Electrode adaptive control system based on multi-mode signal driving

Abstract

The invention discloses an electrode adaptive control system based on multi-modal signal driving, which is applied to multi-electrode adaptive design in a helmet-type electrode cap voltage acquisition system, and integrates a high-precision data acquisition module, a first contact quality evaluation coefficient generation module, a second contact quality evaluation coefficient generation module, a fusion decision module, an adaptive control module and a database. Self-adjustment of the electrodes in real time and electroencephalogram feedback data acquisition of the electrodes are realized. According to automatic adjustment of the electrode, the signal acquisition stability and the wearing comfort of the electrode are optimized by detecting the contact impedance coefficient of the electrode and the scalp, the electrode pressure gradient coefficient and the like, and a personalized electrode adjustment strategy is formulated.

Description

Technical Field

[0001] The present invention relates to the technical field of helmet - type electrode caps, and more specifically, to an electrode adaptive control system driven by multi - modal signals. Background Art

[0002] With the increasing attention of humans to brain science, the 21st century has been considered by the scientific community as the era of biological science and brain science, and voltage signals are widely used in clinical medicine. The voltage signal acquisition and analysis technology based on helmet - type electrode caps is applied to the assisted driving warning of drivers. First, through the change of voltage waves, driver fatigue and distraction can be detected and warned in advance, reducing accidents caused by slow reaction. Secondly, the system can also analyze the driver's emotional state and remind the driver to relax when nervous or road - raging, improving driving safety.

[0003] However, during the long - term driving of drivers, situations such as bumps, sweating, or heating of the user during driving can cause poor contact between the electrodes and the scalp. Moreover, due to differences in the head circumferences of users and the fact that the scalp at the data acquisition points is covered with hair, the hair layer will also affect the full contact between the electrodes and the scalp, which will also lead to unstable contact between the electrodes and the scalp, resulting in poor signal quality and inaccurate acquisition. Therefore, we propose an electrode adaptive control system driven by multi - modal signals. Summary of the Invention

[0004] An object of the present invention is to provide a new technical solution for an electrode adaptive control system driven by multi - modal signals.

[0005] According to a first aspect of the present invention, there is provided an electrode adaptive control system driven by multi - modal signals, which is applied to the position adaptive design of multiple electrodes in a helmet - type electrode cap voltage acquisition system, including:

[0006] A high - precision data acquisition module: used to acquire the electroencephalogram feedback data of multiple electrodes during the current detection time period and the change value of the contact pressure between each electrode and the head contact part; including the electrical contact stability coefficient, electrode synchronization index, and contact quality coefficient during the current detection time period;

[0007] A first contact quality evaluation coefficient generation module: used to receive the electroencephalogram feedback data acquired by each electrode during the current detection time period, and perform analysis to separately calculate the scalp contact impedance coefficient of each electrode, and the scalp contact impedance coefficient is used to evaluate the acquisition state of each electrode during the current detection time period;

[0008] The second contact quality assessment coefficient generation module: It is used to receive the contact pressure change values collected by each electrode during the current detection time period, analyze them, and obtain the electrode contact pressure gradient coefficients of each electrode. The contact pressure gradient coefficients are used to characterize the pressure contact states of the respective electrodes during the current detection time period;

[0009] The fusion decision-making module: It is used to obtain the scalp contact impedance coefficients of each electrode and the contact pressure gradient coefficients of each electrode, correct the scalp contact impedance coefficients through the contact pressure gradient coefficients, calculate the pressure output coefficients of each electrode separately, and formulate the pressure adjustment strategies for each electrode;

[0010] The adaptive control module: It is used to obtain the electrode pressure adjustment strategies and execute the implementation of the electrode pressure adjustment;

[0011] The database: It is used to obtain the electrode pressure adjustment records, generate adjustment logs, record the adjustment changes of each electrode, and is used to identify this user next time.

[0012] Optionally, each electrode is marked with a unique index to form a monomer marking set {1, 2,..., i,..., N}, where i represents the index marking of the electrode and N represents the total number of electrodes;

[0013] Each electrode is divided into a regional array to form an array marking set {1, 2,..., l,..., L}, where l represents the index marking of the target regional array and L represents the total number of regional arrays;

[0014] The regional array is divided according to the brain anatomical regions.

