Ultrasonic Stimulation Desynchronization Simulation Method and System Device Based on Neural Network Model

Abstract

The present invention relates to an ultrasonic stimulation desynchronization simulation method and system device based on a neural network model. It includes: obtaining ultrasonic data, where the ultrasonic data includes sound pressure and / or frequency; inputting the ultrasonic data into an intravascular cavitation model constructed based on the intravascular cavitation effect to obtain an intravascular capacitance displacement current; then inputting the intravascular capacitance displacement current into the neural network model to obtain the potentials of each neuron; calculating a network synchronization error based on the potentials of each neuron; and then quantifying the desynchronization of the neural network model from the network synchronization error. The method of the present invention aims to achieve desynchronization based on the ultrasonic stimulation simulation of the neural network model, and explore its inhibitory ability and potential application value for pathological seizures.

Description

Technical Field

[0001] The present invention relates to the field of neuroscience, and more specifically, to a method, device, system, computer-readable storage medium, and application thereof for simulating ultrasonic stimulation desynchronization based on a neural network model. Background Art

[0002] Neurons are the fundamental structural and functional units of the nervous system, and signal transmission in the nervous system relies on clusters of neurons. Biological information in the nervous system is transmitted from one neuron to the next. Neurons are coupled through common chemical and electrical coupling methods. Electrical coupling transmits action potentials from one neuron to the next faster than chemical coupling, so it is generally believed that electrical coupling enables synchronized movement between neurons. Synchronous discharges in coupled neural networks or neural clusters can cause neuropsychiatric disorders such as epilepsy and Parkinson's disease. Recent studies have shown that using external physical techniques to stimulate nerves (e.g., applied current stimulation and magnetic acoustic stimulation) can alter the firing rhythm of neurons.

[0003] In recent years, high-frequency, high-intensity ultrasound has been widely used in medicine, such as ultrasound imaging, ultrasonic lithotripsy, and ultrasonic scalpels. Low-intensity transcranial ultrasound stimulation, a novel neuromodulatory technique, has attracted considerable attention due to its noninvasive nature, high penetration depth, and high spatial resolution. Research has focused on different ultrasound parameters (frequency, peak pressure amplitude, duty cycle, duration, and pulse repetition frequency) and their therapeutic and regulatory effects on psychiatric disorders. Numerous animal models and human experiments have demonstrated that ultrasound modulation of brain tissue can induce in situ neural activity. Low-intensity transcranial ultrasound stimulation of the motor cortex of healthy mice can induce motor responses in the tail, limbs, and whiskers, alter local field potentials, and enhance cerebral blood flow. Low-intensity transcranial ultrasound neuromodulation can also suppress the onset of psychiatric disorders such as epilepsy and Parkinson's disease. In the nervous system of an organism, numerous neurons and neuronal groups form a complex network structure. The coupling between neurons is not a one-to-one relationship, allowing for the construction of complex neural networks using a small-world network approach. Summary of the Invention

[0004] The method of the present invention is based on the theoretical basis of the ultrasonic intra-membrane cavitation effect, and combines the intra-membrane cavitation model and the neural network model. It is a theoretical innovation in the theoretical explanation and mechanism of the ultrasonic intra-membrane cavitation effect in the desynchronization of neuronal and neural network discharges. It provides a theoretical research method for finding the optimal parameters of different ultrasonic stimulations for the treatment of mental illnesses, and explores the application value of ultrasonic stimulation in the clinical treatment of functional encephalopathy.

[0005] The present application discloses a method for simulating ultrasonic stimulation desynchronization based on a neural network model, comprising:

[0006] Obtain ultrasonic data, where the ultrasonic data includes sound pressure and / or frequency;

[0007] Input the ultrasonic data into an intracavitary cavitation model constructed based on the intracavitary cavitation effect to obtain an intracavitary capacitance displacement current;

[0008] Input the intracavitary capacitance displacement current into a neural network model to obtain the potentials of each neuron;

[0009] Calculate the network synchronization error of the neural network model based on the potentials of each neuron;

[0010] Obtain the desynchronization of the neural network model based on the network synchronization error of the neural network model.

