CN111599437A - Dynamic beta power signal tracking method, system and terminal - Google Patents

Dynamic beta power signal tracking method, system and terminal Download PDF

Info

Publication number
CN111599437A
CN111599437A CN202010417020.4A CN202010417020A CN111599437A CN 111599437 A CN111599437 A CN 111599437A CN 202010417020 A CN202010417020 A CN 202010417020A CN 111599437 A CN111599437 A CN 111599437A
Authority
CN
China
Prior art keywords
signal
output
model
beta power
power signal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010417020.4A
Other languages
Chinese (zh)
Inventor
苏斐
王红
陈民
祖林禄
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Agricultural University
Original Assignee
Shandong Agricultural University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shandong Agricultural University filed Critical Shandong Agricultural University
Priority to CN202010417020.4A priority Critical patent/CN111599437A/en
Publication of CN111599437A publication Critical patent/CN111599437A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/40—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction

Landscapes

  • Health & Medical Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Surgery (AREA)
  • Urology & Nephrology (AREA)
  • Engineering & Computer Science (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • Feedback Control In General (AREA)

Abstract

The application discloses a dynamic beta power signal tracking method, a system and a terminal, wherein a Controlled Autoregressive (CAR) model is established, a first output signal of a proportional-integral (PI) controller is received to output a first beta power signal, and a signal difference value between the first beta power signal and an expected beta power signal is obtained; and inputting the signal difference value into a PI controller, outputting a second output signal to the CAR model by the PI controller according to the signal difference value, and outputting a second beta power signal by the CAR model. After the beta power signal output from the CAR model is compared with the expected beta power signal, the obtained difference value is input to a receiving PI controller, and the PI controller obtains an output signal to the CAR model according to the received difference value signal. The CAR model outputs an output signal that approximates the desired beta power signal based on the received signal. Thus, the beta power signal can be tracked whether the expected beta power signal is changed or not.

