Individualized nerve stimulation parameter generation method based on electroencephalogram analysis

Abstract

The invention provides a personalized nerve stimulation parameter generation method based on electroencephalogram analysis, and aims to overcome the problem of parameter immobilization in the prior art. According to the method, a discomfort threshold value (DT) of an individual is determined through a stimulation tolerance test, and a personalized current intensity safety range is established. Then, combining an electroencephalogram risk assessment result, and utilizing a quantization formula to dynamically calculate stimulation parameters, including current intensity, stimulation duration and frequency; and finally, generating and outputting a personalized stimulation scheme to the stimulation equipment. According to the method, the defects of fixed parameters and one-step cutting in the prior art are overcome, dual individuation (safety and effectiveness) of stimulation parameters is realized, the safety and accuracy of transcranial electrical stimulation treatment are remarkably improved, and the method can be widely applied to auxiliary treatment of neurological and mental diseases such as depression, cognitive impairment and sleep disorder.

Description

Technical Field

[0001] This invention relates to the field of neuromodulation and cognitive intervention, and in particular to a method for generating personalized stimulation parameters based on electroencephalogram (EEG) analysis, which can be applied to the auxiliary intervention of neuropsychiatric diseases such as depression, cognitive impairment, and sleep disorders. Background Technology

[0002] Transcranial electrical stimulation (TCS), including tDCS, tACS, and tRNS, has been widely used in recent years as an adjunct treatment for neuropsychiatric disorders such as depression, Alzheimer's disease, and mild cognitive impairment due to its high safety, non-invasiveness, and potential for neuroplasticity modulation. However, current clinical practice often uses standardized or empirical stimulation parameters, such as fixed current intensity, stimulation duration, or frequency, neglecting individual differences in brain electrical activity patterns, cognitive states, and stimulation tolerance.

[0003] In practical applications, there are significant differences in the subjective perception and tolerance of stimuli among individuals. Some people experience significant discomfort even at low intensities, while others can tolerate higher intensities. Using uniform parameter settings may result in insufficient stimulation for some individuals, leading to insignificant therapeutic effects; or it may lead to overstimulation, causing side effects such as scalp discomfort, headaches, or even insomnia. Furthermore, while electroencephalogram (EEG) signals, as an important objective indicator reflecting brain function, can provide multi-dimensional information about cognitive load, emotion regulation, and sleep status, current methods have not yet deeply integrated EEG analysis results with the stimulation parameter generation process, making it difficult to dynamically optimize intervention programs based on individual real-time states.

[0004] Therefore, there is an urgent need for a method that, while ensuring safety, can dynamically calculate and output personalized stimulation parameters based on an individual's EEG risk characteristics and stimulation tolerance, thereby achieving truly precise electrical stimulation intervention. Summary of the Invention

[0005] The purpose of this invention is to overcome the problem of parameter fixation in the prior art and to propose a method for generating individualized neural stimulation parameters based on electroencephalography (EEG) analysis. By conducting stimulation tolerance tests on subjects, an individualized current intensity threshold range is established, and the stimulation intensity, duration, and frequency are dynamically quantified by combining EEG risk scores, thereby generating a stimulation program that conforms to individual characteristics, ensuring safety while improving the pertinence and effectiveness of intervention.

[0006] This invention provides a method for generating personalized stimulation parameters based on electroencephalogram (EEG) analysis, characterized by the following steps:

[0007] Step 1: Determination of the safe range of individual stimulation current intensity. This is achieved by applying a current to the user from the lowest intensity (I... minA progressive current stimulation is initiated, and combined with a standardized user feedback questionnaire, the current intensity value at which the user experiences unbearable stinging, burning, or significant pain is dynamically determined, i.e., the discomfort threshold (DT). Based on this discomfort threshold, a configurable safety margin ΔS (ranging from 0.1mA to 0.3mA) is introduced, thereby determining the individualized safe intensity range as [I min [DT-ΔS], the current intensity of all subsequent stimulation parameters must be strictly limited to this range.

[0008] Step 2: Quantitative calculation of parameters based on EEG risk assessment. After obtaining the individual's safety range, the parameters are combined with the user's EEG assessment results (including but not limited to: overall risk score R). total Risk scores for each cognitive domain and the upper limit of risk score R max (etc.), the specific stimulus parameters are calculated using the following formula:

[0009] 1) Formula for calculating current intensity (I):

[0010]

[0011] Where I is the calculated final stimulation current intensity; R total This is a comprehensive risk score obtained by weighted fusion of Z-scores from multiple EEG features, used to characterize the degree of abnormality in a user's brain function state; R max k is the preset upper limit of the risk score. i This is the current intensity adjustment coefficient, and its value is optimized through deep learning models or experimental data.