[0011] Further, the step of inputting the ultrasonic data into an intracavitary cavitation model constructed based on the intracavitary cavitation effect to obtain an intracavitary capacitance displacement current is specifically as follows: Input the ultrasonic data into the BLS. The BLS drives the bubble to generate a dynamically deformed bubble radius Z(t) based on the ultrasonic sound pressure. When Z(t) reaches a stable periodic solution, substitute it into the capacitance Cm(Z) in the form of a Fourier series, thereby changing the average capacitance Cm of the membrane and generating an intracavitary capacitance displacement current I Cm ; Preferably, the equation of the BLS is as follows:

[0012]

[0013]

[0014]

[0015]

[0016]

[0017] where P ec is the effective pressure exerted on the membrane by the charges on both sides of the membrane, P in is the gas pressure inside the bubble, P0, P v , P A ·sin(ωt) are the hydrostatic pressure in the external medium of the bubble, the saturated gas pressure inside the bubble, and the external driving ultrasonic sound pressure respectively, Vm is the membrane potential, t is time, f is the frequency of the ultrasonic wave, δ is the surface tension coefficient, μ is the viscosity coefficient, ɑ is the radius of the bubble boundary, Z0 is the initial radius of the bubble boundary, Δ is the initial gap between the two membranes, c is the speed of sound in the liquid medium, ε r is the relative dielectric constant of the intracavitary cavity of the membrane, ε0 is the dielectric constant of the intracavitary cavity of the membrane, C m0 is the cell membrane capacitance under the initial conditions, ρl is the tissue density of the conductive medium.

[0018] Furthermore, the neural network model is constructed in the way of a small-world neural network. The small-world neural network uses N neurons of finite natural numbers as network nodes. Each node is symmetrically connected to its 2m (m << N) nearest adjacent nodes, and then two nodes are randomly selected from the N nodes with probability p for edge addition processing to obtain a small-world neural network with a connection degree of 2m / N without isolated nodes. The construction process of the small-world neural network is shown by the following differential equations, that is:

[0019]

[0020] Among them, x represents the membrane potential of the neuron, y represents the recovery variable related to the internal current, z represents the slow-varying regulatory current related to the potassium ion current activated by calcium ions, C represents the coupling strength, N represents the number of neurons, I is the external DC excitation, and I Cm is the membrane capacitance displacement current. a, b, c, d, r, s, χ are common parameters for model construction, and matrix(i,j) is the connection matrix: when neuron Ni and neuron Nj are mutually coupled, matrix(i,j) = 1, otherwise matrix(i,j) = 0.

[0021] Specifically, the neurons are obtained by coupling double-neuron models. Optionally, the double-neuron model includes one or more of the following models: Hindmarsh-Rose, FitzHugh-Nagumo, Wilson-Cowan, Hodgkin-Huxey, McCullochandPitts, Mapmodels, Morris-Lecar, PhaseOscillatormodels; Preferably, the double-neuron model is Hindmarsh-Rose.

[0022] Furthermore, the network synchronization error includes the network synchronization errors of all neurons in the network and is obtained from the consistency difference of the discharge activities of the potentials of each neuron; Preferably, the calculation process is as follows:

[0023]

[0024] Among them, the network synchronization error e i represents the difference between the potential of the i-th neuron and the average potential of the neuron network. x, y, z represent the potentials of each neuron of the i-th neuron.

[0025] Furthermore, the desynchronization of the neural network model is obtained through the average network synchronization error, and the average network synchronization error is obtained by calculating the arithmetic mean of the network synchronization errors of all neurons in the network.

[0026] An ultrasonic stimulation desynchronization simulation system based on a neural network model, characterized in that the system comprises:

[0027] An acquisition module that acquires ultrasonic data, where the ultrasonic data includes sound pressure and / or frequency;

[0028] A first generation module that inputs the ultrasonic data into an intracavitary cavitation model constructed based on the intracavitary cavitation effect to obtain an intracavitary capacitive displacement current;

[0029] A second generation module that inputs the intracavitary capacitive displacement current into a neural network model to obtain the potentials of each neuron;

[0030] A calculation module that calculates the network synchronization error of the neural network model based on the potentials of each neuron;

[0031] An output module that obtains the desynchronization of the neural network model based on the network synchronization error of the neural network model.

