A dual-mode implantable neuromodulation system for closed-loop intervention of epilepsy

Abstract

The present application relates to a kind of dual-mode implantable neuromodulation system for epilepsy closed-loop intervention, comprising: ultra-flexible nerve probe, for implanting target brain area, collection nerve electrophysiological signal, and according to the preset electric stimulation signal received to target brain area output electric stimulation;Subcutaneous drug release capsule, for storing antiepileptic drug, and release antiepileptic drug drug after receiving control signal;Neural signal acquisition module, for the nerve electrophysiological signal collected to carry out analog-digital conversion;Embedded control module, for the nerve electrophysiological signal after analog-digital conversion carries out time-frequency analysis, and utilizes neural network model to determine whether current neural activity is in epilepsy state;When in epilepsy state, control and ultra-flexible nerve probe connected electric stimulation output module exports preset electric stimulation signal, and exports control signal to subcutaneous drug release capsule.The present application can realize the quick, continuous and stable control of epilepsy attack.

Description

Technical Field

[0001] This invention relates to the field of neuromodulation technology, and in particular to a bimodal implantable neuromodulation system for closed-loop intervention in epilepsy. Background Technology

[0002] Epilepsy is a chronic neurological disorder caused by abnormal synchronous discharges of neurons in the brain. Its seizures are sudden and unpredictable, and in severe cases can lead to status epilepticus, loss of consciousness, secondary injuries, and even death. For some patients, conventional antiepileptic drugs are ineffective in controlling seizures, resulting in refractory epilepsy. Therefore, a closed-loop neuromodulation system capable of real-time seizure identification and timely intervention is of significant clinical importance.

[0003] Current treatment methods mainly include antiepileptic drugs, deep brain stimulation, vagus nerve stimulation, transcranial magnetic stimulation, and epilepsy focus resection. While antiepileptic drugs are widely used, they suffer from delayed onset of action, require long-term regular medication, are prone to drug tolerance, and have systemic side effects. Electrostimulation therapy can quickly suppress abnormal nerve discharges, but its duration of action is limited, and epilepsy may recur after stimulation stops. Most existing closed-loop epilepsy intervention systems rely on a single intervention method. For example, systems using electrostimulation alone can quickly suppress seizures but struggle to provide sustained pharmacological control; systems using drug delivery alone can maintain efficacy for a longer period, but there is a significant delay between drug release and onset of action, failing to cover the rapid intervention window in the early stages of a seizure. Furthermore, some existing systems rely on external high-performance computing equipment for epilepsy identification, which is insufficient to meet the requirements of implantable or portable closed-loop devices for low power consumption, miniaturization, and real-time performance. Traditional rigid electrodes may also exhibit mechanical mismatch with brain tissue, leading to inflammatory responses, decreased signal quality, and reduced stimulation stability during long-term implantation. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a dual-modal implantable neuromodulation system for closed-loop intervention of epilepsy, which can achieve rapid, continuous and stable control of epileptic seizures.

[0005] The technical solution adopted by this invention to solve its technical problem is: to provide a dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy, comprising:

[0006] An ultra-flexible neural probe is used to implant into a target brain region, collect neurophysiological signals, and output electrical stimulation to the target brain region according to the received preset electrical stimulation signal.

[0007] A subcutaneous drug-release capsule for storing an antiepileptic drug and releasing the antiepileptic drug upon receiving a control signal;

[0008] A neural signal acquisition module, connected to the ultra-flexible neural probe, is used to perform analog-to-digital conversion on the acquired neural electrophysiological signals;

[0009] An embedded control module, connected to the neural signal acquisition module, the ultra-flexible neural probe, and the subcutaneous drug release capsule, is used to perform time-frequency analysis on the analog-to-digital converted neural electrophysiological signals and to use a neural network model to determine whether the current neural activity is in an epileptic state. When in an epileptic state, the module controls the electrical stimulation output module connected to the ultra-flexible neural probe to output a preset electrical stimulation signal and outputs the control signal to the subcutaneous drug release capsule.