Description

Dynamic beta power signal tracking method, system and terminal
Technical Field
The present application relates to the field of technologies, and in particular, to a method, a system, and a terminal for tracking a dynamic beta power signal.
Background
Parkinson's Disease (PD) is a common degenerative disease of the nervous system, and the application of Deep Brain Stimulation (DBS) to the internal Globus Pallidus (GPi) or subthalamic nucleus (STN) can alleviate PD symptoms. DBS stimulation parameters are selected primarily based on clinician experience and cannot be adjusted based on any neural or motor feedback, thus consuming more energy and increasing the frequency with which the patient changes stimulators. If the relationship between the input stimulation frequency and the output beta power signal of the substrate kernel model can be identified, appropriate stimulation parameters are selected according to the change of the beta power signal, and the control curative effect can be improved.
Closed-loop DBS can better inhibit the motor symptoms of PD patients than open-loop high-frequency stimulation, if a reference signal which is suitable for reflecting pathological states can be found, stimulation according to needs is realized, and energy consumption can be reduced while side effects can be reduced. Therefore, how to accurately track the β power signal is a technical problem to be solved in the art.
Disclosure of Invention
In order to solve the technical problems, the following technical scheme is provided:
in a first aspect, an embodiment of the present application provides a dynamic β power signal tracking method, where the method includes: establishing a Controlled Autoregressive (CAR) model, wherein the CAR model is used for receiving a first output signal of a PI controller and outputting a first beta power signal, and the first output signal is a signal output by the PI controller at any moment; acquiring a signal difference value of the first beta power signal and an expected beta power signal; and inputting the signal difference value into the PI controller, outputting a second output signal to the CAR model by the PI controller according to the signal difference value, and outputting a second beta power signal by the CAR model.
By adopting the implementation mode, after the beta power signal output from the CAR model is compared with the expected beta power signal, the obtained difference value is input to the PI receiving controller, and the PI receiving controller obtains an output signal according to the received difference value signal and sends the output signal to the CAR model. The CAR model outputs an output signal that approximates the desired beta power signal based on the received signal. Thus, the beta power signal can be tracked whether the expected beta power signal is changed or not.
With reference to the first aspect, in a first possible implementation manner of the first aspect, the establishing a controlled autoregressive CAR model includes: identifying the relation between the input stimulation frequency of the neural network and the beta power of the output GPi neuron, and collecting input and output data from the CTx-BG-Th neural network, wherein the input and output are expressed as follows:
Figure BDA0002495372380000021
where u (k) denotes the input signal stimulus frequency, y (k) denotes the output GPi neuron β band oscillation power, naAnd nbIs the order of the output input signal, (k) represents white noise; determining model parameters of the CAR model and an order of the model parameters.
With reference to the first possible implementation manner of the first aspect, in a second possible implementation manner of the first aspect, the determining a model parameter of a CAR model and an order of the model parameter includes: estimating CAR model parameters using a recursive least squares method; obtaining an output root mean square error of the CAR model through a real output signal and an output signal of the CAR model, wherein the output root mean square error is used for quantifying the prediction precision of the CAR model; and determining the order of the model parameter according to the output root mean square error.
With reference to the first aspect, in a third possible implementation manner of the first aspect, the obtaining a signal difference value of the first β power signal and a desired β power signal includes: inputting the first beta power signal and the desired beta power signal to a subtractor; the subtractor outputs the signal difference.
With reference to the first aspect, in a fourth possible implementation manner of the first aspect, the method further includes determining a parameter of the PI controller, so that the PI controller is stable.
In a second aspect, an embodiment of the present application provides a dynamic β -power signal tracking system, including: the model establishing module is used for establishing a Controlled Autoregressive (CAR) model, the CAR model is used for receiving a first output signal of a PI controller and outputting a first beta power signal, and the first output signal is a signal output by the PI controller at any moment; an obtaining module, configured to obtain a signal difference between the first β power signal and an expected β power signal; and the signal tracking module is used for inputting the signal difference value into the PI controller, the PI controller outputs a second output signal to the CAR model according to the signal difference value, and the CAR model outputs a second beta power signal.