[0032] An ultrasonic stimulation desynchronization simulation device based on a neural network model, the device comprising:

[0033] A memory and a processor;

[0034] The memory is used to store program instructions;

[0035] The processor is used to call the program instructions, and when the program instructions are executed, it is used to execute the above-mentioned ultrasonic stimulation desynchronization simulation method based on the neural network model.

[0036] A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the above-mentioned ultrasonic stimulation desynchronization simulation method based on the neural network model.

[0037] The above device is of great significance in assisting the research on the desynchronization problem of ultrasound on excitable neurons; the research includes exploring and analyzing the desynchronization regulation effect of low-intensity transcranial ultrasound stimulation from the microscopic cell level to the macroscopic mental disease levels such as epilepsy and Parkinson's disease using a neural network model;

[0038] The application of the above device in analyzing the synchronous state or asynchronous state of neurons;

[0039] Use of the above-mentioned device in suppressing seizures of mental diseases such as epilepsy and Parkinson's; optionally, the use includes exploring the optimal parameters for transcranial ultrasound stimulation in treating seizures of mental diseases such as epilepsy and Parkinson's;

[0040] The above-mentioned device provides a new approach for theoretical exploration of the use of transcranial ultrasound stimulation in the treatment of other mental diseases, promoting the application of ultrasound stimulation in the clinical treatment of functional encephalopathy;

[0041] The above-mentioned system is of great significance in assisting the study of the desynchronization problem of excitable neurons by ultrasound; the study includes using a mathematical model to explore and analyze the desynchronization regulation effect of low-intensity transcranial ultrasound stimulation from the microscopic cell level to the macroscopic level of mental diseases such as epilepsy and Parkinson's;

[0042] Use of the above-mentioned system in analyzing the synchronous or asynchronous state of neurons;

[0043] Use of the above-mentioned system in suppressing seizures of epilepsy and Parkinson's; optionally, the use includes exploring the optimal parameters for transcranial ultrasound stimulation in treating seizures of mental diseases such as epilepsy and Parkinson's;

[0044] The above-mentioned system provides a new approach for theoretical exploration of the use of transcranial ultrasound stimulation in the treatment of other mental diseases, promoting the application of ultrasound stimulation in the clinical treatment of functional encephalopathy.

[0045] The present invention conducts joint training on high-quality ultrasound data with clinical significance using a cavitation effect model and a neural network model, which has an important role and significance in solving the desynchronization problem of excitable neurons. Based on the theoretical basis of ultrasonic intracavitary cavitation effect, from the microscopic cell level to the macroscopic level of mental diseases such as epilepsy and Parkinson's, a neural network model is used to explore and analyze the desynchronization regulation effect of low-intensity transcranial ultrasound stimulation, which not only provides a biophysical mechanism for ultrasound regulation, but also provides a theoretical method for finding the optimal ultrasound parameters, further providing theoretical guidance for the treatment of mental diseases, being innovative in the field of life sciences, and having a beneficial promoting effect on the synchronous analysis and state regulation research of ultrasound data.

[0046] Advantages of the present application:

[0047] 1. The present application innovatively discloses a new method for simulating ultrasound stimulation desynchronization, introducing the intracavitary cavitation effect into the neural network model, and using a mathematical model to explore and analyze the desynchronization regulation effect of low-intensity transcranial ultrasound stimulation from the microscopic cell level to the macroscopic level of neurological diseases such as epilepsy, which can not only provide a biophysical mechanism for ultrasound regulation, but also provide a theoretical method for finding the optimal ultrasound parameters, and further provide theoretical guidance for the treatment of neurological diseases.

[0048] 2. This application innovatively uses an improved BLS membrane cavitation model and a small-world neural network for simulation. By solving the Fourier periodic solution and substituting it into a neural network model constructed using a small-world network approach, it is concluded that ultrasound stimulation has a desynchronizing effect on neural discharges and has a significant inhibitory effect on the onset of psychiatric disorders such as epilepsy.