[0015] Optionally, an electrical contact stability coefficient is defined to characterize the stability of the conductance path between the single electrode and the head contact part during the current detection time period, and the electrical contact stability coefficient is denoted as ECSC-ER. The numerical value of the output ECSC-ER is coefficient-adjusted, and the output range is mapped to the interval (0, 1);

[0016] The specific expression of the electrical contact stability coefficient ECSC-ER is as follows:

[0017]

[0018] Among them: ECSC k is the steady-state conductance value at the kth sampling point within the analysis window, and is the skin conductance value collected by the electrode in real time;

[0019] ECSC t is the instantaneous steady-state conductance value at the current moment t;

[0020] ECSC max is the preset maximum conductance value, which is set according to the conductance limit value of the human skin;

[0021] Δt is the sampling interval;

[0022] T is the duration of the analysis window;

[0023] σ SSEC is the standard deviation of the ECSC;

[0024] When ECSC-ER approaches 0 more and more, it means that the stability of the conductance path between the electrode and the head contact part is worse;

[0025] When ECSC-ER approaches 1 more and more, it means that the stability of the conductance path between the electrode and the head contact part is better.

[0026] Optionally, an electrode synchronization index is defined to characterize the synchronous response ability of multiple electrodes in the regional array to the same physiological signal during the current detection period, and the electrode synchronization index is denoted as STCR-PCI, and the value range of STCR-PCI is set to the interval (0,1);

[0027] The expression of the electrode synchronization index STCR-PCI is specifically as follows:

[0028]

[0029] where: Δf m is the phase difference, f m is the instantaneous phase angle, specifically:

[0030] is the average phase;

[0031] ETCR m is the instantaneous response amplitude of the m-th electrode;

[0032] is the Hilbert transform of ETCR;

[0033] A m is the instantaneous amplitude;

[0034] A max is the amplitude normalization reference;

[0035] M is the number of pairs of adjacent electrodes;

[0036] When STCR-PCI approaches 0 more and more, it means that there are contact problems between local electrodes and the head contact part in the regional array;

[0037] When STCR-PCI approaches 1 more and more, it means that the cooperation between the electrodes in the regional array is better.

[0038] Optionally, a contact quality coefficient is defined to characterize the mechanical coupling state between a single electrode and the head contact part during the current detection period, and the electrical contact stability coefficient is denoted as SIDV-SM, and the value range of SIDV-SM is set to the interval (0,1);

[0039] The expression of the contact quality coefficient SIDV-SM is specifically:

[0040]

[0041] where: K is a normalization adjustment factor,

[0042]

[0043] M n is the nth order spectral moment, specifically:

[0044]

[0045] f c is the cut-off frequency, which is set according to the skin characteristics;

[0046] SIDV(f) is the impedance spectrum amplitude;

[0047] f is the frequency component;

[0048] ω n is the dynamic weight coefficient,

[0049] When SIDV-SM approaches 0 more, it indicates that the contact between the electrode and the head contact part is better;

[0050] When SIDV-SM approaches 1 more, it indicates that the contact between the electrode and the head contact part is worse.

[0051] Optionally, the electrode scalp contact impedance coefficient is denoted as C E and the expression is specifically:

[0052]

[0053] where: k1, k2, k3 are weight coefficients.

[0054] Optionally, the electrode contact pressure gradient coefficient is denoted as C P and the expression is specifically:

[0055]

[0056] where: ΔP is the pressure change gradient;

[0057] σ P is the pressure standard deviation;

[0058] D CE is the contact ellipticity;

[0059] D max is the contact ellipticity threshold;

[0060] F var is the dynamic friction coefficient variation;

[0061] F stable is the friction stability threshold;

[0062] K A , K B , K C are respectively the pressure change sensitivity, the contact shape weight, and the friction compensation coefficient.

[0063] Optionally, the correction of the scalp contact impedance coefficient by the contact pressure gradient coefficient is specifically as follows:

[0064] 1) Obtain the output data of the first contact quality evaluation coefficient generation module and the second contact quality evaluation coefficient generation module;

[0065] 2) Calculate the pressure coupling correction factor of each electrode by using the contact pressure gradient coefficient of each electrode;

[0066] 3) Correct the scalp contact impedance coefficient by the pressure coupling correction factor;

[0067] 4) Output the pressure output coefficient of each electrode;

[0068] 5) Formulate the pressure adjustment strategy of each electrode according to the preset strategy threshold.

[0069] Optionally, the expression of the pressure coupling correction factor is specifically:

[0070]

[0071] When C P > C E : d ∈ (0, 1], indicating that the contact pressure state of a single electrode in the current detection time period is better than its electrical contact state;

[0072] When C P = C E : d = 0

[0073] When C P < C E : d ∈ [-1, 0), indicating that the contact pressure state of a single electrode in the current detection time period is worse than its electrical contact state;

[0074] The correction of the scalp contact impedance coefficient is specifically:

[0075] C E ′=C E ·(1+|d| 1.2 ·sgn(d))

[0076] Among them: sgn function is:

[0077]

[0078] Optionally, the electrode pressure output coefficient is specifically:

[0079] P out =clip(P base +ΔQ+H,P min ,P max )

[0080] Among them: clip is the truncation function;

[0081] P base As the basic pressure,

[0082] P base =10·(C E ′-0.5)