[0012] 2) Formula for calculating stimulus duration (T):

[0013]

[0014] Where T is the calculated final stimulus duration; T min and T max These are the preset lower and upper duration limits for the corresponding stimulus types; k t This is the duration adjustment factor, whose value is optimized through deep learning or experimental data.

[0015] For the tACS protocol, the stimulation frequency is set according to the intervention goal: if to improve attention and executive function, beta waves are selected with a frequency of 13Hz to 20Hz; if to regulate mood or promote neural network balance and plasticity, gamma waves are selected with a frequency of 30Hz to 45Hz; and if to enhance memory function and learning ability, theta waves are selected with a frequency of 4Hz to 8Hz.

[0016] Step 3: Personalized Stimulation Protocol Generation and Output. The calculated current intensity I, stimulation duration T, and frequency F are combined with the pre-defined stimulation target and stimulation type to generate a complete and executable personalized stimulation protocol instruction, which is then output to the stimulation device.

[0017] The beneficial effects of this invention are as follows:

[0018] This invention establishes a safe, accurate, and scalable framework for generating stimulation parameters by combining individualized tolerance testing with EEG risk analysis. Compared with traditional methods that rely on experience or fixed parameter settings, this invention has the following advantages:

[0019] First, by dynamically measuring an individual's discomfort threshold and introducing a safety margin, the current intensity is kept within an acceptable range, effectively avoiding adverse reactions caused by overstimulation. Second, by introducing an EEG risk score as a quantitative factor, the stimulation parameters can reflect the individual's current cognitive and emotional state, thereby improving the targeting of the intervention. Third, by using an adaptive parameter adjustment coefficient, the method can be continuously optimized during long-term intervention, meeting the requirements of closed-loop neuromodulation.

[0020] Finally, this method can be applied to the adjunctive treatment of depression and cognitive impairment, as well as sleep disorders and other neuropsychiatric diseases, and has broad clinical application value. Attached Figure Description

[0021] Figure 1 The overall flowchart of the personalized parameter generation method described in this invention. Detailed Implementation

[0022] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be understood that the embodiments described are for illustrative purposes only and are not intended to limit the scope of the invention.

[0023] like Figure 1 As shown in the figure, this embodiment provides a method for generating personalized stimulus parameters for individualized tACS intervention for the risk of mild cognitive impairment (MCI). The implementation process is as follows:

[0024] Step 1: Determination of the safe range of individual stimulus intensity

[0025] The user is placed in a quiet environment and puts on the transcranial electrical stimulation device. The electrodes are placed in a safe area according to the international 10-20 system to ensure the reliability and safety of the test. The device enters the "endurance test" mode, starting from the lowest current intensity that the system can output, and gradually increasing in a gentle ramp to avoid sudden current surges.

[0026] During the gradual increase, the system periodically collects subjective feedback from users through a human-computer interaction interface. When a user first reports significant discomfort, the current intensity is recorded as the discomfort threshold (DT). Subsequently, combined with a configurable safety margin, the system limits the individualized safe intensity range to between the minimum intensity and the discomfort threshold minus the safety margin. This safety range will serve as the basis for setting all subsequent stimulation parameters to ensure safety and tolerability.

[0027] Step 2: EEG signal acquisition and risk assessment

[0028] Users wear multi-channel EEG acquisition devices to collect resting-state, eye-closed EEG signals in a quiet environment. The system preprocesses the acquired raw EEG data, including bandpass filtering, artifact removal (eye movement, electromyography, and power line interference), and baseline correction, to ensure the effectiveness and stability of feature extraction.

[0029] After preprocessing, the system automatically extracts key features (such as prefrontal power in the theta band, peak α frequency, and prefrontal-parietal functional connectivity) and compares them with a healthy norm database of the same age, using the Z-score calculation formula:

[0030] Z=(X-μ) / σ

[0031] Where X represents the user's feature value, and μ and σ are the mean and standard deviation of the norm database, respectively. The degree to which a user deviates from the normal range is determined by |Z|<1 being normal, 1≤|Z|<2 being slightly abnormal, and |Z|≥2 being abnormal.

[0032] Then, based on the preset weighted model, the comprehensive risk score R is calculated. total and the set risk limit R max Normalization is performed to reflect the user's overall perceived risk level.