[0049] 3. This application creatively discloses an ultrasonic stimulation desynchronization simulation device and system based on a neural network model. Through the in-depth interpretation of ultrasonic data and synchronization state control by the neural network model, combined with the collaborative analysis of the intra-membrane cavitation model, it can effectively solve the desynchronization problem of excitable neurons and significantly improve performance, making this application more accurately applied to the auxiliary control of the occurrence and development of mental illnesses related to ultrasonic data and the auxiliary selection of solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 is a schematic flow chart of ultrasound stimulation desynchronization simulation based on a neural network model provided by an embodiment of the present invention;

[0052] Figure 2 This is a structural diagram of a neuron double-layer acoustic cluster model based on the intra-membrane cavitation effect provided by an embodiment of the present invention;

[0053] Figure 3 This is a simulation diagram of the ultrasonic membrane cavitation mechanism based on the membrane cavitation model provided by an embodiment of the present invention;

[0054] Figure 4 Schematic diagram of an edge-added small-world network based on a neural network model provided by an embodiment of the present invention;

[0055] Figure 5 This is a diagram of the desynchronization effect based on the neural network model under different ultrasound stimulations provided by an embodiment of the present invention;

[0056] Figure 6 Schematic diagram of an ultrasonic stimulation desynchronization simulation device based on a neural network model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0058] In some of the processes described in the specification, claims, and above-mentioned drawings of the present invention, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The serial numbers of the operations, such as S101, S102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. Additionally, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel.

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0060] Figure 1 It is a schematic flowchart of ultrasonic stimulation desynchronization simulation based on a neural network model provided by an embodiment of the present invention. Specifically, the method includes the following steps:

[0061] S101: Obtain ultrasonic data.

[0062] In one embodiment, the obtained ultrasonic data includes sound pressure and / or frequency. For example, ultrasonic stimulation (frequency 500KHz, sound pressure 0.15MPa) is added; optionally, the ultrasonic data further includes power density, peak pressure amplitude, duty cycle, duration, and pulse repetition frequency.

[0063] S102: Input the ultrasonic data into an intracavitary cavitation model constructed based on the intracavitary cavitation effect to obtain an intracavitary capacitance displacement current.

[0064] In one embodiment, the ultrasonic data is input into BLS. BLS is a specific intracavitary cavitation model constructed based on the intracavitary cavitation effect. Based on the ultrasonic sound pressure, the bubble undergoes dynamic deformation, and the bubble radius Z(t). When Z(t) reaches a stable periodic solution, it is substituted into the capacitance Cm(Z) in the form of a Fourier series, thereby changing the average capacitance C of the membrane m , generating an intracavitary capacitance displacement current I Cm ; preferably, the equation of BLS is as follows:

[0065]

[0066]

[0067]

[0068]

[0069]

[0070] wherein, P ec is the effective pressure exerted by the charges on both sides of the membrane on the membrane, P in is the gas pressure inside the bubble, P0, P v , P A ·sin(ωt) are respectively the hydrostatic pressure in the external medium of the bubble, the saturated gas pressure inside the bubble, and the sound pressure of the external driving ultrasonic wave. Vm is the membrane potential, t is the time, f is the frequency of the ultrasonic wave, δ is the surface tension coefficient, μ is the viscosity coefficient, ɑ is the radius of the bubble boundary, Z0 is the radius of the initial boundary of the bubble, Δ is the initial gap between the two membranes, c is the speed of sound in the liquid medium, ε r is the relative permittivity of the membrane lumen, ε0 is the permittivity of the membrane lumen, C m0 is the cell membrane capacitance under the initial conditions, ρ l is the tissue density of the conductive medium.

[0071] Specifically, the BLS model (modified Rayleigh - Plesset equation of bubble dynamics) is the biophysical mechanism of cavitation effect in the membrane (the BLS model is set according to the original parameters, and these parameters are selected based on known or measured physical and biological quantities or ranges). The ultrasonic wave propagates into neurons at the nanoscale, and the positive and negative pressures cause the phospholipid bilayer membrane to form vacuoles. Electrically, the bilayer membrane can be approximately regarded as a capacitor. As Figure 2 is the structural diagram of the neuron bilayer acoustic cluster model based on the cavitation effect in the membrane provided by the embodiment of the present invention, showing the biomechanical and bioelectrical structures of the bilayer phospholipid molecular layer of cavitation in the membrane. In the BLS model of nanoscale bubbles, a circular and uniform phospholipid bilayer membrane patch is surrounded by a transmembrane protein constraint ring (maximum radius 32 nm). Generally, the incoming ultrasound induces the formation of bubbles (radius R) between specific phospholipid bilayers (bilayer sonophores BLS) on the cell membrane, generating cavitation.

[0072] In a specific example, when the bubble radius reaches a stable periodic solution, the capacitance Cm(Z) is brought in the form of a Fourier series, thereby changing the average capacitance Cm of the membrane and generating a membrane capacitance displacement current I Cm , and further causing the membrane potential Vm to change. As Figure 3 shows the detailed simulation process of the ultrasonic cavitation mechanism in the membrane. Specifically, the ODE113 function in MATLAB is used for solving, the solution step size is set to dt = 0.025 / f μs, f is the frequency of the ultrasonic wave, with the unit of MHz, and the charge is updated every 500 μs. From Figure 3As can be seen from (a), the alternating positive and negative pressures of ultrasound cause the bubbles between the phospholipid bilayers to compress and expand. The periodic solution of the Fourier series is substituted into the H-H model for solution. The M gate is the first to respond to the change in membrane potential. When the membrane potential exceeds -50 mV, an action potential is rapidly generated. Figure 3 (b) shows a schematic diagram of membrane discharge. In ultrasonic cavitation bubble dynamics, the negative pressure of ultrasonic waves pulls the molecular layer apart, and the positive pressure compresses it. The dynamic deformation of the cavitation bubble can induce the oscillation of the membrane capacitance Cm, the accumulation of charges on the cell membrane, and the membrane capacitance displacement current I Cm is generated. Under the drive of I Cm , the membrane potential Vm of the cavitation bubble oscillates between -200 mV and -60 mV. At the resting potential, the M gate related to the sodium ion channel is closed, the H gate is fully open, and the P and N gates related to the potassium ion channel are closed. When the ultrasonic stimulation ends, the membrane capacitance returns to the reference value, and the membrane potential exceeds -50 mV due to the accumulated charges. The M gate quickly responds and generates an action potential.

[0073] S103: Input the membrane capacitance displacement current into the neural network model to obtain the potentials of each neuron.

[0074] In one embodiment, the neural network model is constructed in the way of a small-world neural network. The small-world neural network uses N neurons of finite natural numbers as network nodes. Each node is symmetrically connected to its 2m (m << N) nearest adjacent nodes, and then two nodes are randomly selected from N nodes with a probability p for edge addition processing to obtain a small-world neural network with a connectivity of 2m / N without isolated nodes. The construction process of the small-world neural network is shown by the following differential equations, that is:

[0075]

[0076] Among them, x is the cell membrane potential of the neuron, y represents the recovery variable related to the internal current (such as sodium ions and potassium ions, etc.), z represents the slow-varying regulatory current related to the potassium ion current activated by calcium ions, C represents the coupling strength, N represents the number of neurons, I is the external DC excitation, and I Cm is the membrane capacitance displacement current. a, b, c, d, r, s, χ are common parameters for model construction, and matrix(i, j) is the connection matrix: when neurons Ni and Nj are coupled to each other, matrix(i, j) = 1; otherwise, matrix(i, j) = 0. Specifically, in the construction parameters of an example, they are as follows: a = 1.0, b = 3.0, c = 1.0, d = 5.0, r = 0.006, s = 4.0, χ = -1.6, C = 0.02, I = 1.4 mA.

[0077] Furthermore, neurons are obtained by coupling with a two-neuron model. Optionally, the two-neuron model includes one or more of the following models: Hindmarsh-Rose, FitzHugh-Nagumo, Wilson-Cowan, Hodgkin-Huxey, McCullochandPitts, Morris-Lecar, Phaseoscillatormodels; preferably, the two-neuron model is Hindmarsh-Rose.

[0078] Hindmarsh-Rose is a classical neuron model that can better simulate the characteristics of neurons in the hippocampus region of the brain. Existing research on it involves issues such as bifurcation, chaos, and synchronization.

[0079] FitzHugh-Nagumo contains a simplified model that describes oscillatory peak discharge neural dynamics with bistability.

[0080] Mapmodels is a discrete-time mapping model and also a simple neuron model that can generate spike states and bursting states. It calculates very fast, but lacks a biophysical basis.

[0081] Morris-Lecar is a simplified model that reduces the number of dynamic variables in the HH model. It shows the generation of action potentials when changing the current I, resulting in a saddle-node bifurcation to a limit cycle.

[0082] More preferably, the Hindmarsh-Rose neural network model includes a two-neuron coupled system that is completely synchronized. The membrane capacitance displacement current I Cm is input into the two-neuron coupled system that is completely synchronized, and then the time corresponding to the peak values of the potentials of the two neurons is calculated based on the firing activities of the neurons; preferably, the construction process of the two-neuron coupled system that is completely synchronized is as follows:

[0083]

[0084] where x is the membrane potential of the neuron, y represents the recovery variable related to the internal current, z represents the slow-varying regulatory current related to the calcium-activated potassium current, I is the capacitive position current generated by the ultrasonic stimulation acting on the neuron, C is the coupling strength between the two neurons, a, b, c, d, r, s, χ are parameters commonly used in the model construction process. In a specific construction, the parameters are as follows: a = 1.0, b = 3.0, c = 1.0, d = 5.0, r = 0.006, s = 4.0, χ = -1.6, C = 0.02, I = 1.4 mA.

[0085] In a specific embodiment, such as Figure 4The following is a schematic diagram of a small-world network with added edges based on a neural network model provided by an embodiment of the present invention, showing that the small-world neural network uses 20 neurons as network nodes, and each node is symmetrically connected to its 2m (m << 20) nearest adjacent nodes. Then, with a probability p = 0.8, 2 nodes are randomly selected from the 20 nodes for edge addition processing to obtain a small-world neural network with a connectivity of m / 10 and no isolated nodes.

[0086] S104: Calculate the network synchronization error of the neural network model based on the obtained neuron potentials.

[0087] In one example, the network synchronization error of the neural network model includes the network synchronization errors of all neurons in the network, which is obtained through the consistency difference of the discharge activities of the neuron potentials. Preferably, the specific calculation method is as follows:

[0088]

[0089] Among them, x, y, z represent the neuron potentials of the i-th neuron, and the network synchronization error e in the formula i represents the difference between the i-th neuron potential and the average potential of the neuron network. The larger the network synchronization error e i is, the greater the difference between the neuron potential and the average potential of the neural network.

[0090] S105: Quantify the desynchronization of the neural network model based on the obtained network synchronization error of the neural network model.

[0091] In one example, the desynchronization of the neural network model is obtained by calculating the average network synchronization error. The average network synchronization error is obtained by calculating the arithmetic mean of the network synchronization errors of all neurons in the network. The greater the average network synchronization error, the stronger the desynchronization effect.

[0092] The above method is used to assist in suppressing neural discharge synchronization and in the selection of treatment plans for mental diseases such as epilepsy and Parkinson's disease. The intravascular cavitation model BLS is combined with the small-world neural network model, and the simulation research results are as Figure 5 shown, and ultrasonic stimulation is introduced into the model of the small-world neural network. Figure 5 (a) shows the relationship between the average network error and the ultrasonic sound pressure at a frequency of 0.5 MHz. It can be seen from the figure that as the sound pressure increases, the average network error first increases and then decreases. At 0.35 MPa, the average network error reaches a maximum value of 0.6383, and the desynchronization effect of the neural network is the best, indicating that for ultrasonic stimulation at the same frequency, the optimal parameter of the sound pressure is 0.3 MPa. Figure 5(b) shows the relationship between the average network error and the ultrasonic frequency when the sound pressure is 0.15 MPa. As the frequency increases, although the average network error decreases before reaching the maximum value, the overall trend first gradually increases and then decreases with the increase of the ultrasonic frequency. The average network error reaches the maximum value of 1.7608 at 0.7 MHz, indicating that for the ultrasonic stimulation under the same ultrasonic sound pressure, the optimal parameter of the frequency is 0.7 MPa.