With reference to the second aspect, in a first possible implementation manner of the second aspect, the model building module includes: a first determining unit, configured to identify a relationship between a neural network input stimulation frequency and output GPi neuron beta power, and collect CAR model input and output data from the CTx-BG-Th neural network, where the CAR model input and output are expressed as:
Figure BDA0002495372380000031
where u (k) denotes the input signal stimulus frequency, y (k) denotes the output GPi neuron β band oscillation power, naAnd nbIs the order of the output input signal, (k) represents white noise; and the second determining unit is used for determining the model parameters of the CAR model and the order of the model parameters.
With reference to the first possible implementation manner of the second aspect, in a second possible implementation manner of the second aspect, the second determining unit includes: a parameter estimation subunit for estimating CAR model parameters using a recursive least squares method; the acquisition subunit is used for acquiring an output root mean square error of the CAR model through a real output signal and an output signal of the CAR model, wherein the output root mean square error is used for quantifying the prediction precision of the CAR model; and the determining subunit is used for determining the order of the model parameter according to the output root mean square error.
With reference to the second aspect, in a third possible implementation manner of the second aspect, the obtaining module includes: a signal input unit for inputting the first beta power signal and the desired beta power signal to a subtractor; and the difference output unit is used for outputting the signal difference by the subtracter.
In a third aspect, an embodiment of the present application provides a terminal, including: a processor; a memory for storing computer executable instructions; when the processor executes the computer-executable instructions, the processor performs the method of the first aspect or any one of the possible implementations of the first aspect to achieve tracking of a dynamic beta power signal
Drawings
Fig. 1 is a schematic flowchart of a dynamic beta power signal tracking method according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram of a framework of a closed-loop PI control system according to an embodiment of the present disclosure;
fig. 3 is a schematic diagram of a linear transformation system of a closed-loop DBS system according to an embodiment of the present disclosure;
FIG. 4 shows CAR model parameters n provided in embodiments of the present applicationaAnd nbSchematic diagram of the relation with root mean square error;
FIG. 5 is a schematic diagram of parameter estimation of the CAR model at different iterations according to an embodiment of the present application;
FIG. 6 is a schematic diagram of the CAR model identification effect provided in the embodiments of the present application;
FIG. 7 is a schematic diagram illustrating the response time of the closed-loop control system tracking different beta powers at different initial stimulation frequencies according to an embodiment of the present application;
fig. 8 is a schematic diagram illustrating the effect of tracking other amplitude β power by the closed-loop PI controller robustness test provided in the embodiment of the present application;
FIG. 9 is a schematic diagram of a PI controller tracking a dynamic beta power over time according to an embodiment of the present disclosure;
fig. 10 is a schematic diagram of the tracking of the beta power with the variation of the frequency dynamics of the PI controller provided in the embodiment of the present application;
fig. 11 is a schematic diagram of a dynamic beta power signal tracking system according to an embodiment of the present application;
fig. 12 is a schematic diagram of a terminal according to an embodiment of the present application.
Detailed Description
The present invention will be described with reference to the accompanying drawings and embodiments.
Fig. 1 is a schematic flow chart of a dynamic β power signal tracking method provided in an embodiment of the present application, and referring to fig. 1, the dynamic β power signal tracking method in the embodiment includes:
s101, establishing a Controlled Autoregressive (CAR) model, wherein the CAR model is used for receiving a first output signal of a PI controller and outputting a first beta power signal, and the first output signal is a signal output by the PI controller at any moment.
In order to overcome the defect that the constant beta power is used as the reference signal, the power value which dynamically changes according to the motion state is used as the reference signal, and the PI controller is designed to track the beta power in real time according to the dynamic change of the reference signal. A basal nucleus-cortex-thalamus (CTx-BG-Th) neural network is used as a simulation test bed to estimate the performance of closed-loop control, and simulation results show that the closed-loop control system based on the PI controller can track constant and dynamically-changed beta power so as to inhibit pathological beta oscillation.
The relationship between the input stimulation frequency of the neural network and the output GPi neuron beta power is firstly recognized, and the neural network cannot be expressed by a linear method because of the high nonlinearity of the neural network. A controlled auto-regression (CAR) model is used in the invention, with the inputs and outputs expressed as:
