A dual-mode electrophysiological signal synchronous acquisition telemetry device and method

Abstract

The application discloses a kind of synchronous acquisition telemetry devices and methods of bimodal electrophysiological signal, device includes signal connection electrode assembly, signal acquisition component and signal analysis component;Electrode assembly includes microneedle electrode subassembly, fixed subassembly and wearing subassembly, microneedle electrode subassembly includes first microneedle group, second microneedle group;Signal acquisition component is used to obtain the electrophysiological signal of target object at different preset points by electrode assembly synchronously, and the electrophysiological signal is preprocessed, and monitoring information is given;Signal analysis component is used to receive and analyze monitoring information, and give the telemetry result of bimodal electrophysiological signal.Through electrode design-circuit development-system integration and algorithm build-model analysis application, realize minimally invasive electrode, low-power synchronous circuit, wearable integration, physiological characteristic processing model technical fusion.

Description

Technical Field

[0001] This invention belongs to the field of electrophysiological monitoring technology, specifically relating to a device and method for synchronous acquisition and telemetry of dual-mode electrophysiological signals. Background Technology

[0002] Precise, continuous, and synchronous monitoring of electrophysiological signals such as electrocardiograms and electroencephalograms is a core technical means for studying the pathogenesis of neurogenic cardiovascular diseases, epilepsy, and other diseases, and is of great significance for promoting basic biomedical research.

[0003] Current research methods for heart-brain related diseases include quasi-static phenotypic observation (such as imaging and physiological parameter measurement at specific time points, and pathological section analysis of target organs at the endpoint) and omics analysis (gene sequencing, Western blotting, etc.). However, they lack simultaneous monitoring of dynamic multiple physiological indicators, which limits the study of the dynamic mechanisms of these diseases.

[0004] Currently, there is no electrode design scheme that simultaneously achieves low invasiveness, low impedance, contamination resistance, and rapid installation and removal. Existing implantable and wearable monitoring devices are far heavier than the physiological limits of the target subject's free movement. Forced wearing / implantation interferes with their natural behavior, leading to deviations between the acquired electrophysiological signals and the actual physiological state, thus failing to provide reliable data support for disease mechanism research. Although wearable monitoring solutions can achieve stable acquisition of single ECG or EEG signals, single signal transmission cannot achieve refined electrophysiological signal acquisition across multiple brain regions and leads, making it impossible to achieve precise monitoring and analysis of pattern diseases. The few devices that can achieve multimodal acquisition lack dedicated synchronous acquisition circuits and algorithms, resulting in significant time shifts in ECG and EEG signals.

[0005] For example, patent CN120267294A discloses a flexible multi-channel implantable bioelectrode, implantation device, and method. The bioelectrode consists of at least two single-channel electrodes fixed in a bundle by spiral winding or parallel fixation. Adjacent single-channel electrodes are fixed in contact and insulated from each other. Each single-channel electrode has an exposure site at its implantation segment for collecting neural electrical signals and / or electromyographic signals, or for neural electrical stimulation and / or electromyographic stimulation. Each single-channel electrode has a defined relative position and together form a multi-channel structure. The exposure sites are located at the end and / or side of the implantation segment of the single-channel electrode. However, this method uses implantable electrodes, which, while possessing the ability to monitor multiple physiological signals over a long period, a large number of channels, and high sampling accuracy, are highly invasive, have a large incision area, and the installation and removal of the electrodes are complex and time-consuming, significantly reducing monitoring efficiency. Furthermore, existing non-invasive wet electrodes are susceptible to motion interference and have a low signal-to-noise ratio, failing to meet the requirements for high-precision monitoring.

[0006] Therefore, how to achieve dynamic monitoring and accurate analysis of electrophysiological signals through in-depth research on wearable electrophysiological signals, while ensuring minimally invasive, lightweight, wireless, and synchronous monitoring, is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a device and method for synchronous acquisition and telemetry of dual-mode electrophysiological signals. Through a holistic technical solution encompassing electrode design, circuit development, system integration and algorithm construction, and model analysis and application, it integrates knowledge from multiple disciplines such as biomedicine, electronic circuits, 3D printing, machine learning, and embedded programming. This achieves a fusion of technologies including minimally invasive electrodes, low-power synchronization circuits, wearable integration, and physiological feature processing model adaptation, providing a standardized overall solution for electrophysiological monitoring technology and filling the technological gap in synchronous acquisition of multimodal signals in this field.

[0008] In a first aspect, the present invention provides a telemetry device for synchronous acquisition of dual-mode electrophysiological signals, comprising an electrode assembly for signal connection, a signal acquisition assembly, and a signal analysis assembly; The electrode assembly includes a microneedle electrode sub-assembly, a fixation sub-assembly, and a wearable sub-assembly. The microneedle electrode sub-assembly consists of multiple microneedle groups, including a first microneedle group and a second microneedle group. The first microneedle group and the second microneedle group are used to collect electrophysiological signals at different preset points on the target object. The fixation sub-assembly is used to fix the microneedle electrode sub-assembly at the preset points. The wearable sub-component is used to integrate the microneedle electrode sub-component, the fixation sub-component, and the signal acquisition component, and is worn and fixed to the target object; The signal acquisition component is used to synchronously acquire electrophysiological signals of the target object at different preset points through the electrode assembly, and to preprocess the electrophysiological signals to provide monitoring information; The signal analysis component is used to receive and analyze monitoring information and provide telemetry results of dual-mode electrophysiological signals.