[0093] An ultrasonic stimulation desynchronization simulation system based on a neural network model provided by an embodiment of the present invention includes:

[0094] An acquisition module that acquires ultrasonic data, where the ultrasonic data includes sound pressure and / or frequency;

[0095] A first generation module that inputs the ultrasonic data into an intracavitary cavitation model constructed based on the intracavitary cavitation effect to obtain an intracavitary capacitance displacement current;

[0096] A second generation module that inputs the intracavitary capacitance displacement current into the neural network model to obtain the potentials of each neuron;

[0097] A calculation module that calculates the network synchronization error of the neural network model based on the potentials of each neuron;

[0098] An output module that obtains the desynchronization of the neural network model based on the network synchronization error of the neural network model.

[0099] Figure 6 An ultrasonic stimulation desynchronization simulation device provided by an embodiment of the present invention includes: a memory and a processor; the device may further include: an input device and an output device.

[0100] The memory, the processor, the input device, and the output device may be connected through a bus or other means. Figure 6 Taking the connection through the bus as an example;

[0101] The memory is used to store program instructions;

[0102] The processor is used to call the program instructions. When the program instructions are executed, it is used to execute the above-mentioned ultrasonic stimulation desynchronization simulation method based on the neural network model.

[0103] A computer-readable storage medium provided by the present invention stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned ultrasonic stimulation desynchronization simulation method based on the neural network model.

[0104] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems, devices, and modules can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0105] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be indirect couplings or communication connections through some interfaces, devices, or modules, and can be electrical, mechanical, or other forms.

[0106] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they can be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0107] In addition, in each embodiment of the present invention, the functional modules can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0108] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. The storage medium can include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disk, or optical disc, etc.

[0109] Those of ordinary skill in the art can understand that all or part of the steps in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. The storage media mentioned above can be read-only memory, magnetic disk, or optical disc, etc.

[0110] The above has introduced in detail a computer device provided by the present invention. For those of ordinary skill in the art, according to the idea of the embodiments of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. An ultrasonic stimulation desynchronization simulation method based on a neural network model, characterized in that, Including: Obtain ultrasonic data, where the ultrasonic data includes sound pressure and / or frequency; Input the ultrasonic data into an intracavitary cavitation model constructed based on the intracavitary cavitation effect to obtain an intracavitary capacitive displacement current; Input the intracavitary capacitive displacement current into a neural network model to obtain the potentials of each neuron. The neural network model is constructed in the way of a small-world neural network. The small-world neural network uses a finite natural number N of neurons as network nodes. Each node is symmetrically connected to its 2m nearest adjacent nodes, and then two nodes are randomly selected from the N nodes with a probability p for edge addition processing to obtain a small-world neural network without isolated nodes and with a connectivity of 2m / N; where m << N; Calculate the network synchronization error of the neural network model based on the potentials of each neuron. The network synchronization error is obtained from the consistency difference of the firing activities of the potentials of each neuron. The specific calculation method is as follows: Among them, , , , the network synchronization error represents the difference between the potential of the i-th neuron and the average potential of the neuron network, and x, y, and z represent the respective neuron potentials of the i-th neuron; Obtain the desynchronization of the neural network model based on the network synchronization error of the neural network model.

2. The ultrasonic stimulation desynchronization simulation method based on a neural network model according to claim 1, wherein Inputting the ultrasonic data into the in-membrane cavitation model constructed based on the in-membrane cavitation effect to obtain the membrane capacitance displacement current specifically includes: inputting the ultrasonic data into the BLS. The BLS drives the bubble to generate dynamic deformation of the bubble radius Z(t) based on the ultrasonic sound pressure. When Z(t) reaches a stable periodic solution, it is brought into the capacitance C m (Z) in the form of a Fourier series, thereby changing the average capacitance C of the membrane m , generating the membrane capacitance displacement current I Cm .