[0083] ΔQ is the dynamic adjustment component,

[0084]

[0085] erf is the error function;

[0086] The pressure adjustment strategy includes:

[0087] Fine-tuning mode uses progressive pressure adjustment and pauses after completing an adjustment to verify the effect;

[0088] Active compensation mode to increase pressure regulation rate and sampling rate, and conduct real-time monitoring verification

[0089] According to one embodiment of the present disclosure, an electrode adaptive control system driven by multimodal signals realizes adaptive control of electrode pressure through multimodal fusion analysis and graded pressure adjustment strategy. Firstly, based on real-time collaborative analysis of ECSC-ER, STCR-PCI and SIDV-SM parameters, the accuracy of electrode contact state assessment is improved to ensure the quality of voltage signal acquisition. Secondly, a progressive pressure adjustment algorithm and angle fine-tuning linkage control are adopted to prevent user discomfort while maintaining good contact pressure, thereby avoiding excessive compression of the electrode and eliminating interference from external factors, so that the system can maintain an effective data acquisition rate under normal activity conditions.

[0090] The automatic adjustment of the electrodes of the present invention optimizes the signal acquisition stability and wearing comfort of the electrodes by detecting the contact impedance coefficient between the electrodes and the scalp, the electrode pressure gradient coefficient, etc., and formulates a personalized electrode adjustment strategy.

[0091] Other features and advantages of the present invention will become clear from the following detailed description of exemplary embodiments of the present invention with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] The drawings incorporated in and constituting a part of this specification illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0093] Figure 1 FIG. is a schematic diagram of the working process of an electrode adaptive control system based on multi-modal signal driving in one embodiment;

[0094] Figure 2 FIG. is a schematic diagram of the structural framework of an electrode adaptive control system based on multi-modal signal driving in another embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0095] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that: unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions and values set forth in these embodiments do not limit the scope of the present invention.

[0096] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0097] Technologies, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and devices should be considered as part of the specification.

[0098] In all examples shown and discussed herein, any specific values should be construed as merely exemplary and not as limitations. Thus, other examples of exemplary embodiments may have different values.

[0099] As Figure 1-2 shown, an electrode adaptive control system based on multi-modal signal driving is applied to the position adaptive design of multiple electrodes in a helmet-type electrode cap voltage acquisition system, and is characterized in that it includes:

[0100] A high-precision data acquisition module: used to acquire the electroencephalogram feedback data of multiple electrodes in the current detection time period, and the contact pressure change value between each electrode and the contact part of the head;

[0101] High-precision acquisition module, integrated with a 24-bit or 16-bit ADC to optimize the analog front end and an instrumentation amplifier with a common-mode rejection ratio > 120dB, directly embedded in the electrode base to reduce lead interference;

[0102] The EEG feedback data includes the electrical contact stability coefficient, electrode synchronization index, and contact quality coefficient for the current detection period;

[0103] The first contact quality assessment coefficient generation module: used to receive the EEG feedback data collected by each electrode during the current detection period and analyze it to separately calculate the scalp contact impedance coefficient of each electrode, and the scalp contact impedance coefficient is used to evaluate the acquisition status of each electrode during the current detection period;

[0104] The second contact quality assessment coefficient generation module: used to receive the contact pressure change values collected by each electrode during the current detection period and analyze them to obtain the electrode contact pressure gradient coefficient of each electrode, and the contact pressure gradient coefficient is used to characterize the pressure contact state of each electrode during the current detection period;

[0105] The fusion decision module: used to obtain the scalp contact impedance coefficient of each electrode and the electrode contact pressure gradient coefficient of each electrode, correct the scalp contact impedance coefficient through the electrode contact pressure gradient coefficient, separately calculate the pressure output coefficient of each electrode, and formulate the pressure adjustment strategy for each electrode;

[0106] The adaptive control module: used to obtain the electrode pressure adjustment strategy and execute the electrode pressure adjustment; the adaptive control module uses extended Kalman filter to fuse ISI / PAS, drives a micro servo motor to adjust the electrode pressure, and simultaneously monitors the inter-electrode coherence (> 0.8) in the Alpha band (8 - 13Hz);

[0107] The database: used to obtain the electrode pressure adjustment records, generate adjustment logs, record the adjustment changes of each electrode, and be used to identify the user next time.

[0108] It should be noted that the present invention is based on the position adaptive design of multiple electrodes in a helmet-type electrode cap voltage acquisition system. First, by using multi-parameter fusion technologies such as ECSC-ER, STCR-PCI, and SIDV-SM, the electrical contact states of each electrode are judged, and the pressure output coefficient is obtained by combining the pressure contact states of the electrodes. Finally, by formulating a pressure adjustment strategy, while ensuring the acquisition quality, the wearing comfort is guaranteed, it adapts to the head circumferences and hair blockages of different users, compensates in real time for signal attenuation caused by environmental factors such as sweating and hair blockage, eliminates poor electrode contact caused by individual head shape differences and micro-movements, reduces contact impedance fluctuations, and improves the effective data acquisition rate.