[0009] Furthermore, the microneedle assembly includes electrodes, microtubes, and wires; The electrode has a barbed front end and a wire connection at the rear end. A microtube is fitted over the rear end of the electrode. The electrode substrate is stainless steel or titanium alloy, and the electrode surface is coated with a plating layer, which is at least one of elemental gold, elemental platinum, and platinum-iridium alloy. The microtube is made of an insulating polymer.

[0010] Furthermore, the electrode is prepared by the following steps: S1, the electrode substrate is pre-cut to obtain the first needle body; S2, based on the type of electrode substrate, a corresponding processing technology is adapted to process a three-dimensional barb structure on the first needle body to obtain the second needle body; S3, based on the material of the coating, a corresponding coating process is applied to prepare a coating on the surface of the second needle body to obtain an electrode.

[0011] Further, in step S2, a three-dimensional barb structure is processed on the first needle body to obtain the second needle body, specifically including the following steps: The first needle body is ablated and cut using a femtosecond laser and a multi-axis rotating fixture. The femtosecond laser pulse width is controlled to be <500fs, the single pulse energy is 1-100μJ, and the repetition frequency is 10-100kHz. A three-dimensional barbed structure is formed to obtain the second needle body.

[0012] Further, in step S2, a three-dimensional barb structure is processed on the first needle body to obtain the second needle body, specifically including the following steps: The first needle body is etched layer by layer using a micro-electro-discharge process and a multi-axis rotating fixture. A tungsten or copper-tungsten tool electrode with a diameter of 30-100μm is used to control the discharge current of 0.1-5A, the pulse width of 1-100μs, the pulse interval of 1-200μs, and the discharge gap of 2-10μm to form a three-dimensional barbed structure, thus obtaining the second needle body.

[0013] Further, in step S3, a coating is prepared on the surface of the second needle body, specifically including the following steps: A transition layer is sputtered onto the surface of the second needle body to obtain an electrode intermediate, wherein the thickness of the transition layer is 10-40 nm. Magnetron sputtering is employed, with the intermediate electrode placed at a first tilt angle on the sputtering disk and continuously rotated. The sputtering target is Pt or a Pt-Ir alloy, and the background vacuum is controlled to be <5×10⁻⁶ under an argon atmosphere. -4 Pa, working gas pressure 0.3-0.8 Pa, sputtering power 50-150 W, deposition rate 0.2-0.5 nm / s, wherein the first tilt angle is 40-60°.

[0014] Furthermore, the transition layer material is Ti or Cr.

[0015] Further, in step S3, a coating is prepared on the surface of the second needle body, specifically including the following steps: The second needle body is subjected to a nickel underlayer treatment to obtain an electrode intermediate. The electrode intermediate was subjected to pulse electroplating in a pre-prepared plating solution, with the cathode current density controlled at 0.5-2 A / dm³. 2 The pulse duty cycle is 10%-30%, the pulse frequency is 100-1000 Hz, the plating bath temperature is 45-55℃, the pH value is 8-9, and mechanical stirring is used in conjunction with cathode movement to prepare a coating with a thickness of 0.5-2 μm. The pre-prepared plating solution includes gold ions, sodium sulfite, and a leveling agent and / or a wetting agent. The concentration of gold ions is 8-12 g / L, the concentration of sodium sulfite is 80-120 g / L, the concentration of the leveling agent is 10-100 mg / L, and the concentration of the wetting agent is 50-200 mg / L. The leveling agent is at least one of a nitrogen-containing heterocyclic compound or a thiourea derivative, and the wetting agent is at least one of a fatty alcohol polyoxyethylene ether or a sulfonate surfactant.

[0016] Furthermore, the fixing sub-assembly includes a first fixing group and a second fixing group; The first fixing group includes a fixing patch, a fixing gel, and a base. The fixing patch adheres and fixes the first microneedle group to a first preset point on the target object. The fixing gel is used to fix the electrodes of the first microneedle group to the base. The first microneedle group is arranged in a single-lead or multi-lead manner. The second fixing group includes a base and a magnetic suction group. The magnetic suction group is used to magnetically fix the base. The base is provided with an array of positioning holes. The electrodes of the second microneedle group are embedded in the positioning holes. The base is positioned at the second preset point of the target object.

[0017] Furthermore, the wearable sub-component includes a main wearable structure, which is detachably connected to the microneedle electrode sub-component, the fixing sub-component, and the signal acquisition component; The main wearable structure includes a first part and a second part. The first part is adapted to a first preset point of the target object, and the second part is adapted to a second preset point of the target object. The first component is used to integrate the first microneedle group, the first fixing group, and the signal acquisition component; The second component is used to integrate the second microneedle group, the second fixing group, and the signal acquisition component.

[0018] Furthermore, the electrophysiological signal includes a first electrophysiological signal and a second electrophysiological signal; The signal acquisition component includes a first acquisition module, a second acquisition module, and a clock trigger module; The electrophysiological signals of the target object are collected synchronously at different preset points, specifically including the following steps: At the beginning of each sampling period, the clock trigger module simultaneously sends pulse signals to the first acquisition module and the second acquisition module; In response to the pulse signal, the first acquisition module and the second acquisition module start sampling at the same clock edge, latch the electrophysiological signal in real time, and attach the timestamp generated by the clock trigger module to give the first electrophysiological signal and the second electrophysiological signal.

[0019] Furthermore, the signal acquisition component also includes a dual-mode preprocessing module; The electrophysiological signals are preprocessed to provide monitoring information, specifically including the following steps: Obtain control commands and determine the preprocessing mode; Based on the preprocessing mode, perform either the first preprocessing or the second preprocessing; The first preprocessing step specifically includes the following steps: The first and second electrophysiological signals are filtered and compressed respectively to provide the first monitoring information; The second preprocessing step includes the following steps: The first and second electrophysiological signals were filtered respectively to eliminate motion artifacts and extract physiological features, and the second monitoring information was given. Motion artifact removal involves the following steps: The system performs time-frequency domain feature analysis using a pre-defined neural network model, identifies motion artifact segments and non-motion artifact segments, and generates mask values ​​for motion artifact segments. Based on the mask value, an adaptive elimination strategy is matched to correct the motion artifact clips; A sliding window overlap fusion mechanism is used to stitch non-motion artifact segments with corrected motion artifact segments to complete motion artifact elimination.

[0020] Furthermore, the signal analysis component includes an information recognition module and a feature analysis module; The information identification module identifies and processes the monitoring information to form information to be analyzed; The feature analysis module performs feature analysis on the information to be analyzed and provides the telemetry results of the dual-mode electrophysiological signal; The monitoring information is identified and processed to form information to be analyzed, which specifically includes the following steps: The first monitoring information is decompressed, motion artifacts are eliminated, and the first spatiotemporal alignment is performed. Then, feature extraction is performed to form the information to be analyzed; and / or, The second monitoring information is spatiotemporally aligned to form the information to be analyzed.

[0021] Furthermore, the first spatiotemporal alignment specifically includes the following steps: Based on the sampling window length, the first monitoring information is segmented and processed to give signal segment pairs; Perform cross-correlation analysis on all signal segment pairs and give the first cross-correlation coefficient; Remove signal segment pairs whose first cross-correlation coefficient is lower than a first preset threshold, and give a coarsely registered signal segment group; For all signal segment pairs in the coarsely registered signal segment group, interpolation upsampling is performed, followed by cross-correlation analysis to give the second cross-correlation coefficient; Retain signal segment pairs whose second cross-correlation coefficient is higher than the second preset threshold, and give a finely registered signal segment group; For all signal segment pairs in the finely registered signal segment group, extract the first feature point and the second feature point; Calculate the time difference between each first feature point and its corresponding second feature point, and then average the results to obtain the first translation amount; Based on the first translation amount, the signal segment is time-axis shifted using an interpolation algorithm to give aligned electrophysiological signal pairs.

[0022] Furthermore, the second spatiotemporal alignment specifically includes the following steps: Determine the first feature point sequence and the second feature point sequence in the second monitoring information; Electrophysiological events are identified from the first feature point sequence, and multiple electrophysiological event windows are provided. Based on each electrophysiological event window, the first feature point sequence and the second feature point sequence are divided into segments, and the first feature point sub-sequence and the second feature point sequence corresponding to each electrophysiological event window are given. Within each electrophysiological event window, a corresponding electrophysiological coupling weighted cost matrix is ​​constructed based on the first feature point subsequence and the second feature point subsequence. Based on the electrophysiological coupling weighted cost matrix, a dynamic time warping algorithm is used to search for the optimal alignment path within the corresponding electrophysiological event window; Based on the optimal alignment path within each electrophysiological event window, the corresponding local time offset sequence is determined, bidirectional coupling consistency verification is performed, and the verified local time offset sequence is given. Based on the validated local time offset sequence, the second feature point subsequence within each electrophysiological event window is reconstructed using timestamps, and aligned feature point pairs are given.

[0023] Furthermore, based on the electrophysiological coupling weighted cost matrix, a dynamic time warping algorithm is used to search for the optimal alignment path within the corresponding electrophysiological event window, specifically including the following steps: Determine the search start and end boundaries corresponding to the electrophysiological event window; Based on the electrophysiological coupling weighted cost matrix, the initial cumulative cost of the search starting point is determined, and the corresponding initial cumulative cost matrix is ​​given. Within the electrophysiological event window, combining monotonicity and continuity constraints, a dynamic programming algorithm is used to recursively calculate the cumulative cost point by point and update the corresponding initial cumulative cost matrix, thus providing the cumulative cost matrix. Based on the cumulative cost matrix, the position of the matrix element with the minimum cumulative cost is selected as the path endpoint on the termination boundary; Starting from the end of the path, and combining the preceding direction information, we backtrack to the starting point of the search to obtain the optimal alignment path.

[0024] Furthermore, the signal analysis component is also used to perform feature analysis based on a preset physiological feature processing model, and to provide telemetry results of dual-mode electrophysiological signals.

[0025] Feature analysis is performed based on a pre-defined physiological feature processing model to provide telemetry results of dual-mode electrophysiological signals. The specific steps include the following: Extract the first and second physiological characteristics from the information to be analyzed; Based on the first and second physiological characteristics, a correlation analysis is performed to determine the physiological coupling and regulatory relationships between the first and second physiological characteristics, and the results of the correlation analysis are given. Based on the correlation analysis results, the telemetry results of the dual-mode electrophysiological signals are presented.

[0026] Secondly, the present invention also provides a method for synchronous acquisition and telemetry of dual-mode electrophysiological signals, using the aforementioned synchronous acquisition and telemetry device for dual-mode electrophysiological signals, the method comprising the following steps: The microneedle electrode sub-assembly is fixed to a preset point on the target object by a fixing sub-assembly, and the microneedle electrode sub-assembly, fixing sub-assembly, and signal acquisition sub-assembly are integrated and worn on the target object by a wearable sub-assembly; The signal acquisition component is activated, and the electrophysiological signals of the target object at different preset points are acquired synchronously through the electrode component. The electrophysiological signals are then preprocessed to provide monitoring information. The monitoring information is received and analyzed by the signal analysis component, and the telemetry results of the dual-mode electrophysiological signal are given.

[0027] The dual-mode electrophysiological signal synchronous acquisition and telemetry device and method provided by the present invention have at least the following beneficial effects: (1) This invention adopts an integrated architecture of "electrode assembly - signal acquisition assembly - signal analysis assembly" for signal connection. Through the modular integration of microneedle electrode sub-assemblies, fixation sub-assemblies and wearable sub-assemblies, combined with the synchronous acquisition and preprocessing functions of the signal acquisition assembly and the intelligent analysis function of the signal analysis assembly, it realizes the synchronous, accurate and continuous acquisition and telemetry analysis of dual-mode electrophysiological signals in the free movement state of the target object. The device as a whole adopts a lightweight and modular design, which is adapted to the natural movement characteristics of the target object, effectively solving the problems of excessive load and data distortion caused by wire restraint in existing equipment, and ensuring the authenticity of electrophysiological signals.

[0028] (2) The microneedle electrode of the present invention adopts a barbed microneedle structure with a transdermal depth of less than 1 mm, reducing the wound area by more than 90% compared with traditional implantable electrodes, significantly reducing trauma to the target object; the barbed structure can effectively prevent electrode detachment caused by the target object's movement, ensuring the stability of dynamic signal acquisition. At the same time, the electrode is prepared by using a biocompatible metal substrate and a noble metal plating process, with good interface impedance control, ensuring high-fidelity acquisition of weak electrophysiological signals. The microneedle electrode sub-component and fixation sub-component adopt a modular, 3D-printed customized design, which shortens the device installation / removal time and significantly improves monitoring efficiency; the wearable sub-component adopts medical flexible materials and a 3D-printed customized structure, adapting to the natural movement characteristics of the target object, further ensuring the authenticity of electrophysiological signal acquisition.

[0029] (3) This invention achieves high-precision synchronous acquisition of electrophysiological signals at different preset points through a dual synchronization strategy of hardware unified clock triggering and software spatiotemporal alignment algorithm. Among them, the first spatiotemporal alignment and the second spatiotemporal alignment perform post-calibration of residual time offset at the signal level and feature level, respectively. The dual synchronization mechanism controls the synchronization error of dual-mode electrophysiological signals to less than 1 ms, effectively solving the core problem of lack of synchronization of multimodal signals and providing an accurate time reference for the study of physiological signal correlation mechanism. The second spatiotemporal alignment fully integrates the physiological characteristics of dual-mode signals. Through electrophysiological event window segmentation, construction of electrophysiological coupling weighted cost matrix and open endpoint strategy, combined with bidirectional coupling consistency verification mechanism, the alignment result is more consistent with the actual interaction of point physiological signals.

[0030] (4) This invention employs a dual strategy of hardware fixation optimization and algorithm elimination to achieve precise elimination of motion artifacts. By using a differentiated dual-group fixation strategy for the fixed sub-components, electrode displacement caused by target object movement is reduced at the source, thus lowering the generation of motion artifacts. A lightweight neural network model is used in the dual-mode preprocessing module to perform time-frequency domain feature analysis on the acquired electrophysiological signals, identifying motion artifact segments and non-motion artifact segments, matching adaptive elimination strategies for hierarchical correction, and employing a sliding window overlapping fusion mechanism to achieve seamless stitching. The synergistic effect of the dual strategies effectively suppresses motion interference introduced during free movement, improving the signal-to-noise ratio.

[0031] (5) The synchronous acquisition and telemetry device and method of the present invention can be directly used as a scientific research tool and applied to the basic biomedical research of universities and research institutes. The equipment adopts a modular design, is compatible with laboratory research and development and small-batch production, has low manufacturing cost and is easy to operate. It can be promoted and applied to multiple research fields such as neuroscience and cardiovascular disease. At the same time, it provides technical reference for the subsequent research and development of clinical-grade wearable electrophysiological monitoring devices, and has significant scientific research transformation value and market application prospects. Attached Figure Description

[0032] Figure 1 A schematic diagram of the structure of a telemetry device for synchronous acquisition of dual-mode electrophysiological signals provided by the present invention; Figure 2 This is a schematic diagram of a microneedle assembly structure according to one embodiment of the present invention; Figure 3 This is a schematic diagram of an electrode fabrication process according to one embodiment of the present invention; Figure 4 This is a schematic diagram of the first fixed group structure according to a certain embodiment of the present invention; Figure 5 (a) to (d) are respectively a schematic diagram, top view, partial internal view and bottom view of the second fixing group in a certain embodiment of the present invention; Figure 6 This is a schematic diagram of a wearable sub-component structure according to one embodiment of the present invention; Figure 7 A schematic diagram of the second spatiotemporal alignment process according to a certain embodiment of the present invention; Figure 8 A flowchart illustrating a method for synchronous acquisition and telemetry of dual-mode electrophysiological signals provided by the present invention; Figure 9 , Figure 10 These are the electrocardiogram (ECG) signal results and electroencephalogram (EEG) signal results synchronously acquired in this invention.

[0033] Explanation of reference numerals in the attached drawings: 1-microneedle group, 11-electrode, 12-microtube, 13-lead wire, 21-first fixation group, 211-fixation gel, 212-base, 22-second fixation group, 221-base, 2211-positioning hole, 222-magnetic suction group, 3-wearable sub-assembly, 31-first component, 32-second component. Detailed Implementation

[0034] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0035] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0036] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0037] Research in the field of wearable physiological electrical signal telemetry has made phased breakthroughs in two major directions: the integration of minimally invasive electrode technology and adaptability technology, and the integration of low-power circuits and wireless transmission technology. However, there are still problems such as the contradiction between invasiveness and signal quality, the pain point of device weight and wireless technology, the monitoring synchronization of multi-mode signals, and the lack of correlation analysis of specific pathological features.

[0038] Based on this, the present invention provides an overall technical solution of "electrode design - circuit development - system integration and algorithm construction - model analysis and application". By integrating knowledge from multiple disciplines such as biomedicine, electronic circuits, 3D printing, machine learning, and embedded programming, it realizes the technical integration of minimally invasive electrodes, low-power synchronous circuits, wearable integration, and physiological feature processing model adaptation, providing a standardized overall solution for electrophysiological monitoring technology and filling the technical gap in multimodal signal synchronous acquisition in this field.

[0039] In a first aspect, the present invention provides a telemetry device for synchronous acquisition of dual-mode electrophysiological signals, such as... Figure 1 As shown, the device includes an electrode assembly for signal connection, a signal acquisition assembly, and a signal analysis assembly; The electrode assembly includes a microneedle electrode sub-assembly, a fixation sub-assembly, and a wearable sub-assembly. The microneedle electrode sub-assembly consists of multiple microneedle groups, including a first microneedle group and a second microneedle group. The first microneedle group and the second microneedle group are used to collect electrophysiological signals at different preset points on the target object. The fixation sub-assembly is used to fix the microneedle electrode sub-assembly at the preset points. The wearable sub-component is used to integrate the microneedle electrode sub-component, the fixation sub-component, and the signal acquisition component, and is worn and fixed to the target object; The signal acquisition component is used to synchronously acquire electrophysiological signals of the target object at different preset points through the electrode assembly, and to preprocess the electrophysiological signals to provide monitoring information; The signal analysis component is used to receive and analyze monitoring information and provide telemetry results of dual-mode electrophysiological signals.

[0040] The dual-mode electrophysiological signal synchronous acquisition and telemetry device provided by the present invention adopts an integrated architecture of "electrode assembly - signal acquisition assembly - signal analysis assembly". Each assembly establishes an electrical signal transmission path through a signal transmission link, enabling the transmission of electrophysiological signals and related monitoring information between the assemblies, and realizing a complete technical closed loop from signal acquisition, synchronous preprocessing to intelligent analysis.

[0041] The target subject refers to the active object whose electrophysiological signals need to be monitored. In a specific embodiment, this is an experimental animal, such as a mouse or rat. The overall weight of the device is controlled within 15%-20% of the target subject's body weight to ensure that the target subject can freely eat, nest, and move after wearing the device, without obvious behavioral abnormalities. In a specific embodiment, the dual-mode electrophysiological signals include electrocardiogram (ECG) and electroencephalogram (EEG). The ECG signal reflects cardiac electrical activity, and the EEG signal reflects central nervous system electrical activity. Both are collected simultaneously to study the heart-brain interaction mechanism. Telemetry results refer to the final analysis conclusions output by the signal analysis component, which may include synchronized and aligned electrophysiological characteristic parameters, electrophysiological correlation analysis results, and risk warning information for specific diseases.

[0042] The electrode assembly serves as the front end for electrophysiological signal acquisition and includes a microneedle electrode subassembly, a fixation subassembly, and a wearable subassembly.

[0043] The microneedle electrode sub-assembly includes multiple microneedle groups, each comprising at least a first microneedle group and a second microneedle group, used to collect electrophysiological signals from the target subject at different preset points. Preset points refer to electrophysiological signal acquisition locations determined in advance according to monitoring needs. The first preset point may correspond to the chest, abdomen, or back of the target subject for electrocardiogram (ECG) signal acquisition; the second preset point may correspond to a specific brain region on the head of the target subject for electroencephalogram (EEG) signal acquisition.

[0044] The fixing sub-assembly is used to fix the microneedle electrode sub-assembly at preset points to ensure the stability of signal acquisition. The wearable sub-assembly is used to integrate the microneedle electrode sub-assembly, the fixing sub-assembly, and the signal acquisition assembly, and to wear and fix the entire device to the target object, achieving device lightweighting and modularity.

[0045] Furthermore, such as Figure 2 As shown, each microneedle assembly 1 includes an electrode 11, a microtube 12, and a wire 13; The front end of electrode 11 has a barbed structure, the rear end of electrode 11 is connected to wire 13, and a microtube 12 is sleeved on the rear end of electrode 11. The substrate of electrode 11 is stainless steel or titanium alloy, and a coating is attached to the surface of electrode 11. The coating is at least one of elemental gold, elemental platinum, and platinum-iridium alloy. The microtube 12 is made of insulating polymer.

[0046] The electrode, as the core sensing element, employs a barbed microneedle structure to achieve low-impedance contact with biological tissue. A lead wire is positioned at the rear end of the electrode to transmit the electrophysiological signals acquired by the electrode to the signal acquisition component.

[0047] The electrode tip features a barbed structure, meaning three-dimensional barbs are machined onto the metal needle body to enhance mechanical anchoring with superficial skin tissue. Once the electrode is inserted transdermally, the barbs hook onto the dermal tissue beneath the stratum corneum, forming a mechanical locking anchor. This effectively prevents electrode displacement and detachment caused by muscle contraction, scratching, or equipment pulling during free movement of the target subject, ensuring the continuity and stability of signal acquisition.

[0048] Preferably, the electrode uses a barbed dry microneedle with a transdermal depth of less than 1 mm, penetrating only the stratum corneum of the epidermis and anchoring to the dermis through the barbed structure, avoiding damage to deep subcutaneous tissues and blood vessels and nerves. The wound area is reduced by more than 90% compared to traditional implantable electrodes, achieving minimally invasive signal acquisition. The actual effective insertion depth is determined based on the skin thickness at the preset points on the target subject, through tissue sectioning and insertion depth experiments, avoiding entry into deep subcutaneous tissues. The barbed structure enhances the anchoring ability of superficial skin; the size, angle, and number of barbs are optimized between anchoring force and tissue damage, and verified through insertion force, pull-out force, histological inflammation score, and impedance stability.

[0049] The electrode substrate is made of biocompatible stainless steel or titanium alloy. Stainless steel (such as medical-grade 316L stainless steel) has excellent mechanical strength and processing performance, is well-suited for laser micromachining processes, and is suitable for fabricating slender microneedles at a relatively low cost. Titanium alloy (such as Ti-6Al-4V ELI) has good biocompatibility and corrosion resistance, and its low density helps reduce the overall weight of the device, making it well-suited for micro-electro-discharge machining processes. The appropriate substrate material can be selected based on the target object's body size, activity level, and acquisition cycle in different acquisition scenarios.

[0050] A noble metal coating is prepared on the surface of a barbed microneedle electrode. The coating material includes at least one of elemental gold, elemental platinum, and a platinum-iridium alloy. The noble metal coating exhibits high chemical stability and excellent conductivity, significantly reducing the electrode-tissue interface impedance (10 kΩ·cm). 2 (The following) This reduces electrode polarization and ensures high-fidelity acquisition of weak electrophysiological signals (especially microvolt-level EEG signals). Furthermore, the noble metal coating has good biocompatibility, reducing tissue inflammation and fibrous encapsulation caused by long-term implantation, thus ensuring the stability of electrode signal acquisition quality during continuous testing.

[0051] Among them, gold plating has low resistivity and is easy to prepare quickly using electroplating processes; platinum and platinum-iridium alloy plating has higher hardness and stronger adhesion to the needle body, making it suitable for scenarios that require repeated use.

[0052] The rear end of the electrode is fitted with a microtube made of an insulating polymer, specifically an organic polymer insulating polymer, such as polyimide (PI), polytetrafluoroethylene (PTFE), polyurethane (PU), and other high-performance organic polymer insulating polymers. These materials possess excellent insulation, biocompatibility, flexibility, corrosion resistance, and high-temperature stability, making them ideal for the insulation and protection of minimally invasive medical electrodes.

[0053] For example, in one embodiment, a polyimide microtube is fitted onto the rod of the electrode (the rear 1 / 3 to 1 / 2 of the needle length), initially positioned outside the skin. The polyimide microtube is used to assist in the non-invasive removal of the electrode: when the electrode needs to be removed, a small amount of medical lubricant is applied around the puncture point, and the pre-placed polyimide microtube is pushed along the electrode towards the skin. The tip of the microtube completely covers all barbed areas, opening up the barbed structure, allowing the microtube and electrode to be smoothly pulled out together, avoiding secondary damage to the tissue caused by the barbs.

[0054] Polyimide possesses excellent mechanical strength, chemical stability, and biocompatibility. Its minimal deformation capability can accommodate the slight oscillations of the microneedle electrode, ensuring that the microtube remains in a stable position during electrode use without affecting signal acquisition quality.

[0055] like Figure 3 As shown, the electrodes in the aforementioned microneedle assembly can be prepared through three main processes: needle body cutting, barb structure processing, and surface metal plating. Specifically, the electrodes are prepared through the following steps: S1, the electrode substrate is pre-cut to obtain the first needle body; S2, based on the type of electrode substrate, a corresponding processing technology is adapted to process a three-dimensional barb structure on the first needle body to obtain the second needle body; S3, based on the material of the coating, a corresponding coating process is applied to prepare a coating on the surface of the second needle body to obtain an electrode.

[0056] Specifically, the electrode substrate is pre-cut to a preset size to obtain the first needle. The substrate is preferably made of stainless steel or titanium alloy, and the needle diameter is controlled within the range of 200-1000 μm. The pre-cutting process includes cutting the substrate to a fixed length and flattening the end face to meet the requirements of subsequent precision machining.

[0057] Next, a three-dimensional barb structure is processed on the first needle body. Different processing techniques can be used to process the three-dimensional barb structure.

[0058] In a specific embodiment, step S2, processing a three-dimensional barb structure on the first needle body to obtain the second needle body, can be achieved through the following steps: A femtosecond laser is used in conjunction with a multi-axis rotating fixture to ablate and cut the first needle body. The femtosecond laser pulse width is controlled to be <500fs, the wavelength can be 1030nm or 515nm, the single pulse energy is 1-100μJ, and the repetition frequency is 10-100kHz. A three-dimensional barbed structure is formed to obtain the second needle body.

[0059] Specifically, the first needle is clamped on a multi-axis rotating fixture, and its surface is ablated and cut using a femtosecond laser to form a three-dimensional barbed structure. The femtosecond laser processing parameters include: pulse width of 100 fs to 10 ps, ​​preferably <500 fs; wavelength preferably 1030 nm or 515 nm; single pulse energy of 1-100 μJ; and repetition frequency of 10-100 kHz. This process allows for a minimum linewidth control of 10-20 μm, resulting in a heat-affected zone of less than 5 μm, preferably less than 1 μm, effectively avoiding molten metal burrs. Furthermore, it eliminates the need for molds, offering flexible design suitable for small-batch, rapid iteration. After femtosecond laser processing, ultrasonic cleaning, electrochemical polishing, plasma cleaning, or a combination thereof are performed to remove processing residues, redeposited particles, and edge microburrs, thereby improving electrode surface uniformity and bio-contact stability. When stainless steel is used as the substrate, femtosecond laser processing is preferred.