[0010] The ultra-flexible neural probe includes a probe body made of an insulating layer, on which multiple electrode points are distributed; a metal wire layer is disposed inside the probe body, and each electrode point is connected to a connecting pad through the metal wire layer. The connecting pad is used to connect to the neural signal acquisition module and the electrical stimulation output module.

[0011] A surface modification layer is provided on the surface of the electrode point, and the surface modification layer is used to reduce the impedance of the electrode point.

[0012] The subcutaneous drug-release capsule includes a drug storage cavity with a drug release outlet; an electrolytic electrode is disposed in the drug storage cavity, the electrolytic electrode is connected to the embedded control module, and an encapsulation layer is disposed outside the drug storage cavity.

[0013] The gap between the drug storage cavity and the electrolytic electrode is sealed with epoxy resin.

[0014] When the embedded control module performs time-frequency analysis on the analog-to-digital converted neurophysiological signal, it first performs regional averaging on the analog-to-digital converted neurophysiological signal, and then uses short-time Fourier transform to convert the regionally averaged neurophysiological signal into a time-frequency graph.

[0015] The neural network model is one or a combination of several of the following: LSTM network model, convolutional neural network model, gated recurrent unit model, Transformer model, temporal convolutional network model, support vector machine model, and random forest model.

[0016] The preset electrical stimulation signal output by the electrical stimulation output module is a charge-balanced biphasic current pulse signal.

[0017] Beneficial effects

[0018] Due to the adoption of the above-mentioned technical solutions, this invention has the following advantages and positive effects compared with the prior art: This invention constructs a dual-modal closed-loop regulatory mechanism that coordinates electrical stimulation and drug release. Electrical stimulation rapidly inhibits abnormal nerve discharge in the early stages of epileptic seizures, while simultaneously triggering subcutaneous drug release to provide continuous pharmacological control. The two complement each other on a time scale, compensating for the shortcomings of single electrical stimulation (short duration and easy relapse after stimulation cessation) and solving the problems of delayed onset and difficulty in covering the early intervention window of single drugs. This effectively shortens the duration of seizures and significantly improves the timeliness and long-term stability of epilepsy intervention. This invention uses an embedded neural network to achieve real-time identification and autonomous control of epileptic states, without relying on external high-performance computing equipment. It adapts to the miniaturized and low-power application requirements of implantable systems, achieving autonomous operation of the entire process of acquisition, identification, and intervention, ensuring the real-time nature of closed-loop intervention and device independence. The ultra-flexible neural probe used in this invention combines signal acquisition and electrical stimulation output functions, reducing mechanical damage to brain tissue and improving long-term signal stability. The subcutaneous electrolytic release drug capsule generates pressure through electrolysis to release the antiepileptic drug stored in the capsule on demand, which can be directly linked with the neural signal recognition results. Attached Figure Description

[0019] Figure 1 This is a framework diagram of a dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy according to an embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram of the response of the same neuron recorded by a flexible neural probe before and after electrical stimulation. Detailed Implementation

[0022] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0023] Embodiments of the present invention relate to a bimodal implantable neuromodulation system for closed-loop intervention in epilepsy, such as... Figure 1 As shown, it includes an ultra-flexible neural probe, a subcutaneous drug release capsule, a neural signal acquisition module, and an embedded control module.

[0024] The ultra-flexible neural probe is implanted into a target brain region to collect neurophysiological signals and can output electrical stimulation to the target brain region according to the received preset electrical stimulation signal. In this embodiment, the ultra-flexible neural probe includes a flexible insulating layer, a metal wire layer, electrode sites, connecting pads, and a surface modification layer.

[0025] The flexible neural probe in this embodiment includes 32 electrode channels. The electrode sites are located along the surface of the probe body, which is composed of a flexible insulating layer. These channels can be used to record local field potentials and neuronal firing in the target brain region, and also to apply electrical stimulation to the target brain region. The target brain region can be the hippocampus, thalamus, cortex, amygdala, or other brain regions associated with the occurrence and spread of epilepsy. The insulating layer in this embodiment can be made of polyimide, and a metal wire layer, made of materials such as titanium, gold, or nickel, is disposed within the insulating layer. The electrode sites in this embodiment have a surface modification layer to reduce electrode impedance, improve charge injection capability, and enhance electrical stimulation stability. This surface modification layer can be a platinum-iridium alloy. After platinum-iridium modification, the average impedance of the electrode sites at 1 kHz is approximately 87.21 kΩ. Lower electrode impedance is beneficial for improving the quality of neural signal acquisition and enhancing the stability of electrical stimulation output.

[0026] Because flexible neural probes have low mechanical strength and are not easily inserted directly into brain tissue, they can be implanted using a rigid guide device. For example, a tungsten wire with a diameter of approximately 50 μm can be used as a temporary guide device, and polyethylene glycol can be used to temporarily fix the flexible probe to the tungsten wire. After implantation into the target brain region, the tungsten wire is removed, leaving the flexible probe in the brain tissue. Figure 2 This demonstrates the response of the same neuron recorded before and after electrical stimulation using flexible neural electrodes.

[0027] The subcutaneous drug-release capsule in this embodiment is used to store antiepileptic drugs and release them upon receiving a control signal. The subcutaneous drug-release capsule includes a drug storage cavity, an electrolytic electrode, a drug release outlet, and an encapsulation layer.

[0028] In this embodiment, the drug storage chamber can be fabricated using 3D printing and includes a drug release outlet. Two platinum wires serve as electrolytic electrodes within the chamber. These platinum wires can be connected to an embedded control module via a flexible interconnect circuit. When the embedded control module detects a seizure, it outputs a control signal to apply a voltage to the electrolytic electrodes (i.e., the platinum wires), causing an electrolytic reaction in the water within the drug solution to produce oxygen and hydrogen. The gas generates pressure within the chamber, thereby propelling the drug solution out of the drug release outlet.

[0029] In this embodiment, the external casing of the drug storage chamber can be encapsulated with polydimethylsiloxane to form an encapsulation layer. This encapsulation layer can improve the device's biocompatibility, water resistance, electrical insulation, and mechanical protection performance. The gap between the drug storage chamber and the platinum wire can be sealed with epoxy resin to prevent drug leakage.

[0030] In this embodiment, the subcutaneous drug-release capsule can contain approximately 120 μL of antiepileptic drug solution. The drug can be diazepam solution, or other antiepileptic drugs suitable for subcutaneous or local release, such as benzodiazepines, barbiturates, levetiracetam, valproate, or combinations thereof. When a seizure is detected, an embedded control module applies approximately 3.3V to the platinum wire within the capsule to drive the electrolytic release process. In in vitro testing, the capsule releases approximately 90% of the stored liquid within approximately 75 seconds.

[0031] In this embodiment, the neural signal acquisition module is connected to the ultra-flexible neural probe and is used to perform analog-to-digital conversion on the acquired neural electrophysiological signals. It is worth mentioning that, before performing analog-to-digital conversion, the neural signal acquisition module in this embodiment can also amplify and filter the acquired neural electrophysiological signals.

[0032] In this embodiment, the embedded control module is connected to the neural signal acquisition module, the ultra-flexible neural probe, and the subcutaneous drug release capsule, respectively. It is used to perform time-frequency analysis on the analog-to-digital converted neural electrophysiological signals and use a neural network model to determine whether the current neural activity is in an epileptic state. When in an epileptic state, it controls the electrical stimulation output module connected to the ultra-flexible neural probe to output a preset electrical stimulation signal and outputs the control signal to the subcutaneous drug release capsule.

[0033] In this embodiment, the system acquires 16 channels of neural signals and uses the most recent 3 seconds of neural signals as the recognition window. The embedded control module first performs regional averaging or channel selection on the signals to reduce computational burden. Subsequently, a short-time Fourier transform is used to convert the time-domain signal into a time-frequency graph, and this time-frequency graph is input into the neural network model for epileptic state recognition.

[0034] The neural network model can be one or a combination of several of the following: LSTM network model, convolutional neural network model, gated recurrent unit model, Transformer model, temporal convolutional network model, support vector machine model, and random forest model. This embodiment uses an LSTM network model. The LSTM network is used to classify neural signals into non-epileptic states, interictal states, and epileptic states, assigning values ​​of 0, 1, and 2 to these states, respectively. The embedded control module calculates the average epilepsy score over a recent period. When the average epilepsy score of any region exceeds a preset threshold within the last 30 seconds, it is determined to be in an epileptic seizure state, at which point electrical stimulation and drug release are immediately triggered.

