限公司 | Anesthesia depth monitoring device and method |
| CN103908249A (en) * | 2014-01-28 | 2014-07-09 | 广西威利方舟科技有限公司 | Anaesthetic balance control device and control method |
| CN105852804A (en) * | 2016-03-24 | 2016-08-17 | 美合实业（苏州）有限公司 | Portable anesthesia depth monitor |
| CN106902421A (en) * | 2017-01-17 | 2017-06-30 | 燕山大学 | The individualized anesthesia closed-loop control system of one kind |
-
2018
- 2018-03-09 CN CN201810193928.4A patent/CN108325020B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102793534A (en) * | 2012-07-11 | 2012-11-28 | 广西威利方舟科技有限公司 | Anesthesia depth monitoring device and method |
| CN103908249A (en) * | 2014-01-28 | 2014-07-09 | 广西威利方舟科技有限公司 | Anaesthetic balance control device and control method |
| CN105852804A (en) * | 2016-03-24 | 2016-08-17 | 美合实业（苏州）有限公司 | Portable anesthesia depth monitor |
| CN106902421A (en) * | 2017-01-17 | 2017-06-30 | 燕山大学 | The individualized anesthesia closed-loop control system of one kind |
Non-Patent Citations (2)
| Title |
|---|
| ZHENHU LIANG,YUE GU,XUEJING DUAN: "Design of multichannel functional near-infrared spectroscopy system with application to propofol and sevoflurane anesthesia monitoring", 《NEUROPHOTONICS》 * |
| 梁振虎: "EEG熵算法及麻醉状态监测应用研究", 《中国博士学位论文全文数据库医药卫生科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109157231A (en) * | 2018-10-24 | 2019-01-08 | 阿呆科技（北京）有限公司 | Portable multi-channel Depression trend assessment system based on emotional distress task |
| CN109887599A (en) * | 2019-02-25 | 2019-06-14 | 中国计量大学 | A kind of tcm prescription curative effect deduction method neural network based |
| CN112190233A (en) * | 2020-10-20 | 2021-01-08 | 黄淮学院 | Remote analysis management system for observing postoperative pain at home of patient |
| CN112870480A (en) * | 2021-01-22 | 2021-06-01 | 燕山大学 | Fuzzy control remote control anesthetic drug infusion pump system |
| CN112870480B (en) * | 2021-01-22 | 2021-12-14 | 燕山大学 | Fuzzy control remote control anesthetic drug infusion pump system |
| CN113440690A (en) * | 2021-08-03 | 2021-09-28 | 复旦大学 | Intelligent quantitative drug administration electric needle injection device based on electromyographic signal feedback |
| CN113648484A (en) * | 2021-09-06 | 2021-11-16 | 北京航空航天大学 | Portable target-controlled infusion intravenous anesthesia device and target-controlled infusion control method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108325020B (en) | 2021-01-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108325020A (en) | A kind of Intravenous Anesthesia multi-parameter index supervisory system of closed | |
| CN109009099A (en) | A kind of intelligent anesthesia system based on EEG-NIRS | |
| Izard et al. | Infant cardiac activity: Developmental changes and relations with attachment. | |
| CN105147248B (en) | Depression assessment system and its appraisal procedure based on physiologic information | |
| Katz | Fractals and the analysis of waveforms | |
| Lu et al. | Automatic selection of the threshold value $ r $ for approximate entropy | |
| CN110547868B (en) | Continuous anesthesia device | |
| CN104825135A (en) | Method for establishing novel pain assessment system | |
| CN101449974A (en) | Method for automatic real-time estimating anesthesia depth | |
| CN103153178A (en) | An apparatus for combining drug effect interaction between anaesthetics and analgesics and electroencephalogram features for precise assessment of the level of consciousness during anaesthesia | |
| CN111616681B (en) | Anesthesia state monitoring system based on portable electroencephalogram acquisition equipment and deep learning | |
| CN110193113A (en) | Multi-channel intelligent drug delivery system | |
| CN114343573B (en) | Closed-loop digital medicine system, equipment and medium for drug addiction patients | |
| CN112397176A (en) | Intelligent oxytocin dose regulation and control method and system based on uterine contraction signals and LightGBM | |
| CN116747377A (en) | Closed-loop intelligent anesthesia control system | |
| CN104799814A (en) | Qi deficiency, blood deficiency state quantitative measurement method and device | |
| CN109620263B (en) | Physical sign safety analysis method for post workers in enterprises and public institutions | |
| Banerjee et al. | Detecting eye movement direction from stimulated Electro-oculogram by intelligent algorithms | |
| CN108241431B (en) | Task adjusting method and device | |
| CN117219227B (en) | Anesthesia administration control system based on fuzzy neural network | |
| Zhou et al. | Application of back propagation neural network and information entropy in deep detection of anesthesia | |
| Spotts et al. | Drugs and personality: Comparison of drug users, nonusers, and other clinical groups on the 16PF | |
| Van Den Broek | Monitoring Anesthetic Depth Modification, Evaluation and Application of the Correlation Dimension | |
| Nunes et al. | Fuzzy modelling for controlled anaesthesia in hospital operating theatres | |
| Belgacem et al. | ECG based human identification using random forests |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |