Combined regulation and control method based on PV neurons

Abstract

The invention relates to a combined regulation and control method based on PV neurons, and the method comprises the following steps: building a thalamic cortical neuron group dynamic model, the thalamic cortical neuron group comprises a cortical main neuron, an inhibitory intermediate neuron, a specific relay nucleus and a thalamic reticulate nucleus; constructing a four-state photogenetic channel model, calibrating photogenetic channel parameters, and performing targeted photostimulation on PV neurons; and performing single-parameter scanning and parameter combination optimization on the parameters of the optical genetic channel, and realizing the parameter optimization of the optical genetic channel on the basis of the neuron activity state simulated by the thalamic cortex neuron group dynamic model. Compared with the prior art, the method has the advantages that accurate neuron regulation and control based on a targeted target can be realized, and the like.

Description

Technical Field

[0001] The present invention relates to the field of neural regulation technology, and in particular to a combined regulation method based on PV neurons. Background Art

[0002] Optogenetics, a technology that integrates optics and genetics, can precisely control the activity of specific neurons. In recent years, due to its high spatiotemporal precision and cell-specificity, optogenetics has been widely used to precisely control the activity of specific neurons. Neuromodulation is a method of regulation that is studied and applied to the nervous system and is of great significance for neuroscience research and the treatment of neurological diseases. CN116808445A provides a neuron regulation system and method based on adaptive optogenetics, comprising: an adaptive control module that adjusts controller parameters based on the error between the system's output and the underlying neural information state and physiological state; an optical path module that monitors the light beam state of neuronal activity, manipulates the light signal from the stimulation module, and guides the light signal to the sensor element; a CMOS photosensitive imaging module that captures the light signal and converts it into a digital image signal for image processing; a data perception module that records and captures the fluorescent signals released by neurons, collects and records them in real time, and extracts and analyzes information about neuronal activity; a stimulation module that regulates neuronal activity and studies neural circuits and behavioral manifestations; and a light information module that monitors the binding of photosensitive molecules or fluorescent markers to biomolecules or cells to obtain information about the biological system. This system achieves precise control of neuronal activity. However, this technology does not provide a specific implementation for parameter adjustment using adaptive control algorithms and real-time feedback mechanisms. Instead, it relies on actual data collected by external devices for adaptive control, lacking a theoretical basis for parameter optimization. Furthermore, the technology lacks precise targeting of specific neuronal populations, making it difficult to achieve precise control of different states. Summary of the Invention

[0003] The purpose of the present invention is to provide a combined regulation method based on PV neurons, using optogenetics technology to achieve precise neuronal targeted regulation.

[0004] The purpose of the present invention can be achieved by the following technical solutions:

[0005] A combined regulation method based on PV neurons, comprising the following steps:

[0006] establishing a dynamic model of a thalamocortical neuron population, wherein the thalamocortical neuron population includes cortical principal neurons, inhibitory interneurons, specific relay nuclei, and thalamic reticular nuclei;

[0007] Construct a four-state optogenetic channel model, calibrate optogenetic channel parameters, and perform targeted photostimulation of PV neurons;

[0008] Single parameter scanning and parameter combination optimization are performed on the optogenetic channel parameters, and the optogenetic channel parameter optimization is achieved based on the neuronal activity state simulated by the thalamic cortical neuron group dynamics model.

[0009] The dynamic model of the cortical principal neuron is expressed as:

[0010]

[0011] Among them, PN is the cortical principal neuron related variable, t is time, ∈ pn is the resting potential of PN, p1-p4 are the neuronal connection strength parameters, f is the linear transformation function, PV, SOM, and SRN are the related variables of PV interneurons, SOM interneurons, and specific relay nuclei, respectively, E is the electromagnetic induction stimulus, τ pn is the time constant of PN.

[0012] The inhibitory interneurons include PV interneurons, SOM interneurons and VIP interneurons.