[0109] Specifically, each electrode is marked with a unique index to form a monomer marking set {1, 2, …, i, …, N}, where i represents the index mark of the electrode and N represents the total number of electrodes;

[0110] Each electrode is divided into a regional array to form an array marking set {1, 2, …, l, …, L}, where l represents the index mark of the target regional array and L represents the total number of regional arrays;

[0111] The regional array is divided according to the brain anatomical regions.

[0112] Furthermore, a dual-index marking system of monomer marking set and regional array marking set is implemented for the electrodes. First, the array marking based on the brain anatomical partition can quickly locate the abnormal signal source and improve the efficiency of voltage feature recognition, especially beneficial for scenarios such as event-related potentials that require regional analysis. Second, the mapping relationship between the monomer index and the regional mark supports a two-layer adjustment strategy of "regional coordination - monomer fine-tuning". When the overall impedance of a certain region increases, the pressure of all electrodes within the array can be adjusted in batches, while precise independent control can be implemented for local abnormalities of specific electrodes.

[0113] It should be noted that in the helmet-type electrode cap voltage acquisition system, it mainly consists of electrodes and a helmet-type cap body for installing the electrodes. Among them, the electrode array is arranged on the cap body, and the electrodes proposed in this application are arranged and divided according to the brain anatomical regions;

[0114] The regional array includes the frontal lobe region, temporal lobe region, parietal lobe region, occipital lobe region, and central region, and differential arrangement is adopted based on the brain region functions to achieve precise signal acquisition;

[0115] Exemplarily, the frontal lobe region is concentratedly covered by 6 - 8 electrodes on the prefrontal cortex and premotor area, focusing on capturing features of decreased attention and decision-making delay; the temporal lobe region is arranged with 4 - 6 electrodes, mainly covering the primary auditory cortex and the parahippocampal gyrus; the parietal lobe region is covered by 4 electrodes on the somatosensory cortex to detect somatic tactile feedback. The occipital lobe region is configured with 2 - 3 electrodes; the central region is deployed with 3 electrodes to synchronously collect voltage data around the central sulcus.

[0116] Specifically, an electrical contact stability coefficient is defined to characterize the stability of the conductance path between a single electrode and the head contact site during the current detection time period, and the electrical contact stability coefficient is denoted as ECSC-ER. The output ECSC-ER value is coefficient-adjusted to map the output range to the interval (0, 1);

[0117] The expression of the electrical contact stability coefficient ECSC-ER is specifically as follows:

[0118]

[0119] Where: ECSC k is the steady-state conductance value of the k-th sampling point within the analysis window, and is the skin conductance value collected by the electrode in real time;

[0120] ECSC t is the instantaneous steady-state conductance value at the current time t;

[0121] ECSC max is the preset maximum conductance value, which is set according to the conductance limit value of the human skin;

[0122] Δt is the sampling interval;

[0123] T is the duration of the analysis window;

[0124] σ SSEC is the standard deviation of ECSC;

[0125] Furthermore, the dynamic stability of the contact part between the electrode and the head is quantified by ECSC-ER, and whether the contact between the electrode and the head contact part is uniform and stable is judged according to the disorder degree of the conductance signal;

[0126] When ECSC-ER approaches 0 more and more, it indicates that the stability of the conductance path between the electrode and the head contact part is worse, manifested as: the conductance fluctuates violently, and the signal may be distorted due to electrode looseness, sweat interference or motion artifacts.

[0127] When ECSC-ER approaches 1 more and more, it indicates that the stability of the conductance path between the electrode and the head contact part is better, manifested as: the conductance signal is highly stable, indicating good contact between the electrode and the skin and reliable signal quality.

[0128] Specifically, the electrode synchronization index is defined to characterize the synchronous response ability of multiple electrodes in the regional array to the same physiological signal during the current detection period, and the electrode synchronization index is denoted as STCR-PCI, and the value range of STCR-PCI is set as the interval (0,1);

[0129] The expression of the electrode synchronization index STCR-PCI is specifically as follows:

[0130]

[0131] Where: Δf m is the phase difference, f m is the instantaneous phase angle, specifically:

[0132] is the average phase;

[0133] ETCR m is the instantaneous response amplitude of the m-th electrode;

[0134] is the Hilbert transform of ETCR;

[0135] A m is the instantaneous amplitude;

[0136] A max is the amplitude normalization reference;

[0137] M is the number of adjacent electrode pairs;