[0060] In another specific embodiment, step S2, processing a three-dimensional barb structure on the first needle body to obtain the second needle body, can be achieved through the following steps: The first needle body is etched layer by layer using a micro-electro-discharge process and a multi-axis rotating fixture. A tungsten or copper-tungsten tool electrode with a diameter of 30-100μm is used to control the discharge current of 0.1-5A, the pulse width of 1-100μs, the pulse interval of 1-200μs, and the discharge gap of 2-10μm to form a three-dimensional barbed structure, thus obtaining the second needle body.

[0061] Specifically, the first needle body is clamped on a multi-axis rotary fixture, preferably using a tungsten or copper-tungsten tool electrode with a diameter of 30-100 μm. Pulsed discharge is applied to the needle body in deionized water, kerosene working fluid, or other EDM working fluid. The multi-axis rotary fixture then etches the metal needle body layer by layer, employing a combination of roughing and finishing to form a three-dimensional barbed structure. Typical discharge parameters are: current 0.1-5 A, pulse width 1-100 μs, pulse interval 1-200 μs, and discharge gap 2-10 μm. This process achieves a machining accuracy controllable within ±5-10 μm and a surface roughness Ra of 0.2-0.5 μm, demonstrating high precision. When titanium alloy is used as the substrate, a micro-EDM process is preferred. Roughing is used to quickly remove localized material from the needle body, while finishing is used to trim the barb edges, reduce the recast layer thickness, and improve surface roughness. Afterwards, ultrasonic cleaning, electrochemical polishing, plasma cleaning, or a combination thereof are performed to remove the recast layer, machining debris, and microburrs.

[0062] Subsequently, a coating is prepared on the surface of the second needle body, comprising at least one of elemental gold, elemental platinum, and a platinum-iridium alloy. By controlling the coating material, thickness, and uniformity, the electrode-tissue interface impedance of the final electrode is controlled at 10 kΩ·cm. 2 The following describes the barbed microneedle electrode that can be used for electrophysiological signal acquisition.

[0063] In one embodiment, step S3, preparing a coating on the surface of the second needle body, specifically includes the following steps: A transition layer is sputtered onto the surface of the second needle body to obtain an electrode intermediate. The transition layer is made of Ti or Cr and has a thickness of 10-40 nm. Magnetron sputtering was employed, with the electrode intermediate placed at a first tilt angle on the sputtering disk and continuously rotated. Under an argon (Ar) atmosphere, the background vacuum was controlled to be <5 × 10⁻⁶. -4 The working pressure is 0.3-0.8 Pa, the sputtering power is 50-150 W, the deposition rate is 0.2-0.5 nm / s, the first tilt angle is 40-60°, the coating is elemental platinum or platinum-iridium alloy, and the sputtering target is Pt or Pt-Ir alloy.

[0064] Specifically, to improve the adhesion between the platinum elemental or platinum-iridium alloy coating and the surface of the second needle, a transition layer is sputtered onto the surface of the second needle before the formal sputtering coating. The transition layer is preferably made of titanium (Ti) or chromium (Cr), and its thickness is controlled between 10-40 nm, preferably 20 nm. The transition layer is sputtered using a DC or RF magnetron sputtering system. The second needle is placed at a 40-60° angle in the sputtering disk, and sputtering is performed while continuously rotating to ensure good film coverage on the needle surface, especially on the sidewalls of the barbed structure.

[0065] As an intermediate layer, the transition layer effectively bridges the differences in lattice constants and thermal expansion coefficients between the coating and the needle substrate, reducing the risk of film cracking or peeling due to stress accumulation during sputtering. Simultaneously, the active metal elements (Ti or Cr) in the transition layer can form chemical bonds or diffusion layers with the substrate surface, enhancing the interfacial bonding strength between the coating and the needle, ensuring the coating is less prone to detachment during long-term repeated use and maintaining stable low interfacial impedance characteristics. Furthermore, the transition layer can also prevent the migration and diffusion of substrate elements to the surface of the noble metal coating, avoiding interfacial oxidation or contamination that could lead to electrode performance degradation and extend the electrode's lifespan.

[0066] Then, a DC or RF magnetron sputtering system is used, with pure platinum (Pt) targets or platinum-iridium (Pt-Ir) alloy targets as the sputtering targets. The electrode intermediate is placed on the sputtering disk at an angle of 40-60° and continuously rotated to improve the uniformity of the coating coverage on the barbed sidewalls. The sputtering process parameters include: a base vacuum level controlled to be less than 5 × 10⁻⁶. -4 The working gas is argon (Ar), with a working pressure of 0.3-0.8 Pa, a sputtering power of 50-150 W, and a deposition rate of 0.2-0.5 nm / s. By controlling the sputtering time, the target film thickness can be achieved to 100-300 nm. The resulting platinum or platinum-iridium alloy coating exhibits strong adhesion to the needle body and high hardness, making it suitable for applications requiring repeated use.

[0067] In another embodiment, step S3 involves preparing a coating on the surface of the second needle body, specifically including the following steps: The second needle body is subjected to a nickel underlayer treatment to obtain an electrode intermediate. The electrode intermediate was subjected to pulse electroplating in a pre-prepared plating solution, with the cathode current density controlled at 0.5-2 A / dm³. 2 With a pulse duty cycle of 10%-30%, a pulse frequency of 100-1000 Hz, a plating bath temperature of 45-55℃, and a pH value of 8-9, a coating with a thickness of 0.5-2 μm is prepared by mechanical stirring combined with cathode movement.

[0068] Furthermore, the coating is made of elemental gold, and the pre-prepared plating solution contains gold ions at a concentration of 8-12 g / L and sodium sulfite at a concentration of 80-120 g / L.