Figure BDA0002495372380000051
where u (k) denotes the input signal stimulus frequency, y (k) denotes the output GPi neuron β band oscillation power, naAnd nbThe order of the output input signal is (k) white noise.
The identification process is as follows: collecting input and output data from the CTx-BG-Th neural network; estimating model parameters
Figure BDA0002495372380000052
And
Figure BDA0002495372380000053
selecting the appropriate order naAnd nb(ii) a The prediction accuracy of the CAR model is identified.
Since the model accuracy depends on the choice of input data, 1000 samples were randomly chosen from 5-200Hz as input stimulation signals, each frequency lasting 0.4s (to ensure two pulses per random frequency), and the input data lasted 400s in total. Input data are transmitted to the CTx-BG-Th model, and the discharging activity of the GPi neurons is recorded to obtain output beta power.
FIG. 2 shows a block diagram of a closed-loop PI control system, wherein the power of the GPi neuron β is selected as a feedback signal y (k), and the desired power y βsp(k) The difference e (k) between the power y (k) and the real β is used as an error signal to be input into the PI controller, and the stimulation frequency u (k) is calculated and determines the next stimulation current Idbs(t) the time of transmitting to the physiological model CTx-BG-Th neural network, and obtaining real β power after the CTx-BG-Th is stimulated by current, thus completing the construction of a complete closed-loop control system.
FIG. 3 shows a linear transformation system of a closed-loop DBS system, i.e. a highly non-linear CTx-BG-Th model is identified using the CAR model, and the transformed linear system is used to determine the parameters of the PI controller.
CAR model parameters were estimated using a recursive least squares method:
Figure BDA0002495372380000061
wherein:
Figure BDA0002495372380000062
Figure BDA0002495372380000063
is a model unknown parameter vector.
The unknown model parameters are estimated using the RLS method,
Figure BDA0002495372380000064
wherein
Figure BDA0002495372380000065
Expressed using the following equation:
Figure BDA0002495372380000066
Figure BDA0002495372380000067
Figure BDA0002495372380000068
the prediction accuracy of the CAR model is quantified using the Root Mean Square Error (RMSE) of the true output signal and the predicted output of the CAR model, expressed as,
Figure BDA0002495372380000069
FIG. 4 shows the root mean square error eRMSEAnd naAnd nbCan be seen from FIG. 4 as the CAR model order (n)aAnd nb) Increase, eRMSEGradually decreases. The model order is selected using Akaike's Information Criterion (AIC),
Figure BDA0002495372380000071
wherein K is na+nb+1, N is the length of the prediction data,
Figure BDA0002495372380000072
when n isa＝3,nbAt 3, the value of AIC is minimal, so the CAR model can be expressed as,
Figure BDA0002495372380000073
from n toa3 and nbIn the case of 3, fig. 5 shows eRMSEMinimal CAR model a1、a2、a3And b1、b2、b3The parameter estimation of (2).
The selection of appropriate PI controller parameters follows.
The structure of the discrete PI controller is as follows:
u(k)＝u(k-1)+kp[e(k)-e(k-1)]+kie(k)
solving the PI parameter by using a Router-Hurwitz stability criterion to ensure the system stability, namely solving the parameter kpAnd ki. The system transfer function is:
Figure BDA0002495372380000074
the closed loop transfer function is:
Figure BDA0002495372380000075
the system characteristic equation is as follows:
D(z)＝1+G(z)＝[1+b0(kp+ki)]z4+[(a1-1)+b1(kp+ki)-b0kp]z3+[(a2-a1)+b2(kp+ki)-b1kp]z2+[(a3-a2)+b3(kp+ki)-b2kp]z-a3-b3kp＝0
finally, k is obtained by utilizing a Router-Hurwitz stability criterionp＝0.80,ki＝0.05。
S102, acquiring a signal difference value of the first beta power signal and an expected beta power signal.
The first beta power signal and the desired beta power signal are input to a subtractor, which outputs the signal difference.
And S103, inputting the signal difference value into the PI controller, outputting a second output signal to the CAR model by the PI controller according to the signal difference value, and outputting a second beta power signal by the CAR model.
FIG. 6 shows the CAR model recognition effect. As shown in FIG. 6, (a) is a training diagram of a model, and a correlation coefficient r (y, y) between a prediction output and a real output is obtainede) 0.84. in addition,two groups of stimulation frequency data are randomly generated and input into the output β power calculated by the neural network model, the same stimulation frequency sequence is also input into the CAR model, wherein the identification effect is shown in the graphs (b) and (c), the correlation coefficients of the two groups are respectively 0.82 and 0.80, and therefore the prediction accuracy of the CAR model is about 80%.
Fig. 7 shows a closed loop PI controller tracking the beta power 110 generated by an open loop 115Hz DBS. Where (a) shows the variation of the closed loop DBS stimulation frequency, stimulation is started after 2s and the initial stimulation frequency is set to 5 Hz.
Fig. 7(b), (c) show the change in the GPi neuron beta power when tracking the beta power 110 both closed and open loops, and it can be seen that the response time open loop (0.09s) is shorter than the closed loop (0.66 s). The calculated closed loop average stimulation frequency was 118.7Hz, the average beta power was 114.3, and the open loop average beta power was 111.3.