[0033] Step 3: Assessment of cognitive function risk

[0034] The system is based on R total With the preset risk threshold R threshold Make a judgment if R total ≥R threshold If the risk is positive, it is determined that "there is a cognitive / functional risk"; otherwise, it is determined that "there is no significant risk".

[0035] Step 4: Calculation of stimulus parameters based on quantification formula

[0036] Based on the individual's safety intensity range and risk score, the system calls the parameter quantification formula to calculate the current intensity I, stimulation duration T, and stimulation frequency F determined under the tACS condition.

[0037] 1) Formula for calculating current intensity (I):

[0038]

[0039] Where I is the calculated final stimulation current intensity; R total This is a comprehensive risk score obtained by weighted fusion of Z-scores from multiple EEG features, used to characterize the degree of abnormality in a user's brain function state; R max k is the preset upper limit of the risk score. i This is the current intensity adjustment coefficient, and its value is optimized through deep learning models or experimental data.

[0040] 2) Formula for calculating stimulus duration (T):

[0041]

[0042] Where T is the calculated final stimulus duration; T min and T max These are the preset lower and upper duration limits for the corresponding stimulus types; k t This is the duration adjustment factor, whose value is optimized through deep learning or experimental data.

[0043] Step 5: Solution Generation and Execution

[0044] The system ultimately generates a personalized stimulation instruction set, including stimulation mode, target brain region, current intensity, duration, and frequency, and sends it to the stimulation device. The user follows the intervention plan, and the system prompts for a follow-up EEG assessment after the intervention period to update the risk score R. total The stimulation parameters are recalculated to achieve dynamic optimization and closed-loop regulation.

Claims

1. A method for generating personalized stimulation parameters based on electroencephalogram (EEG) analysis, characterized in that, Includes the following steps: Step 1: Determination of the safe range of individual stimulation current intensity. This is achieved by applying a current to the user from the lowest intensity (I... min A progressive current stimulation is initiated, and combined with a standardized user feedback questionnaire, the current intensity value at which the user experiences unbearable stinging, burning, or significant pain is dynamically determined, i.e., the discomfort threshold (DT). Based on this discomfort threshold, a configurable safety margin ΔS (ranging from 0.1mA to 0.3mA) is introduced, thereby determining the individualized safe intensity range as [I min [DT-ΔS], the current intensity of all subsequent stimulation parameters must be strictly limited within this range. Step 2: Quantitative calculation of parameters based on EEG risk assessment. After obtaining the individual's safety range, the parameters are combined with the user's EEG assessment results (including but not limited to: overall risk score R). total Risk scores for each cognitive domain and the upper limit of risk score R max (etc.), the specific stimulus parameters are calculated using the following formula: 1) Formula for calculating current intensity (I): Where I is the calculated final stimulation current intensity; R total This is a comprehensive risk score obtained by weighted fusion of Z-scores from multiple EEG features, used to characterize the degree of abnormality in a user's brain function state; R max k is the preset upper limit of the risk score. i This is the current intensity adjustment coefficient, and its value is optimized through deep learning models or experimental data. 2) Formula for calculating stimulus duration (T): Where T is the calculated final stimulus duration; T min and T max These are the preset lower and upper duration limits for the corresponding stimulus types; k t This is the duration adjustment factor, whose value is optimized through deep learning or experimental data. For the tACS protocol, the stimulation frequency is set according to the intervention goal: if to improve attention and executive function, beta waves are selected with a frequency of 13Hz to 20Hz; if to regulate mood or promote neural network balance and plasticity, gamma waves are selected with a frequency of 30Hz to 45Hz; and if to enhance memory function and learning ability, theta waves are selected with a frequency of 4Hz to 8Hz. Step 3: Personalized Stimulation Protocol Generation and Output. The calculated current intensity I, stimulation duration T, and frequency F are combined with the pre-defined stimulation target and stimulation type to generate a complete and executable personalized stimulation protocol instruction, which is then output to the stimulation device.

2. The method according to claim 1, characterized in that, The "stimulation tolerance test" in step 1 can be conducted before the first use. The stimulation intensity starts from a low intensity and increases in fixed increments. After each increase, the stimulation is maintained for a period of time and user feedback is obtained.

3. The method according to claim 1, characterized in that, The current intensity adjustment coefficient k i And duration adjustment factor k t It is not a fixed value and can be adaptively optimized based on the user's historical treatment response data.

4. The method according to claim 1, characterized in that, The lower limit of the safety range is the minimum output intensity I of the device. min The upper limit is the unsuitable threshold DT minus a safety margin ΔS (its value range is 0.1mA to 0.3mA).

Images