[0138] When STCR-PCI approaches 0 more and more, it indicates that there are contact problems at the contact parts between local electrodes and the head within the array of this area;

[0139] When STCR-PCI approaches 1 more and more, it indicates that the cooperation between the electrodes within the array of this area is better;

[0140] Specifically, a contact quality coefficient is defined to characterize the mechanical coupling state of a single electrode with the head contact part during the current detection time period, and the electrical contact stability coefficient is denoted as SIDV-SM, and the value range of SIDV-SM is set as the interval (0,1);

[0141] The expression of the contact quality coefficient SIDV-SM is specifically:

[0142]

[0143] Where: K is the normalization adjustment factor,

[0144]

[0145] M n is the nth-order spectral moment, specifically:

[0146]

[0147] f c is the cut-off frequency, which is set according to the skin characteristics;

[0148] SIDV(f) is the impedance spectrum amplitude;

[0149] f is the frequency component;

[0150] ω n is the dynamic weight coefficient,

[0151] When SIDV-SM approaches 0 more and more, it indicates that the contact between the electrode and the head contact part is better;

[0152] When SIDV-SM approaches 1 more and more, it indicates that the contact between the electrode and the head contact part is worse.

[0153] Specifically, the electrode scalp contact impedance coefficient is denoted as C E , and the specific expression is:

[0154]

[0155] where: k1, k2, k3 are weight coefficients.

[0156] It should be noted that k1 + k2 + k3 = 1;

[0157] Furthermore, the electrical contact stability coefficient ECSC-ER reflects the stability of the conductance path between a single electrode and the contact site on the head during the current detection period. According to scalp impedance experiment statistics, and impedance fluctuations will directly cause signal amplitude distortion, which has a significant impact on time-domain analysis such as ERP. It is concluded that the basic weight of the electrical contact stability coefficient ECSC-ER is 0.4. When a high-frequency voltage signal is detected, an additional adjustment weight of 0.05 is added

[0158] The electrode synchronization index STCR-PCI is used to characterize the synchronous response ability of multiple electrodes in a regional array to the same physiological signal during the current detection period. Based on neuroscience research support: Brain functional network research shows that the synchronization of local field potential LFP is crucial for cognitive tasks. In a 128-lead EEG system, when STCR-PCI < 0.4, the cross-electrode signal correlation ICC drops to 0.65, affecting the spatial resolution. Therefore, the reference weight of the electrode synchronization index STCR-PCI is 0.3, and the dynamic gain weight is 0.05;

[0159] The contact quality coefficient SIDV-SM is used to characterize the mechanical coupling state between a single electrode and the contact site on the head during the current detection period. Based on pressure sensor data, when SIDV-SM < 0.2, the risk of electrode detachment increases by 4 times, and poor mechanical coupling may cause electrode displacement and even scalp discomfort. Therefore, the basic weight of the contact quality coefficient SIDV-SM is 0.15, and the elastic weight is 0.05.

[0160] Specifically, the electrode contact pressure gradient coefficient is denoted as C P , and the specific expression is:

[0161]

[0162] where: ΔP is the pressure change gradient;

[0163] σ P is the pressure standard deviation;

[0164] D CE is the contact ellipticity;

[0165] Dmax is the contact ellipticity threshold;

[0166] F var is the dynamic friction variation coefficient;

[0167] F stable is the friction stability threshold;

[0168] K A K B K C are the pressure change sensitivity, contact shape weight, and friction compensation coefficient respectively.

[0169] Furthermore, where K A +K B +K C = 1, electrode c comprehensively quantifies and evaluates the contact state by integrating key parameters such as the standard deviation of fusion pressure, pressure change gradient, dynamic friction variation coefficient, and contact ellipticity, reducing the misjudgment rate of pressure value detection;

[0170] Specifically, the correction of the scalp contact impedance coefficient by the contact pressure gradient coefficient is as follows:

[0171] 1) Obtain the output data of the first contact quality evaluation coefficient generation module and the second contact quality evaluation coefficient generation module;

[0172] 2) Calculate the pressure coupling correction factor of each electrode using the contact pressure gradient coefficient of each electrode;

[0173] 3) Correct the scalp contact impedance coefficient by the pressure coupling correction factor;

[0174] 4) Output the pressure output coefficient of each electrode;

[0175] 5) Develop the pressure adjustment strategy for each electrode according to the preset strategy threshold.

[0176] Furthermore, ECSC-ER characterizes the stability of the conductance path between a single electrode and the head contact site during the current detection period, STCR-PCI characterizes the synchronous response ability of multiple electrodes within the regional array to the same physiological signal during the current detection period, and SIDV-SM characterizes the mechanical coupling state between a single electrode and the head contact site during the current detection period. The three jointly cover the time-frequency-space multi-dimensional characteristics of electrical contact;

[0177] Contact pressure gradient coefficient: Pressure identifies hair interference, evaluates the uniformity of pressure distribution, and avoids local overpressure or underpressure.