[0035] To accommodate the computational resource limitations of embedded platforms, neural network models can be trained on a GPU platform and then converted into IP cores deployable in programmable logic resources through high-level synthesis. Model parameters can be in half-precision floating-point format to reduce storage footprint and computational power consumption. The IP cores can communicate with the ARM processor or main control chip via direct memory access, enabling real-time inference and control.

[0036] In this embodiment, the preset electrical stimulation signal generated by the electrical stimulation output module can be a charge-balanced biphasic current pulse signal with a stimulation frequency of 130 Hz, a pulse width of 300 μs per phase, and a total pulse width of 600 μs for both phases. There can be no interval between the two phases. The duration of a single stimulation can be 3-6 seconds. The stimulation amplitude can be titrated according to individual circumstances, using the minimum current that can produce an anti-epileptic effect as the stimulation amplitude, and can be adjusted in 50 μA increments.

[0037] It should be noted that the stimulation parameters can be adjusted according to the target brain region, epilepsy type, individual threshold, and clinical needs. For example, the stimulation frequency can be set to 1–250 Hz, the monophasic pulse width can be set to 50–1000 μs, and the duration of a single stimulation can be set to 1–30 s.

[0038] This embodiment utilizes a dual-modal implantable neuromodulation system for closed-loop epilepsy intervention. After implantation, a flexible neural probe is positioned in the target brain region and continuously records neural electrical signals. A neural signal acquisition module converts analog neural electrical signals into digital signals and transmits them to an embedded control module. The embedded control module performs time-frequency analysis on the latest acquired neural signals and uses a neural network model to determine whether the current neural activity indicates an epileptic state. When the epileptic state score exceeds a preset threshold, the system assumes an epileptic seizure has occurred and immediately outputs preset electrical stimulation to the target brain region via the flexible neural probe. Electrical stimulation can rapidly intervene in abnormal synchronous discharges, shortening the duration of epileptic seizures. Simultaneously, the system outputs a control voltage to a subcutaneous drug release capsule. Platinum wire electrodes inside the capsule cause an electrolytic reaction in the water in the drug solution, generating gas and increasing the internal pressure of the capsule, propelling the drug solution out of the release port. After subcutaneous release, the drug is gradually absorbed and exerts its anti-epileptic effect. Due to a certain delay in drug onset, the system continues to identify neural signals before the drug has fully taken effect; if another epileptic seizure is detected, the system continues to trigger electrical stimulation. Once the drug takes effect, abnormal neural activity decreases, reducing the risk of epilepsy recurrence.

[0039] Therefore, the core control logic of the bimodal implantable neuromodulation system for closed-loop epilepsy intervention in this embodiment is as follows: continuously acquiring neural signals, identifying the epileptic state in real time, and triggering electrical stimulation and drug release simultaneously or in a preset sequence when the epilepsy score exceeds a threshold. Thus, the system can operate according to the following steps:

[0040] (1) The neural signal acquisition module continuously acquires neural signals from the target brain region;

[0041] (2) The embedded control module extracts the neural signals from the most recent 3 seconds;

[0042] (3) The embedded control module performs regional averaging or channel filtering on multi-channel neural signals;

[0043] (4) The embedded control module calculates the time-frequency characteristics of the neural signal through STFT;

[0044] (5) The LSTM network model outputs the epileptic state classification results;

[0045] (6) The embedded control module determines whether epilepsy has occurred based on the average epilepsy score of the most recent 30 seconds;

[0046] (7) When the epilepsy score exceeds the threshold, the embedded control module triggers electrical stimulation of the target channel;

[0047] (8) The embedded control module simultaneously triggers the subcutaneous drug release capsule to release antiepileptic drugs;

[0048] (9) During the period before the drug takes effect, the neural signal acquisition module continues to acquire and identify neural signals;

[0049] (10) If another epileptic seizure is detected, the embedded control module continues to trigger electrical stimulation;

[0050] (11) After the drug takes effect, the pre-epilepsy activity and abnormal discharge gradually decrease, and the embedded control module stops or reduces the output of electrical stimulation.

[0051] This control logic utilizes the rapid onset of electrical stimulation to cover the early window of epilepsy, while leveraging the sustained effect of drug release to reduce the risk of recurrence, thereby achieving synergistic control across two time scales.

[0052] The effects of the present invention are further illustrated below through a specific embodiment.

[0053] In this embodiment, a dual-modal closed-loop neuromodulation system is first fabricated. This system includes a 32-channel ultra-flexible neural probe, a subcutaneous drug release capsule, a flexible connection circuit, a neural signal acquisition chip, a PYNQ embedded control platform, an LSTM epilepsy recognition module, an electrical stimulation output module, and a data storage module.

[0054] Blue dye was injected subcutaneously into a drug-release capsule as a drug substitute. When a voltage was applied to a platinum wire electrode inside the capsule, an electrolytic reaction occurred, generating gas that propelled the dye out through the release outlet. Test results showed that approximately 90% of the stored liquid was released after about 75 seconds of voltage application, indicating that the capsule can achieve rapid, controllable, and on-demand release.

[0055] Electroencephalogram (EEG) or focal field potential (FLP) signals were collected from mice with epilepsy and categorized into normal, interictal, and epileptic states. Training data included normal neural signal segments, interictal signal segments, and epileptic seizure signal segments. An LSTM network model was first trained on a GPU workstation, and then the trained model was converted into a model or IP core that could run on an embedded platform.

[0056] On an embedded platform, the system captures the most recent 3 seconds of neural signals in real time, calculates time-frequency features, and inputs them into an LSTM model. The model outputs a classification result for the epileptic state. In this way, the system can perform real-time epilepsy identification on a low-power embedded platform and provide trigger signals for closed-loop intervention.

[0057] In a kainic acid-induced acute epilepsy model, the system monitored hippocampal neural signals in real time. Once the LSTM network model detected a seizure, the electrical stimulation module delivered charge-balanced biphasic pulses to the hippocampus via a flexible neural probe. The stimulation frequency was 130 Hz, with each phase pulse width of 300 μs and a total biphasic pulse width of 600 μs, with no phase intervals. Each stimulation lasted 3-6 seconds. The stimulation amplitude was determined through individualized titration. Animal experiments showed that hippocampal electrical stimulation rapidly inhibited epileptiform neural discharges. In 165 stimulation trials with 6 mice, approximately 75.16% of seizures were terminated by electrical stimulation; of the successfully terminated seizures, approximately 90% were suppressed within 10 seconds of stimulation. The median termination latency was approximately 3.6 seconds, and the average was approximately 4.7 seconds.

[0058] In a kainic acid-induced epilepsy model, drug capsules containing diazepam solution were implanted subcutaneously in the backs of mice. Once the mice reached a preset epileptic behavior score, the capsules released the drug. Changes in animal behavior and neural electrical signals were observed. The results showed that subcutaneous drug release reduced epileptic-like behavior and gradually stabilized neural electrical signals. There was a delay between drug release and onset of action, with an average onset time of approximately 14.59 min and a median of approximately 15.35 min. Compared to the control group receiving saline, the experimental group receiving the antiepileptic drug exhibited longer periods of normal behavior and better survival.

[0059] In the complete system, upon detecting the first epileptic seizure by the embedded recognition module, the system simultaneously triggers hippocampal electrical stimulation and subcutaneous drug release. Due to the delayed onset of drug action, the system continues to perform neural signal detection before the drug takes effect; if another epileptic seizure is detected, electrical stimulation is triggered again. Experimental results show that electrical stimulation can rapidly suppress abnormal discharges in the early stages of an epileptic seizure; after drug release, as the drug effect gradually appears, preepileptic activity and abnormal neural discharges gradually decrease. Dual-modal modulation achieves a synergistic effect of rapid control and sustained stability.