Fig. 7(d) shows the relationship of the closed loop initial stimulation frequency to the response time, and it can be seen from the figure that the response time becomes gradually shorter as the initial stimulation frequency becomes larger, and the closed loop response time is less than 0.15s when the initial stimulation frequency is greater than 60 Hz.
Fig. 8 is a PI controller robustness test graph, i.e. tracking different amplitudes of beta power. (a) The dashed line represents the expected beta power and the solid line represents the actual controller-derived beta power (standard error bars show 50 trials), it can be seen that when the beta power is less than 160, the expected beta power has a smaller error with the controller-derived beta power, whereas the error increases.
Fig. 8(b), (c) show the variation of stimulation frequency and GPi neuron beta power as the PI controller tracks beta power 140. The response time was 0.89s and the average stimulation frequency was 74 Hz. Following beta power 150, the response time was 1.15s and the average stimulation frequency was 56 Hz. When the target beta power is greater than 160, the tracking performance of the PI controller becomes poor. Fig. 8(d), (e) show the variation of stimulation frequency and GPi neuron beta power when the PI controller tracks the beta power 180.
The beta power dynamically changes during the preparation period and the execution period of the voluntary movement, and therefore, the fixed beta power is not suitable for the DBS system control as the target value. The innovation of the invention is that the closed-loop PI controller tracks the dynamically changed beta power in real time according to the dynamically changed power value of the motion state as a reference signal. We tested tracking beta power dynamically over time and frequency, respectively.
Fig. 9 shows that the PI controller tracks the dynamically changing beta power over time, (a), (c), (e) and (g) are changes in the beta power of the GPi neuron, where the beta power reference value changes every 10s, 5s, 2s and 1s, respectively. (b) The (f), (d) and (h) are changes corresponding to the stimulation frequency, respectively. The correlation coefficients of the target beta power and the actual output beta power are 0.83, 0.82, 0.71 and 0.69, respectively. As the target β power duration becomes shorter, the correlation coefficient gradually decreases, and the tracking performance decreases.
Fig. 10 shows that the PI controller tracks the beta power as a function of frequency dynamics, with (a), (c), (e) and (g) being the change in the beta power of the GPi neuron, the beta power reference value changing every 0.05Hz, 0.3Hz, 0.5Hz and 1Hz, respectively. (b) The (d), (f) and (h) are respectively the change of the corresponding closed-loop stimulation frequency. The tracking correlation coefficients are respectively 0.85, 0.65, 0.49 and 0.17, and the tracking performance is reduced as the sine frequency is larger.
As can be seen from the foregoing embodiments, in the present embodiment, a dynamic β power signal tracking method is provided, where after a β power signal output from a CAR model is compared with an expected β power signal, an obtained difference is input to a receive PI controller, and the PI controller obtains an output signal according to the received difference signal and sends the output signal to the CAR model. The CAR model outputs an output signal that approximates the desired beta power signal based on the received signal. Thus, the beta power signal can be tracked whether the expected beta power signal is changed or not.
Corresponding to the dynamic beta power signal tracking method provided by the above embodiment, the present application also provides an embodiment of a dynamic beta power signal tracking system. Referring to fig. 11, the dynamic beta power signal tracking system 20 includes: a model building module 201, an acquisition module 202 and a signal tracking module 203.
The model establishing module 201 is configured to establish a controlled autoregressive CAR model, where the CAR model is configured to receive a first output signal of a PI controller and output a first beta power signal, and the first output signal is a signal output by the PI controller at any time. The obtaining module 202 is configured to obtain a signal difference between the first β power signal and a desired β power signal. The signal tracking module 203 is configured to input the signal difference to the PI controller, the PI controller outputs a second output signal to the CAR model according to the signal difference, and the CAR model outputs a second β power signal.
Further, the model building module 201 includes: a first determination unit and a second determination unit.
A first determining unit, configured to identify a relationship between a neural network input stimulation frequency and output GPi neuron beta power, and collect CAR model input and output data from the CTx-BG-Th neural network, where the CAR model input and output are expressed as:
Figure BDA0002495372380000101
where u (k) denotes the input signal stimulus frequency, y (k) denotes the output GPi neuron β band oscillation power, naAnd nbThe order of the output input signal is (k) white noise. And the second determining unit is used for determining the model parameters of the CAR model and the order of the model parameters.
Wherein the second determination unit includes: the device comprises a parameter estimation subunit, an acquisition subunit and a determination subunit.