[0013] The dynamic model of the PV interneuron is expressed as:

[0014]

[0015] The dynamic model of the SOM interneuron is expressed as:

[0016]

[0017] The dynamic model of the VIP interneuron is expressed as:

[0018]

[0019] Among them, PN, PV, SOM, SRN, and VIP are the variables related to cortical principal neurons, PV interneurons, SOM interneurons, specific relay nuclei, and VIP interneurons, respectively. t is time, ∈ pv ,∈ som ,∈ vip represent the resting potential of the corresponding interneuron, is the control potential of PV interneurons, τ pv , τ som , τ vip They represent the time constants of the corresponding intermediate neurons, p represents the corresponding neuron connection strength parameter, and f is the linear transformation function.

[0020] The dynamic model of the specific relay core is expressed as:

[0021]

[0022] Among them, PN, SRN, and TRN are the variables related to the cortical principal neurons, specific relay nuclei, and thalamic reticular nuclei, respectively. t is time, ∈ srn is the resting potential of the specific relay nucleus, p represents the corresponding neuronal connection strength parameter, f is the linear transformation function, τ srn is the time constant of the specific relay core, and z is the activation function.

[0023] The dynamic model of the thalamic reticular nucleus is expressed as:

[0024]

[0025] Among them, PN, SRN, and TRN are the variables related to the cortical principal neurons, specific relay nuclei, and thalamic reticular nuclei, respectively. t is time, ∈ trn is the resting potential of the thalamic reticular nucleus, p represents the corresponding neuronal connection strength parameter, f is the linear transformation function, τ trn is the time constant of the thalamic reticular nucleus, and z is the activation function.

[0026] The four-state optogenetic channel model is expressed as:

[0027]

[0028] O1+O2+L1+L2=1

[0029] Among them, O1 and O2 represent the open state, L1 and L2 represent the closed state, and H a1 、H a2 , K d1 , K d2 Represent the transition rates from L1 to O1, L2 to O2, O1 to L1, and O2 to L2, respectively, e 12 、e 21 Represent the transition rates from O1 to O2 and from O2 to O1, K r is the transition rate from L2 to L1.

[0030] The targeted light stimulation of PV neurons is specifically performed by calculating the photocurrent based on the state determined by the four-state optogenetic channel model:

[0031]

[0032] S x =ε x x-μ x ,x=PV,SOM,VIP

[0033] in, is the photocurrent corresponding to x, b ChR2 is the maximum conductivity of ChR2, EChR2 is the reversal potential of ChR2, γ is the conductivity ratio, ε x is the slope of the linear transformation corresponding to x, μ x is the intercept of the linear transformation corresponding to x;

[0034] Calculation of the control potential for targeted photostimulation of PV neurons based on photocurrent:

[0035]

[0036] Among them, N ChR2 is the number of ChR2 channels, R m is the average resistance of the stimulation group.

[0037] The optogenetic channel parameters to be optimized include the range of light intensity, pulse width, and stimulation frequency.

[0038] The objective function of the optogenetic channel parameter combination optimization is:

[0039] F=w1E1+w2E2+w3E3

[0040] Among them, F is the objective function, w1, w2, and w3 are weights, E1 represents the inhibitory effect index on the regulation target, E2 represents the energy consumption index, and E3 represents the side effect index.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] (1) By establishing a dynamic model of thalamic cortical neuron groups, the present invention can precisely target and regulate specific neuron groups such as cortical principal neurons, inhibitory interneurons, specific relay nuclei and thalamic reticular nuclei. Subsequently, by constructing a four-state optogenetic channel model for targeted photostimulation of PV neurons, it can act more accurately on the target neurons and achieve precise regulation of specific nerves.

[0043] (2) The present invention optimizes the optogenetic channel parameters by performing single parameter scanning and parameter combination optimization on the parameters of the optogenetic channel, and simulating the neuronal activity state based on the dynamic model of the thalamic cortical neuron group. It does not rely on the actual data collected by external equipment, but determines the optimal parameters through the constructed model and the optimization method based on the objective function, thereby reducing the interference of external factors and improving the autonomy and stability of the regulation.

[0044] (3) The objective function of the parameter optimization method of the present invention takes into account the inhibition effect, energy consumption and side effects, which can achieve effective control target inhibition, reduce energy consumption and side effects caused by control, reduce control costs, and improve practical applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flow chart of the method of the present invention;

[0046] Figure 2 This is a schematic diagram of the thalamocortical neural network structure;

[0047] Figure 3 Schematic diagram of the extreme value changes of the PV neuron group under different parameter conditions in one embodiment. DETAILED DESCRIPTION

[0048] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0049] Example 1

[0050] This embodiment provides a joint regulation method based on PV neurons, such as Figure 1 As shown, the following steps are included:

[0051] S1. Establish a dynamic model of the thalamocortical neuron population, which includes cortical principal neurons (PN), three types of inhibitory interneurons (PV, SOM, VIP), specific relay nuclei (SRN) and thalamic reticular nucleus (TRN).

[0052] Among them, the dynamic model of the cortical principal neuron is expressed as:

[0053]

[0054] The dynamic model of PV interneurons is expressed as:

[0055]

[0056] The dynamic model of SOM interneurons is expressed as:

[0057]

[0058] The dynamic model of VIP interneurons is expressed as:

[0059]

[0060] The dynamic model of the specific relay nucleus is expressed as:

[0061]

[0062] The dynamic model of the thalamic reticular nucleus is expressed as:

[0063]

[0064] Among them, PN, PV, SOM, VIP, SRN, and TRN are the relevant variables of cortical principal neurons, PV interneurons, SOM interneurons, VIP interneurons, specific relay nuclei, and thalamic reticular nuclei, respectively. t is time, ∈ is the resting potential of the neuron represented by its subscript, f is the transformation function, E is the electromagnetic induction stimulus, and τ is the time constant of the neuron represented by its subscript. is the control potential of the PV interneuron, z is the activation function, and p represents the corresponding neuron connection strength parameter.

[0065] The connection relationship of each neuron time represented by p is as follows Figure 2 Among them, p1~p4, p6~p8 are the connection strength parameters of the intracortical connection, and their specific meanings and values are as follows:

[0066] PN→PN: p1=1.8 (excitability)

[0067] PV→PN: p2 = 0.1-5.0 (inhibitory, step size 0.1)

[0068] SOM→PN: p3=1.38 / 1.45 (inhibitory)

[0069] PN→PV: p4=4.5 (excitability)

[0070] SOM→PV: p6=0.5 / 2.0 (inhibitory)

[0071] PN→SOM: p7=4.0 (excitability)

[0072] PV→SOM: p8=0.2 (inhibitory)

[0073] p9~p 11 、p 15 is the connection strength parameter of the thalamocortical connection. Its specific meaning and value are as follows:

[0074] SRN→PN: p9=1.0 (excitability)

[0075] TRN→SRN:p 10 =0.5 (inhibitory)

[0076] PN→SRN:p 11 =3.5 (excitability)

[0077] SRN→PV:p 15 =0.5-4.0 (excitability, step size 0.1)

[0078] In this embodiment, the neuron group time constant is as follows:

[0079] τpn =0.026ms-1 (cortical principal neuron)

[0080] τ pv =0.065ms-1 (PV interneuron)

[0081] τ som =0.0325ms-1 (SOM interneuron)

[0082] τ vip =0.0325ms-1 (VIP interneuron)

[0083] τ srn =0.0026ms-1(specific relay core)

[0084] τ trn =0.0026ms-1 (thalamic reticular nucleus)

[0085] The neuron activation function parameters are:

[0086] Sigmoid function slope v = 2.5 × 105

[0087] Constant term β = 0.5

[0088] Gain constant α = 2.8

[0089] The resting potential of a neuron group is as follows:

[0090] PN:∈ pn =-0.35

[0091] PV:∈ pv =-3.5

[0092] SOM:∈ som =-3.4

[0093] VIP:∈ vip =3.3

[0094] SRN:∈ srn =-2.0

[0095] TRN:∈ trn =-5.0

[0096] S2, construct a four-state optogenetic channel model, calibrate the optogenetic channel parameters, and perform targeted photostimulation on PV neurons.

[0097] The four-state optogenetic channel model is expressed as:

[0098]

[0099]

[0100] O1+O2+L1+L2=1

[0101] Among them, O1 and O2 represent the open state, L1 and L2 represent the closed state, and H a1 、H a2 , K d1 , K d2 Represent the transition rates from L1 to O1, L2 to O2, O1 to L1, and O2 to L2, respectively, e 12 、e 21 Represent the transition rates from O1 to O2 and from O2 to O1, K r is the transition rate from L2 to L1.

[0102] H a1 =η1J*P(ms -1 )

[0103] H a2 =η2J*P(ms -1 )

[0104] K d1 =0.075+0.043*tanh(-(S x +20) / 20)(ms -1 )

[0105] K d2 =0.05(ms -1 )

[0106] e 12 =0.011+0.005ln(1+E x / 0.024)(ms -1 )

[0107] e 21 =0.008+0.004ln(1+E x / 0.024)(ms -1 )

[0108] K r =4.34587*10 -5 *exp(-0.0211539274*S x )(ms -1 )

[0109]

[0110] Where Q0(θ) is the ChR2 activation function, Q0(θ)=0.5*(1+tanh(120(100E x -0.1))), τ ChR2is the activation time constant of ChR2, which is 1.3 ms in this embodiment, η1 is the quantum efficiency of photon absorption by L1, which is 0.8535 in this embodiment, η2 is the quantum efficiency of photon absorption by L2, which is 0.14 in this embodiment, J is the photon flux, J = σ ret *E x *λ / (w loss *h*c)(ms -1 ), σ ret is the retinal absorption cross section, which is 10 in this embodiment. -20 m 2 , E x is the illumination intensity of the target group, λ is the optical wavelength, which is 470nm in this embodiment, and w loss is the photon loss scaling factor, which is 1.3 in this embodiment, and h is Planck's constant, h = 6.626*10 -34 Js, c is the speed of light, c = 3*10 8 m / s.

[0111] Targeted photostimulation of PV neurons is performed as follows: Based on the states determined by the four-state optogenetic channel model, the photocurrent is calculated:

[0112]

[0113] S x =ε x x-μ x ,x=PV,SOM,VIP

[0114] in, is the photocurrent corresponding to x, b ChR2 is the maximum conductivity of ChR2, which is 400nS in this embodiment, and E ChR2 is the reversal potential of ChR2, which is 0 mV in this embodiment, γ is the conductivity ratio, which is 0.1 in this embodiment, and ε x is the slope of the linear transformation corresponding to x, which is 60mV in this embodiment, μ x is the intercept of the linear transformation corresponding to x, which is 20 mV in this embodiment;

[0115] Calculation of the control potential for targeted photostimulation of PV neurons based on photocurrent:

[0116]

[0117] Among them, N ChR2 is the number of ChR2 channels, which is 2.5*10 in this embodiment. 2 , R m is the average resistance of the stimulation group, which is 3.2 MΩ in this embodiment.

[0118] S3, single parameter scanning and parameter combination optimization are performed on the optogenetic channel parameters, and the optogenetic channel parameter optimization is achieved based on the neuronal activity state simulated by the thalamic cortical neuron group dynamics model.

[0119] In this embodiment, the optogenetic channel parameters to be optimized include the range of light intensity, pulse width, and stimulation frequency.

[0120] This example first verifies the causal relationship between insufficient feedforward inhibition of PV neurons and the regulatory target through theoretical models and numerical simulations.

[0121] For comparison, this embodiment sets two sets of initial conditions:

[0122] Group A (weak inhibition): p3 = 1.38, p6 = 0.5;

[0123] Group B (strong inhibition): p3=1.45, p6=2.0.

[0124] For each set of conditions, by adjusting p 15 and p2 perform a dual parameter sweep: where p 15 The step size is set to 0.1, the p2 step size is set to 0.1; the membrane potential time series of the PN group is recorded

[0125] Then the system dynamic characteristics are analyzed, including the following steps: extracting extreme value data; calculating the dominant frequency; identifying the bifurcation point; and drawing the phase diagram.

[0126] The results can be obtained from the phase diagram analysis:

[0127] (1) When p 15 When <1.5, the system exhibits a typical pathological state: 2-4Hz spike and slow wave discharges; the amplitude is significantly increased; and the degree of synchronization is improved.

[0128] (2) Dynamic conversion law: p 15 The increase leads to Hopf bifurcation; the periodic solution coexists with the low saturation state; the pathological waveform disappears after the LPC bifurcation.

[0129] In one embodiment, the PV extreme values under different parameter conditions are as follows: Figure 3 shown.

[0130] Based on the above analysis, this embodiment verifies the causal relationship between certain parameters and the pathological state presented by the control target through theoretical models and numerical simulations, that is, when certain connection strength parameters of PV neurons (such as p 15 , p2, etc.) are in a specific range, the system will experience corresponding pathological conditions, so determining PV neurons is the key target.

[0131] On this basis, this embodiment performs optogenetic channel parameter optimization, including:

[0132] (1) Single parameter scan:

[0133] Light intensity: 0.1-10mW / mm 2 , step size 0.1;

[0134] Frequency: 10-40Hz, step size 1Hz;

[0135] Pulse width: 1-10ms, step size 0.5ms.

[0136] (2) Parameter combination optimization:

[0137] The objective function for optimizing the optogenetic channel parameter combination is:

[0138] F=w1E1+w2E2+w3E3

[0139] Among them, F is the objective function, w1, w2, and w3 are weights, E1 represents the inhibitory effect index on the regulation target, E2 represents the energy consumption index, and E3 represents the side effect index.

[0140] The genetic algorithm is used to solve the optimal combination. The process of solving the optimal combination by the genetic algorithm is a common method in this field. In order to avoid ambiguity of the purpose of the present invention, this embodiment does not describe the process of solving the optimal combination by the genetic algorithm in detail.

[0141] This embodiment uses the following three quantitative indicators to evaluate the control effect: SWDs frequency reduction rate, energy utilization efficiency, and tissue temperature rise limit. The optimal parameter range finally obtained is:

[0142] Light intensity: 2-5mW / mm 2 ;

[0143] Stimulation frequency: 20-30 Hz;

[0144] Pulse width: 2-5ms.

[0145] The following is a detailed analysis of the parameter optimization results.

[0146] (1) ChR2 channel kinetic characteristics

[0147] Activation time constant: At the optimal light intensity (3.5mW / mm 2 ) conditions, the channel opening time constant was 0.85±0.12ms, which was significantly lower than that of ordinary ChR2 (1.21±0.15ms);

[0148] Inactivation kinetics: Biexponential fitting curve shows τ fast =9.8±1.2ms, τ slow=38.4±4.5ms

[0149] (2) Single parameter sensitivity analysis

[0150] Light intensity response curve: 0.1-10mW / mm 2 Within the range, the suppression effect shows an S-shaped curve, 2mW / mm 2 The following effects are not obvious, 5mW / mm 2 The above is saturated.

[0151] Frequency dependence: The inhibitory effect is best in the frequency range of 20-30Hz, and the energy consumption is moderate; the effect is significantly reduced below 15Hz, and rebound excitation may be induced above 35Hz.

[0152] Pulse width influence: The suppression efficiency is highest in the 2-5ms pulse width range; <2ms activation is insufficient, >6ms energy is wasted and the thermal effect is enhanced.

[0153] (3) Multi-parameter combination optimization results

[0154] Parameter interaction effect: It was found that there was an obvious interaction between light intensity and pulse width, and the interaction between frequency and the other two parameters was relatively independent.

[0155] Optimization algorithm convergence: The genetic algorithm reached a stable solution in the 47th generation, and the optimal parameter combination was a light intensity of 3.5mW / mm 2 , frequency 25Hz, pulse width 3ms.

[0156] Weight sensitivity: The most acceptable solution is obtained when the weight combination (w1, w2, w3) = (0.6, 0.3, 0.1).

[0157] (4) Verification of long-term control effect

[0158] Durability of efficacy: After continuous application of the optimal parameters for 28 days in the WAG / Rij rat model, the SWDs inhibition efficiency remained at 78.5±5.2% without significant attenuation.

[0159] Energy efficiency: Compared with traditional parameters, the optimized solution reduces energy consumption by 42.6% and controls the fiber temperature rise to <0.8°C.

[0160] Spatial distribution: Through light intensity distribution simulation and actual measurement comparison, it was confirmed that the effective stimulation range covers approximately 92% of the volume of the target area.

[0161] (5) Safety assessment

[0162] Histological examination: After 30 days of continuous stimulation, the neuronal survival rate in the target area was >95%, and there was no obvious glial cell proliferation.

[0163] Off-target effect assessment: Under optimized parameters, the probability of passive activation of adjacent brain regions was <4.5%, significantly lower than that under conventional parameters (15.3%).

[0164] Thermal effect control: The maximum local temperature rise is 0.76±0.09℃, which is far below the tissue damage threshold (2℃).

[0165] Example 2

[0166] An important application area for neuronal regulation is the treatment of neurological diseases. Absence epilepsy is a generalized form of epilepsy, typically characterized by 2.5-4 Hz spike-wave discharges (SWDs). Current research suggests that this type of epilepsy is associated with abnormal interactions between the cortex and thalamus. This example illustrates the effectiveness of the method of the present invention, using the goal of achieving epilepsy suppression as an example.

[0167] 1. State recognition and classification

[0168] (1) Define feature states

[0169] Low saturation health state (I)

[0170] Simple spike and slow waves (II)

[0171] Multiple spike-wave (III)

[0172] Status clonic (IV)

[0173] High saturation state (V)

[0174] (2) State feature extraction

[0175] Time domain characteristics: extreme value, mean, variance

[0176] Frequency domain characteristics: main frequency, energy distribution

[0177] Nonlinear characteristics: Lyapunov exponent

[0178] 2. State transition control strategy

[0179] (1) Determine the target conversion path

[0180] Pathological state → transitional state → healthy state

[0181] Minimize energy consumption

[0182] Avoid unstable oscillations

[0183] (2) Real-time control algorithm

[0184] While (not reaching the target state) {

[0185] Get the current state S(t)

[0186] Calculate target deviation e(t)

[0187] Update control parameters u(t)

[0188] Implementing stimulus packages

[0189] Evaluate control effectiveness

[0190] }

[0191] 3. Verify the results

[0192] (1) State conversion efficiency

[0193] Conversion time 0.5-2s;

[0194] Success rate >85%;

[0195] Energy efficiency >70%.

[0196] (2) Stability analysis

[0197] Disturbance response characteristics

[0198] Parameter sensitivity

[0199] Long-term stability

[0200] 4. Detailed dynamic characteristics analysis

[0201] (1) State space feature mapping

[0202] The 17 extracted characteristic parameters were reduced to a three-dimensional space through principal component analysis (PCA). The state clusters were clearly separable. The coordinates of the characteristic center points of state I were (0.21, 0.15, 0.08), and the coverage radius r1 was 0.12±0.02. The characteristic center points of states II-V were located at (0.45, 0.38, 0.22), (0.68, 0.59, 0.41), (0.82, 0.74, 0.63), and (0.93, 0.85, 0.79), respectively. There were obvious phase transition characteristics between the states, the transition boundaries were clearly discernible, and the critical point parameter fluctuation was <5%.

[0203] (2) State transition dynamics

[0204] The direct conversion from pathological state III to healthy state I takes an average of 2.8±0.4s and consumes 31.7±2.9mJ of energy. By introducing transition state II as an intermediate state, the total conversion time can be shortened to 1.4±0.2s and the energy consumption is reduced to 19.3±1.8mJ. Compared with the direct conversion (IV→I), the multi-step conversion path from state IV to state I (IV→III→II→I) has a 22.7% higher success rate and a 34.2% higher energy efficiency. The optimal state transition path network was determined, achieving 95.6% state accessibility coverage.

[0205] (3) Quantitative indicators of state stability

[0206] The Lyapunov exponent of healthy state I is λ1=-0.023±0.004, showing a strong attractor characteristic; the Lyapunov exponents of pathological states III-V are λ3=0.038±0.006, λ4=0.057±0.009 and λ5=0.082±0.011, respectively, all showing unstable divergent characteristics; the Lyapunov exponent of transition state II is λ2=0.008±0.003, which is close to the critical state and is suitable as a control intermediate state.

[0207] Example 3

[0208] The above is an introduction to the method embodiment. The following further illustrates the solution of the present invention through a system embodiment.

[0209] A PV neuron-based joint regulation system includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the method described in Example 1 when executing the program.

[0210] In one embodiment, the system includes:

[0211] (1) Model building module: responsible for establishing the dynamic model of the thalamocortical neuron group. Based on neuroscience theory and relevant research data, the dynamic equations of neurons such as cortical principal neurons, inhibitory interneurons (including PV neurons, etc.), specific relay nuclei and thalamic reticular nuclei are determined to describe their electrical activity, signal transmission and other characteristics, as well as the relationship between the connections and interactions between neurons.

[0212] (2) Optogenetic regulation module, including:

[0213] (21) Channel model construction unit: Construct a four-state optogenetic channel model. Analyze the channel opening and closing characteristics of light-sensitive proteins under different lighting conditions, and determine the form and range of model parameters based on the physiological characteristics of PV neurons.

[0214] (22) Parameter calibration unit: Calibrate the parameters of the optogenetic channel through experimental measurements or reference to existing research results. For example, the response time of the light-sensitive protein to a specific wavelength of light, channel conductance and other parameters are measured to ensure that the model can accurately reflect the actual behavior of the optogenetic channel.

[0215] (23) Targeted stimulation unit: Based on the constructed model and calibration parameters, targeted light stimulation is performed on PV neurons. Optical fibers and other equipment can be used to accurately transmit light signals of specific wavelength, intensity, and frequency to the area where PV neurons are located, achieving precise control of their activity.

[0216] (3) Parameter optimization module, including:

[0217] (31) Single parameter scanning unit: Scans the parameters of the optogenetic channel one by one. Within the set parameter value range, the value of a single parameter is changed. Based on the dynamic model of the thalamocortical neuron group, the changes in the neuronal activity state are simulated, the simulation results under different parameter values are recorded, and the influence of a single parameter on neuronal activity is analyzed.

[0218] (32) Parameter combination optimization unit: Optimize the optogenetic channel parameter combination using optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms, etc.). Guided by the desired target of the neuronal activity state (such as a specific discharge frequency, synchronization, etc.), the optimal parameter combination is searched in the parameter space to achieve the best control effect of the neuronal activity state simulated based on the thalamocortical neuron group dynamics model.

[0219] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0220] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A joint regulation method based on PV neurons, characterized in that: The following steps are involved: establishing a dynamic model of a thalamocortical neuron population, wherein the thalamocortical neuron population includes cortical principal neurons, inhibitory interneurons, specific relay nuclei, and thalamic reticular nuclei; Construct a four-state optogenetic channel model, calibrate optogenetic channel parameters, and perform targeted photostimulation of PV neurons; Single parameter scanning and parameter combination optimization are performed on the optogenetic channel parameters, and the optogenetic channel parameter optimization is achieved based on the neuronal activity state simulated by the thalamic cortical neuron group dynamics model.

2. A combined regulation method based on PV neurons according to claim 1, characterized in that: The dynamic model of the cortical principal neuron is expressed as: Among them, PN is the cortical principal neuron related variable, t is time, ∈ pn is the resting potential of PN, p1-p4 are the neuronal connection strength parameters, f is the linear transformation function, PV, SOM, and SRN are the related variables of PV interneurons, SOM interneurons, and specific relay nuclei, respectively, E is the electromagnetic induction stimulus, τ pn is the time constant of PN.

3. The combined regulation method based on PV neurons according to claim 1, characterized in that: The inhibitory interneurons include PV interneurons, SOM interneurons and VIP interneurons.

4. A combined regulation method based on PV neurons according to claim 3, characterized in that: The dynamic model of the PV interneuron is expressed as: The dynamic model of the SOM interneuron is expressed as: The dynamic model of the VIP interneuron is expressed as: Among them, PN, PV, SOM, SRN, and VIP are the variables related to cortical principal neurons, PV interneurons, SOM interneurons, specific relay nuclei, and VIP interneurons, respectively. t is time, ∈ pv ,∈ som ,∈ vip represent the resting potential of the corresponding interneuron, is the control potential of PV interneurons, τ pv , τ som , τ vip They represent the time constants of the corresponding intermediate neurons, p represents the corresponding neuron connection strength parameter, and f is the linear transformation function.

5. The combined regulation method based on PV neurons according to claim 1, characterized in that: The dynamic model of the specific relay core is expressed as: Among them, PN, SRN, and TRN are the variables related to the cortical principal neurons, specific relay nuclei, and thalamic reticular nuclei, respectively. t is time, ∈ srn is the resting potential of the specific relay nucleus, p represents the corresponding neuronal connection strength parameter, f is the linear transformation function, τ srn is the time constant of the specific relay core, and z is the activation function.

6. The combined regulation method based on PV neurons according to claim 1, characterized in that: The dynamic model of the thalamic reticular nucleus is expressed as: Among them, PN, SRN, and TRN are the variables related to the cortical principal neurons, specific relay nuclei, and thalamic reticular nuclei, respectively. t is time, ∈ trn is the resting potential of the thalamic reticular nucleus, p represents the corresponding neuronal connection strength parameter, f is the linear transformation function, τ trn is the time constant of the thalamic reticular nucleus, and z is the activation function.

7. The combined regulation method based on PV neurons according to claim 1, characterized in that: The four-state optogenetic channel model is expressed as: O1+O2+L1+L2=1 Among them, O1 and O2 represent the open state, L1 and L2 represent the closed state, and H a1 、H a2 , K d1 , K d2 Represent the transition rates from L1 to O1, L2 to O2, O1 to L1, and O2 to L2, respectively, e 12 、e 21 Represent the transition rates from O1 to O2 and from O2 to O1, K r is the transition rate from L2 to L1.

8. The combined regulation method based on PV neurons according to claim 7, characterized in that: The targeted light stimulation of PV neurons is specifically performed by calculating the photocurrent based on the state determined by the four-state optogenetic channel model: S x =ε x x-μ x ,x=PV,SOM,VIP in, is the photocurrent corresponding to x, b ChR2 is the maximum conductivity of ChR2, E ChR2 is the reversal potential of ChR2, γ is the conductivity ratio, ε x is the slope of the linear transformation corresponding to x, μ x is the intercept of the linear transformation corresponding to x; Calculation of the control potential for targeted photostimulation of PV neurons based on photocurrent: Among them, N ChR2 is the number of ChR2 channels, R m is the average resistance of the stimulation group.

9. The combined regulation method based on PV neurons according to claim 1, characterized in that: The optogenetic channel parameters to be optimized include the range of light intensity, pulse width, and stimulation frequency.

10. The combined regulation method based on PV neurons according to claim 1, characterized in that: The objective function of the optogenetic channel parameter combination optimization is: F=w1E1+w2E2+w3E3 Among them, F is the objective function, w1, w2, and w3 are weights, E1 represents the inhibitory effect index on the regulation target, E2 represents the energy consumption index, and E3 represents the side effect index.

Images