[0060] To evaluate the effect of bimodal modulation on neuronal activity, neural signals were collected one week before epilepsy modeling and one week after bimodal modulation. Typical neuronal waveforms were extracted using a neuronal classification algorithm, and the waveforms, peak-to-peak values, and signal-to-noise ratios (SNR) were compared before and after modulation. Results showed no significant changes in neuronal waveform morphology before and after bimodal modulation. The average peak-to-peak value of the eight typical neurons before modulation was approximately 137.51 μV, and the average SNR was approximately 11.83; after modulation, the average peak-to-peak value was approximately 146.48 μV, and the average SNR was approximately 11.60. Statistical analysis showed no significant difference before and after modulation, indicating that the system can maintain neural signal stability after electrical stimulation and drug release.

[0061] It is evident that this invention constructs a dual-modal closed-loop regulatory mechanism that synergizes electrical stimulation and drug release. Electrical stimulation rapidly inhibits abnormal neural discharges in the early stages of an epileptic seizure, while simultaneously triggering subcutaneous drug release to provide continuous pharmacological control. The two mechanisms complement each other on a timescale, compensating for the shortcomings of single electrical stimulation (short duration and high relapse rate after stimulation cessation) and single drug therapy (delayed onset and difficulty in covering the early intervention window). This effectively shortens seizure duration and significantly improves the timeliness and long-term stability of epilepsy intervention. This invention employs an embedded neural network to achieve real-time identification and autonomous control of epileptic states, eliminating the need for external high-performance computing equipment. It is compatible with the miniaturized and low-power application requirements of implantable systems, enabling autonomous operation of the entire process from data acquisition and identification to intervention, ensuring the real-time nature of the closed-loop intervention and device independence.

Claims

1. A dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy, characterized in that, include: An ultra-flexible neural probe is used to implant into a target brain region, collect neurophysiological signals, and output electrical stimulation to the target brain region according to the received preset electrical stimulation signal. A subcutaneous drug-release capsule for storing an antiepileptic drug and releasing the antiepileptic drug upon receiving a control signal; A neural signal acquisition module, connected to the ultra-flexible neural probe, is used to perform analog-to-digital conversion on the acquired neural electrophysiological signals; An embedded control module, connected to the neural signal acquisition module, the ultra-flexible neural probe, and the subcutaneous drug release capsule, is used to perform time-frequency analysis on the analog-to-digital converted neural electrophysiological signals and to use a neural network model to determine whether the current neural activity is in an epileptic state. When in an epileptic state, the module controls the electrical stimulation output module connected to the ultra-flexible neural probe to output a preset electrical stimulation signal and outputs the control signal to the subcutaneous drug release capsule.

2. The dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy according to claim 1, characterized in that, The ultra-flexible neural probe includes a probe body made of an insulating layer, on which multiple electrode points are distributed; a metal wire layer is disposed inside the probe body, and each electrode point is connected to a connecting pad through the metal wire layer. The connecting pad is used to connect to the neural signal acquisition module and the electrical stimulation output module.

3. The dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy according to claim 2, characterized in that, A surface modification layer is provided on the surface of the electrode point, and the surface modification layer is used to reduce the impedance of the electrode point.

4. The dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy according to claim 1, characterized in that, The subcutaneous drug-release capsule includes a drug storage cavity with a drug release outlet; an electrolytic electrode is disposed in the drug storage cavity, the electrolytic electrode is connected to the embedded control module, and an encapsulation layer is disposed outside the drug storage cavity.

5. The dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy according to claim 4, characterized in that, The gap between the drug storage cavity and the electrolytic electrode is sealed with epoxy resin.

6. The dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy according to claim 1, characterized in that, When the embedded control module performs time-frequency analysis on the analog-to-digital converted neurophysiological signal, it first performs regional averaging on the analog-to-digital converted neurophysiological signal, and then uses short-time Fourier transform to convert the regionally averaged neurophysiological signal into a time-frequency graph.

7. The dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy according to claim 1, characterized in that, The neural network model is one or a combination of several of the following: LSTM network model, convolutional neural network model, gated recurrent unit model, Transformer model, temporal convolutional network model, support vector machine model, and random forest model.

8. The dual-modal implantable neuromodulation system for closed-loop intervention in epilepsy according to claim 1, characterized in that, The preset electrical stimulation signal output by the electrical stimulation output module is a charge-balanced biphasic current pulse signal.

Images