[0178] Specifically, the expression of the pressure coupling correction factor is as follows:

[0179]

[0180] When C P > C E : d ∈ (0, 1], indicating that the contact pressure state of a single electrode in the current detection period is better than its electrical contact state;

[0181] When C P = C E : d = 0

[0182] When C P < C E : d ∈ [-1, 0), indicating that the contact pressure state of a single electrode in the current detection period is worse than its electrical contact state;

[0183] The scalp contact impedance coefficient correction is specifically as follows:

[0184] C E ' = C E ·(1 + |d| 1.2 ·sgn(d))

[0185] where: the sgn function is:

[0186]

[0187] Furthermore, introducing a pressure coupling correction factor can dynamically quantify the non - linear coupling relationship between electrode pressure and bio - electrical signals, thereby significantly improving the accuracy and robustness of adaptive regulation. By deeply binding the mechanical contact state and electrophysiological characteristics: when the increase in pressure leads to a slowdown in the improvement of conductivity, the pressure regulation weight is automatically reduced to avoid discomfort caused by excessive pressure; conversely, when conductivity is sensitive to pressure, the adjustment amplitude is amplified to quickly stabilize the signal, enabling the system to distinguish the pressure that truly improves electrical contact from the pressure that only increases mechanical load.

[0188] Specifically, the electrode pressure output coefficient is specifically as follows:

[0189] P out = clip(P base + ΔQ + H, P min , P max )

[0190] where: clip is a truncation function;

[0191] P base is the base pressure,

[0192] P base = 10·(C E ' - 0.5)

[0193] ΔQ is the dynamic adjustment component,

[0194]

[0195] erf is the error function;

[0196] The pressure adjustment strategy includes:

[0197] Fine-tuning mode uses progressive pressure adjustment and pauses after completing an adjustment to verify the effect;

[0198] Furthermore, the progressive pressure regulation is an exponential decay regulation strategy, the initial step size is set to the reference pressure, and each subsequent adjustment step size is decreased by 0.8 of the previous step until the output pressure coefficient is reached;

[0199] Adjustment direction: Dynamic adjustment based on scalp contact impedance coefficient - contact pressure gradient coefficient. If the current pressure increases and causes the scalp contact impedance coefficient to decrease, reverse fine-tuning is performed;

[0200] Multi-parameter coordination: During the adjustment process, the electrical contact stability coefficient, electrode synchronization index and contact quality coefficient are monitored synchronously to ensure that pressure changes do not destroy other signal indicators;

[0201] Pause duration: After each adjustment, the system pauses and waits for the bioelectric signal to stabilize;

[0202] Verification standard: Short-term verification is within 50ms: Check whether the instantaneous fluctuation of the adjusted electrical contact stability coefficient converges;

[0203] The mid-term verification is after 200ms: evaluate whether the electrode synchronization index has returned to a safe range. If it does not reach the standard, trigger the second fine-tuning;

[0204] After the long-term trend is 1s: Analyze whether the contact quality coefficient shows a stable attenuation trend to avoid high-frequency impedance fluctuations.

[0205] Exit condition: If there is no significant improvement in the scalp contact impedance coefficient after three consecutive fine-tunings, switch to active compensation mode.

[0206] Specifically, the active compensation mode improves the pressure regulation rate and sampling rate, and performs real-time monitoring and verification.

[0207] Furthermore, the active compensation mode is activated when sudden interference is detected, such as intense exercise, sweating causing sudden changes in conductance, or when the fine-tuning mode fails, and uses high-frequency sampling plus real-time feedback control to restore signal quality at a faster adjustment rate;

[0208] Improved adjustment rate: The pressure change speed is increased to 3 to 5 times that of the fine-tuning mode, and dual-mode adjustment is adopted, with large adjustments in the early stage and gradual convergence in the later stage.

[0209] Dynamic step size adjustment: If the scalp contact impedance coefficient continues to deteriorate, it will increase to ensure rapid suppression of the deterioration trend;

[0210] If the scalp contact impedance coefficient begins to improve, it will be adjusted according to the error ratio;

[0211] Furthermore, the pressure adjustment strategy achieves a balance between high-precision stable monitoring and dynamic rapid response through the intelligent switching and collaborative optimization of the fine-tuning mode and the active compensation mode;

[0212] The fine-tuning mode adopts a progressive adjustment and pause verification mechanism to ensure the accuracy of pressure adjustment, avoid overshoot or under-adjustment, and is suitable for long-term stable monitoring and personalized adaptation;

[0213] The active compensation mode can quickly suppress sudden interferences (such as motion artifacts, electrode displacement) by increasing the adjustment rate and sampling frequency and combining real-time feedback control, ensuring signal stability under extreme conditions;

[0214] The seamless switching between the two modes enables the system to not only maintain reliable operation at low power consumption but also quickly recover signal quality in a high-dynamic environment.