The parameter estimation subunit is used for estimating the CAR model parameters by using a recursive least square method. The obtaining subunit is configured to obtain an output root mean square error of the CAR model according to the real output signal and the output signal of the CAR model, where the output root mean square error is used to quantify the prediction accuracy of the CAR model. And the determining subunit is configured to determine the order of the model parameter according to the output root mean square error.
The obtaining module 202 includes: a signal input unit and a difference output unit.
The signal input unit is configured to input the first β power signal and the desired β power signal to a subtractor. And the difference output unit is used for outputting the signal difference by the subtracter.
The embodiment of the present application further provides a terminal 30, referring to fig. 12, where the terminal 30 includes: a processor 301, a memory 302, and a communication interface 303.
In fig. 12, a processor 301, a memory 302, and a communication interface 303 may be connected to each other by a bus; the bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 12, but this is not intended to represent only one bus or type of bus.
The processor 301 generally controls the overall functions of the terminal 30, such as starting the terminal 30 and establishing a controlled autoregressive CAR model after the terminal 30 is started, where the CAR model is configured to receive a first output signal of the PI controller and output a first beta power signal, and the first output signal is a signal output by the PI controller at any time; acquiring a signal difference value of the first beta power signal and an expected beta power signal; and inputting the signal difference value into the PI controller, outputting a second output signal to the CAR model by the PI controller according to the signal difference value, and outputting a second beta power signal by the CAR model.
Further, the processor 301 may be a general-purpose processor, such as a Central Processing Unit (CPU), a Network Processor (NP), or a combination of a CPU and an NP. The processor may also be a Microprocessor (MCU). The processor may also include a hardware chip. The hardware chips may be Application Specific Integrated Circuits (ASICs), Programmable Logic Devices (PLDs), or a combination thereof. The PLD may be a Complex Programmable Logic Device (CPLD), a Field Programmable Gate Array (FPGA), or the like.
The memory 302 is configured to store computer-executable instructions to support the operation of the terminal 30 data. The memory 301 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
After the terminal 30 is started, the processor 301 and the memory 302 are powered on, and the processor 301 reads and executes the computer executable instructions stored in the memory 302 to complete all or part of the steps in the above-mentioned embodiment of the dynamic beta power signal tracking method.
The communication interface 303 is used for the terminal 30 to transmit data, for example, to enable communication with a signal generation device. The communication interface 303 includes a wired communication interface, and may also include a wireless communication interface. The wired communication interface comprises a USB interface, a Micro USB interface and an Ethernet interface. The wireless communication interface may be a WLAN interface, a cellular network communication interface, a combination thereof, or the like.
In an exemplary embodiment, the terminal 30 provided by the embodiments of the present application further includes a power supply component that provides power to the various components of the terminal 30. The power components may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the terminal 30.
A communications component configured to facilitate communications between the terminal 30 and other devices in a wired or wireless manner. The terminal 30 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. The communication component receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. The communication component also includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the terminal 30 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, processors, or other electronic components.
The same and similar parts among the various embodiments in the specification of the present application may be referred to each other. Especially, for the system and terminal embodiments, since the method therein is basically similar to the method embodiments, the description is relatively simple, and the relevant points can be referred to the description in the method embodiments.
It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Of course, the above description is not limited to the above examples, and technical features that are not described in this application may be implemented by or using the prior art, and are not described herein again; the above embodiments and drawings are only for illustrating the technical solutions of the present application and not for limiting the present application, and the present application is only described in detail with reference to the preferred embodiments instead, it should be understood by those skilled in the art that changes, modifications, additions or substitutions within the spirit and scope of the present application may be made by those skilled in the art without departing from the spirit of the present application, and the scope of the claims of the present application should also be covered.