3. The ultrasonic stimulation desynchronization simulation method based on a neural network model according to claim 2, characterized in that The equation of the BLS is as follows: where, P ec is the effective pressure exerted on the membrane by the charges on both sides of the membrane, P in is the gas pressure inside the bubble, P0, P v , are respectively the hydrostatic pressure in the medium outside the bubble, the saturated gas pressure inside the bubble, and the external driving ultrasonic sound pressure, Vm is the membrane potential, t is the time, f is the frequency of the ultrasonic wave, δ is the surface tension coefficient, μ is the viscosity coefficient, ɑ is the radius of the bubble boundary, Z0 is the radius of the initial boundary of the bubble, Δ is the initial gap between the two membranes, c is the speed of sound in the liquid medium, ε r is the relative dielectric constant of the membrane lumen, ε0 is the dielectric constant of the membrane lumen, C m0 is the cell membrane capacitance under the initial conditions, ρ l is the tissue density of the conducting medium.

4. The ultrasonic stimulation desynchronization simulation method based on a neural network model according to claim 1, characterized in that The construction process of the small-world neural network is shown by the following differential equations, that is: where x represents the membrane potential of the neuron, y represents the recovery variable related to the inward current, z represents the slow-varying regulatory current related to the calcium-activated potassium current, C represents the coupling strength, N represents the number of neurons, I is the external DC excitation, I Cm is the membrane capacitance displacement current, a, b, c, d, r, s, χ are common parameters for model construction, and matrix(i,j) is the connection matrix: when neurons Ni and Nj are coupled to each other, matrix(i,j)=1; otherwise, matrix(i,j)=0.

5. The ultrasonic stimulation desynchronization simulation method based on a neural network model according to claim 1, characterized in that, The neurons are coupled through a two-neuron model.

6. The ultrasonic stimulation desynchronization simulation method based on a neural network model according to claim 5, characterized in that, The two-neuron model includes one or several of the following models: Hindmarsh-Rose, FitzHugh-Nagumo, Wilson-Cowan, Hodgkin-Huxey, McCullochandPitts, Morris-Lecar, Phaseoscillatormodels.

7. The ultrasonic stimulation desynchronization simulation method based on a neural network model according to claim 1, characterized in that The network synchronization error of the neural network model also includes the network synchronization errors of all neurons in the network.

8. The ultrasonic stimulation desynchronization simulation method based on a neural network model according to claim 7, wherein Obtain the desynchronization of the neural network model by calculating the arithmetic mean of the network synchronization errors of all neurons in the network.

9. An ultrasonic stimulation desynchronization simulation system based on a double neuron model, characterized in that, The system includes: An acquisition module that acquires ultrasonic data, where the ultrasonic data includes sound pressure and / or frequency; A first generation module that inputs the ultrasonic data into an intracavitary cavitation model constructed based on the intracavitary cavitation effect to obtain an intracavitary capacitive displacement current; A second generation module that inputs the intracavitary capacitive displacement current into a neural network model to obtain the potentials of each neuron. The neural network model is constructed in the way of a small-world neural network. The small-world neural network uses a finite natural number N of neurons as network nodes. Each node is symmetrically connected to its 2m (m << N) nearest adjacent nodes, and then two nodes are randomly selected from the N nodes with a probability p for edge addition processing to obtain a small-world neural network without isolated nodes and with a connectivity of 2m / N; A calculation module that calculates the network synchronization error of the neural network model based on the potentials of each neuron. The network synchronization error is obtained from the consistency difference of the firing activities of the potentials of each neuron. The specific calculation method is as follows: Among them, , , , the network synchronization error represents the difference between the potential of the i-th neuron and the average potential of the neuron network, and x, y, and z represent the potentials of each neuron of the i-th neuron; An output module that obtains the desynchronization of the neural network model based on the network synchronization error of the neural network model.

10. An ultrasonic stimulation desynchronization simulation device based on a neural network model, characterized in that, The device includes: a memory and a processor; The memory is used to store program instructions; The processor is used to call program instructions, and when the program instructions are executed, it is used to execute the ultrasonic stimulation desynchronization simulation method based on the neural network model described in any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the ultrasonic stimulation desynchronization simulation method based on the neural network model described in any one of claims 1-8.

Images