[0215] In summary: The electrode adaptive control system based on multi-modal signal drive realizes the adaptive control of electrode pressure through multi-modal fusion analysis and hierarchical pressure regulation strategies. First, based on the real-time collaborative analysis of parameters such as ECSC-ER, STCR-PCI, and SIDV-SM, it improves the accuracy of electrode contact state assessment and ensures the quality of voltage signal acquisition; second, it adopts a progressive pressure regulation algorithm combined with angle fine-tuning linkage control to prevent discomfort of the user while maintaining good contact pressure, avoiding both excessive electrode compression and interference from external factors, enabling the system to maintain an effective data acquisition rate under normal activity conditions.

[0216] It should be noted that: All calculation formulas in this application document adopt regression analysis including but not limited to machine learning algorithms to deeply analyze the relevant parameters collected, identify their natural trends and interrelationships. Using professional software such as the Scikit-learn library in Python or the R language, a mathematical model matching the data is automatically generated. Then, the performance of the model is objectively evaluated through methods such as cross-validation, and combined with continuous feedback and optimization, to ensure that the created formula truly reflects the internal laws of the data, thus ensuring its effectiveness and accuracy, and ensuring that the calculation process conforms to the constraints of natural laws rather than being based on artificially set rules.

[0217] The technical solution of the present invention can be embodied in the form of a software product in essence or in terms of the part that contributes to the prior art. This computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk or optical disc of a computer, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of various embodiments of the present invention.

[0218] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus or device and execute the instructions), or used in combination with these instruction execution systems, apparatus or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate or transport a program for use by or in combination with an instruction execution system, apparatus or device.

[0219] It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solution of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solution of the present invention, and all of them should be covered by the scope of the claims of the present invention.

[0220] It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solution of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solution of the present invention, and all of them should be covered by the scope of the claims of the present invention.

Claims

1. An electrode adaptive control system driven by multi-modal signals, which is applied to the position adaptive design of multiple electrodes in a helmet-type electrode cap voltage acquisition system, is characterized in that Including: High-precision data acquisition module: used to collect electroencephalogram feedback data of multiple electrodes during the current detection period and the change value of the contact pressure between each electrode and the contact part of the head; And generate the electrical contact stability coefficient, electrode synchronization index, and contact quality coefficient for the current detection period; First contact quality evaluation coefficient generation module: used to receive the electroencephalogram feedback data collected by each electrode during the current detection period and perform analysis to separately calculate the scalp contact impedance coefficient of each electrode, and the scalp contact impedance coefficient is used to evaluate the acquisition state of each electrode during the current detection period; Second contact quality evaluation coefficient generation module: used to receive the change value of the contact pressure collected by each electrode during the current detection period and perform analysis to obtain the electrode contact pressure gradient coefficient of each electrode, and the contact pressure gradient coefficient is used to characterize the pressure contact state of each electrode during the current detection period; Fusion decision module: used to obtain the scalp contact impedance coefficient of each electrode and the electrode contact pressure gradient coefficient of each electrode, and correct the scalp contact impedance coefficient through the electrode contact pressure gradient coefficient to separately calculate the pressure output coefficient of each electrode and formulate the pressure adjustment strategy for each electrode; Adaptive control module: used to obtain the electrode pressure adjustment strategy, execute the electrode pressure adjustment through a micro servo motor, and synchronously monitor the coherence between electrodes; Database: used to obtain the electrode pressure adjustment record, generate an adjustment log, record the adjustment changes of each electrode, and be used to identify the user next time; 2. The electrode adaptive control system based on multi-modal signal driving according to claim 1, characterized in that: Mark each electrode with a unique index to form a monomer mark set {1, 2,..., i,..., N}, where i represents the index mark of the electrode and N represents the total number of electrodes; Divide each electrode into a regional array to form an array mark set {1, 2,..., l,..., L}, where l represents the index mark of the target regional array and L represents the total number of regional arrays; The regional array is divided according to the brain anatomical region.

3. The electrode adaptive control system based on multi-modal signal driving according to claim 1, wherein: Define the electrical contact stability coefficient to characterize the stability of the conductance path between a single electrode and the contact part of the head during the current detection period, and record the electrical contact stability coefficient as ECSC-ER. Adjust the coefficient of the output ECSC-ER value to map the output range to the interval (0, 1); The specific expression of the electrical contact stability coefficient ECSC-ER is as follows: where: ECSC k is the steady-state conductance value of the k-th sampling point within the analysis window, and is the skin conductance value collected by the electrode in real time; ECSC t is the instantaneous steady-state conductance value at the current moment t; ECSC max is a preset maximum conductance value, which is set according to the conductance limit value of human skin; Δt is the sampling interval; T is the analysis window duration; σ SSEC is the standard deviation of the ECSC; When ECSC-ER approaches 0 more, it indicates that the conductance path stability between the electrode and the contact part of the head is worse; When ECSC-ER approaches 1 more, it indicates that the conductance path stability between the electrode and the contact part of the head is better.