Claims (10)

1. A method for dynamic beta power signal tracking, the method comprising:
establishing a Controlled Autoregressive (CAR) model, wherein the CAR model is used for receiving a first output signal of a PI controller and outputting a first beta power signal, and the first output signal is a signal output by the PI controller at any moment;
acquiring a signal difference value of the first beta power signal and an expected beta power signal;
and inputting the signal difference value into the PI controller, outputting a second output signal to the CAR model by the PI controller according to the signal difference value, and outputting a second beta power signal by the CAR model.
2. The method of claim 1, wherein said creating a Controlled Autoregressive (CAR) model comprises:
identifying the relation between the input stimulation frequency of the neural network and the beta power of the output GPi neuron, and collecting input and output data from the CTx-BG-Th neural network, wherein the input and output are expressed as follows:
Figure FDA0002495372370000011
where u (k) denotes the input signal stimulus frequency, y (k) denotes the output GPi neuron β band oscillation power, naAnd nbIs the order of the output input signal, (k) represents white noise;
determining model parameters of the CAR model and an order of the model parameters.
3. The method of claim 2, wherein determining the model parameters of the CAR model and the order of the model parameters comprises:
estimating CAR model parameters using a recursive least squares method;
obtaining an output root mean square error of the CAR model through a real output signal and an output signal of the CAR model, wherein the output root mean square error is used for quantifying the prediction precision of the CAR model;
and determining the order of the model parameter according to the output root mean square error.
4. The method of claim 1, wherein obtaining a signal difference of the first beta power signal and a desired beta power signal comprises:
inputting the first beta power signal and the desired beta power signal to a subtractor;
the subtractor outputs the signal difference.
5. The method of claim 1, further comprising determining parameters of the PI controller to stabilize the PI controller.
6. A dynamic beta power signal tracking system, the system comprising:
the model establishing module is used for establishing a Controlled Autoregressive (CAR) model, the CAR model is used for receiving a first output signal of a PI controller and outputting a first beta power signal, and the first output signal is a signal output by the PI controller at any moment;
an obtaining module, configured to obtain a signal difference between the first β power signal and an expected β power signal;
and the signal tracking module is used for inputting the signal difference value into the PI controller, the PI controller outputs a second output signal to the CAR model according to the signal difference value, and the CAR model outputs a second beta power signal.
7. The system of claim 6, wherein the model building module comprises:
a first determining unit, configured to identify a relationship between a neural network input stimulation frequency and output GPi neuron beta power, and collect CAR model input and output data from the CTx-BG-Th neural network, where the CAR model input and output are expressed as:
Figure FDA0002495372370000021
where u (k) denotes the input signal stimulus frequency, y (k) denotes the output GPi neuron β band oscillation power, naAnd nbIs the order of the output input signal, (k) represents white noise;
and the second determining unit is used for determining the model parameters of the CAR model and the order of the model parameters.
8. The system according to claim 7, wherein the second determination unit comprises:
a parameter estimation subunit for estimating CAR model parameters using a recursive least squares method;
the acquisition subunit is used for acquiring an output root mean square error of the CAR model through a real output signal and an output signal of the CAR model, wherein the output root mean square error is used for quantifying the prediction precision of the CAR model;
and the determining subunit is used for determining the order of the model parameter according to the output root mean square error.
9. The system of claim 6, wherein the acquisition module comprises:
a signal input unit for inputting the first beta power signal and the desired beta power signal to a subtractor;
and the difference output unit is used for outputting the signal difference by the subtracter.
10. A terminal, comprising:
a processor;
a memory for storing computer executable instructions;
when the processor executes the computer-executable instructions, the processor performs the method of any of claims 1-5 to enable tracking of a dynamic beta power signal.
CN202010417020.4A 2020-05-18 2020-05-18 Dynamic beta power signal tracking method, system and terminal Pending CN111599437A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010417020.4A CN111599437A (en) 2020-05-18 2020-05-18 Dynamic beta power signal tracking method, system and terminal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010417020.4A CN111599437A (en) 2020-05-18 2020-05-18 Dynamic beta power signal tracking method, system and terminal