4. The electrode adaptive control system based on multi-modal signal driving according to claim 3, characterized in that: Define the electrode synchronization index to characterize the synchronous response ability of multiple electrodes in the regional array to the same physiological signal during the current detection period, and record the electrode synchronization index as STCR-PCI. Set the value range of STCR-PCI to the interval (0, 1); The specific expression of the electrode synchronization index STCR-PCI is as follows: Where: Δf m is the phase difference, f m is the instantaneous phase angle, specifically: is the average phase; ETCR m is the instantaneous response amplitude of the m-th electrode; is the ETCR Hilbert transform; A m is the instantaneous amplitude; A max is the amplitude normalization reference; M is the number of adjacent electrode pairs; When STCR-PCI approaches 0 more, it indicates that there are local electrodes in the regional array with contact problems with the contact part of the head; When STCR-PCI approaches 1 more closely, it indicates that the cooperation among the electrodes within the regional array is better.

5. The electrode adaptive control system based on multi-modal signal driving according to claim 1, characterized in that: Define the contact quality coefficient to characterize the mechanical coupling state of a single electrode with the head contact part during the current detection time period, and denote the electrical contact stability coefficient as SIDV-SM, and set the value range of SIDV-SM to the interval (0, 1); The specific expression of the contact quality coefficient SIDV-SM is as follows: Where: K is the normalization adjustment factor, M n is the spectral moment of order n, specifically: f c is the cut-off frequency, which is set according to the skin characteristics; SIDV(f) is the impedance spectrum amplitude; f is the frequency component; ω n is the dynamic weight coefficient, When SIDV-SM approaches 0 more closely, it indicates that the contact between the electrode and the head contact part is better; When SIDV-SM approaches 1 more closely, it indicates that the contact between the electrode and the head contact part is worse.

6. The electrode adaptive control system based on multi-modal signal driving according to claim 1, wherein: The electrode scalp contact impedance coefficient is denoted as C E , and the specific expression is as follows: Where: k1, k2, k3 are weight coefficients.

7. The electrode adaptive control system based on multi-modal signal driving according to claim 1, wherein: The electrode contact pressure gradient coefficient is denoted as C P , and the specific expression is as follows: Where: ΔP is the pressure change gradient; σ P is the standard deviation of pressure; D CE is the contact ovality; D max is the contact ovality threshold; F var is the dynamic friction coefficient variation; F stable is the friction stability threshold; K A ,K B ,K C are the pressure change sensitivity, the contact shape weight, and the friction compensation coefficient, respectively.

8. The electrode adaptive control system based on multi-modal signal driving according to claim 1, wherein: The correction of the scalp contact impedance coefficient by the contact pressure gradient coefficient is specifically as follows: 1) Obtain the output data of the first contact quality evaluation coefficient generation module and the second contact quality evaluation coefficient generation module; 2) Calculate the pressure coupling correction factor of each electrode by using the contact pressure gradient coefficient of each electrode; 3) Correct the scalp contact impedance coefficient by the pressure coupling correction factor; 4) Output the pressure output coefficient of each electrode; 5) Formulate the pressure adjustment strategy for each electrode according to the preset strategy threshold.

9. The electrode adaptive control system based on multi-modal signal driving according to claim 8, wherein: The specific expression of the pressure coupling correction factor is as follows: When C P > C E : d ∈ (0, 1], indicating that the contact pressure state of a single electrode in the current detection period is better than its electrical contact state; When C P = C E : d = 0 When C P <C E : d ∈ [-1, 0), indicating that the contact pressure state of a single electrode in the current detection period is worse than its electrical contact state; The correction of the scalp contact impedance coefficient is specifically as follows: C E ′ = C E ·(1 + |d| 1.2 ·sgn(d)) Where: the sgn function is:

10. The electrode adaptive control system based on multi-modal signal driving according to claim 8, characterized in that: The electrode pressure output coefficient is specifically as follows: P out = clip(P base + ΔQ + H, P min , P max ) Where: clip is the truncation function; P base is the base pressure, P base = 10·(C E ′ - 0.5) ΔQ is the dynamic adjustment component, erf is the error function; The pressure adjustment strategy includes: The fine-tuning mode, which adopts progressive pressure adjustment and pauses after one adjustment for effect verification; The active compensation mode, which improves the pressure adjustment rate and the sampling rate and conducts real-time monitoring and verification.

Images