Publications (1)

Publication Number Publication Date
CN111599437A true CN111599437A (en) 2020-08-28

Family

ID=72187220

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010417020.4A Pending CN111599437A (en) 2020-05-18 2020-05-18 Dynamic beta power signal tracking method, system and terminal

Country Status (1)

Country Link
CN (1) CN111599437A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117293860A (en) * 2023-09-28 2023-12-26 暨南大学 Frequency safety index evaluation method, system and storage medium of power system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110444270A (en) * 2019-08-01 2019-11-12 重庆邮电大学 A kind of closed loop DBS control method based on RBF neural
CN110765720A (en) * 2019-09-12 2020-02-07 重庆大学 Power amplifier predistortion method of complex value assembly line recurrent neural network model

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110444270A (en) * 2019-08-01 2019-11-12 重庆邮电大学 A kind of closed loop DBS control method based on RBF neural
CN110765720A (en) * 2019-09-12 2020-02-07 重庆大学 Power amplifier predistortion method of complex value assembly line recurrent neural network model

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
FEI SU 等: ""Model-Based Evaluation of Closed-Loop Deep Brain Stimulation Controller to Adapt to Dynamic Changes in Reference Signal"", 《FRONTIERS IN NEUROSCIENCE》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117293860A (en) * 2023-09-28 2023-12-26 暨南大学 Frequency safety index evaluation method, system and storage medium of power system
CN117293860B (en) * 2023-09-28 2024-04-23 暨南大学 Frequency safety index evaluation method, system and storage medium of power system

Similar Documents

Publication Publication Date Title
US11862340B2 (en) Method and system for cardiovascular disease assessment and management
Bennett et al. Artificial intelligence framework for simulating clinical decision-making: A Markov decision process approach
EP3353947A1 (en) Method and system for crowd-sourced algorithm development
CN109478848A (en) Power output is controlled using battery DC characteristic
Asai et al. Learning an intermittent control strategy for postural balancing using an EMG-based human-computer interface
JP7019127B2 (en) Insulin assessment based on reinforcement learning
US20230241315A1 (en) Medication delivery using calibrated sensor measurements
CN111599437A (en) Dynamic beta power signal tracking method, system and terminal
US20210043315A1 (en) Program control system and method for implantable electronic device
CN110364268A (en) Diabetic Internet-based is health management system arranged and method
US20200391037A1 (en) Method for adaptive control of a medical device using bayesian optimization
Haddock et al. Model predictive control of deep brain stimulation for Parkinsonian tremor
EP3405922A1 (en) A patient procedure schedule throughput optimiser supervised machine learning system
CN102981461A (en) Information processing apparatus and method, server apparatus, server apparatus control method, and program
CN113724824B (en) Chronic patient follow-up method, device, computer equipment and readable storage medium
EP3189454B1 (en) Method for simulating brain stimulation, corresponding computer program and device
Bahari et al. Model-based analysis and forecast of sleep–wake regulatory dynamics: Tools and applications to data
Mathiyazhagan et al. Soft computing approach for predictive blood glucose management using a fuzzy neural network
Ayala et al. Self-Adaptation of mHealth Devices: The Case of the Smart Cane Platform
KR102339382B1 (en) Method and apparatus for regulating brain fucntion using self-adaptive brain model
KR102533835B1 (en) Method of operating scalp care platform using applicator for rental scalp care
EP3111990B1 (en) Stimulation therapy system, in particular for vagus nerve stimulation, by implementing a state transition model with prior learning
US11712564B2 (en) Systems and methods for automated deep brain stimulation programming
Basaklar et al. GEM-RL: Generalized energy management of wearable devices using reinforcement learning
EP4230248A1 (en) Nap environment decision system

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