[0069] Specifically, before the formal gold plating, the second needle body undergoes a nickel (Ni) undercoat treatment to improve the adhesion between the subsequent gold plating layer and the needle body surface. The nickel undercoat can be achieved through conventional electroless nickel plating or electroplating processes, and the thickness of the nickel undercoat is adjusted according to the surface condition of the needle body and the adhesion requirements.

[0070] The second needle is immersed in a chemical nickel plating solution, where a nickel layer is spontaneously deposited on the catalytic surface by the reducing agent in the solution. The chemical nickel plating solution consists of: 20-30 g / L nickel sulfate, 20-30 g / L sodium hypophosphite (reducing agent), and 10-20 g / L sodium citrate (complexing agent), with the pH adjusted to 4.5-5.5 (using ammonia or sodium hydroxide). The plating solution temperature is controlled at 80-90℃, the deposition rate is 10-20 μm / h, and the processing time is 5-20 min. The resulting nickel underlayer has a uniform thickness, making it particularly suitable for needle surfaces with complex three-dimensional barbed structures.

[0071] Then, pulse electroplating is performed using a cyanide-free sulfite gold plating system. The pre-prepared plating solution includes gold ions, sodium sulfite, and a leveling agent and / or a wetting agent. The concentration of gold ions is 8-12 g / L, the concentration of sodium sulfite is 80-120 g / L, the concentration of leveling agent is 10-100 mg / L, and the concentration of wetting agent is 50-200 mg / L. The leveling agent is at least one of a nitrogen-containing heterocyclic compound or a thiourea derivative, and the wetting agent is at least one of a fatty alcohol polyoxyethylene ether or a sulfonate surfactant.

[0072] The performance of the plating bath and the quality of the plating layer can be improved by adding appropriate functional additives to a sulfite-based cyanide-free gold plating system. The additives include at least one of leveling agents and wetting agents.

[0073] The leveling agent is used to suppress tip growth effect, reduce metal ion enrichment in the microneedle tip and barb edge area, avoid excessive gold plating at the tip, and ensure uniform plating thickness across all parts of the barb microneedle surface. In one embodiment, the leveling agent can be a nitrogen-containing heterocyclic compound (e.g., pyridine, C5H5N; imidazole, C3H4N2) or a thiourea derivative (e.g., thiourea, CS(NH2)2; allyl thiourea, C4H8N2S), with a leveling agent concentration of 10-100 mg / L.

[0074] Wetting agents are used to reduce the surface tension of the plating solution, improve the wetting ability of the plating solution on the microneedles and barbed structures, promote the uniform spreading of the plating solution on the microneedle surface, and reduce defects such as incomplete plating or pinholes caused by the adhesion of microbubbles on the needle surface. In one embodiment, the wetting agent may be fatty alcohol polyoxyethylene ether (general formula RO-(CH2CH2O)). n -H, where R is C8-C 18 Alkyl groups (n = 5-20) or sulfonate surfactants (such as sodium dodecylbenzenesulfonate, C... 12 H 25 (C6H4SO3Na), the concentration of wetting agent added is 50-200 mg / L.

[0075] The parameters for pulse electroplating include: cathode current density 0.5-2 A / dm³.2 The pulse duty cycle is 10%-30%, the frequency is 100-1000 Hz, the plating bath temperature is 45-55℃, and the pH value is 8-9. During the electroplating process, mechanical stirring is used in conjunction with cathode movement to enhance plating bath convection, reduce concentration polarization, and improve coating uniformity.

[0076] By controlling the pulse electroplating time, the gold plating thickness can be achieved to 0.5-2 μm. The resulting gold plating has low resistivity, and the interface impedance can be reduced to 10 kΩ·cm at 1 kHz. 2 The process is simple and the cost is low, making it suitable for mass production.

[0077] After the coating preparation is completed, before applying the electrode, a polyimide microtube needs to be installed at its rear end (1 / 3 to 1 / 2 of the length of the needle body) to achieve non-destructive removal of the electrode.

[0078] Understandably, stainless steel or titanium alloy substrates possess excellent biocompatibility and mechanical strength, providing a stable physical carrier and a basic interface for tissue contact with the electrodes. After preparing a coating on the substrate surface, the high chemical stability and excellent conductivity of the coating significantly reduce the electrode-tissue interface impedance and minimize electrode polarization in the electrolyte environment. This is crucial for the high-fidelity acquisition of weak electrophysiological signals (especially microvolt-level EEG signals), ensuring high transmission fidelity from the signal source to the acquisition circuit.

[0079] The microneedle electrodes fabricated using the above-described method exhibit excellent quality and performance in dual-mode electrophysiological signal acquisition scenarios. The three-dimensional barb structure formed on the needle body through precision machining ensures reliable anchoring of the electrode to the dermis after transdermal transmission. Whether applied to back ECG acquisition or head EEG acquisition, the electrode maintains positional stability even under free movement of the target subject, preventing signal interruption due to electrode displacement or detachment. The uniform and dense coating formed through surface plating processes controls the electrode-tissue interface impedance to an extremely low level, meeting the fidelity requirements for millivolt-level amplitude acquisition of ECG signals while ensuring high signal-to-noise ratio acquisition of microvolt-level weak electrical activity in EEG signals. This allows the same electrode structure to simultaneously adapt to the different sensitivity requirements of dual-mode signals. Furthermore, the nickel underlayer treatment and plating process ensure excellent interfacial adhesion between the coating and the needle substrate, preventing cracking and peeling of the coating during long-term use or repeated loading and unloading, maintaining long-term stability of the interfacial impedance, and providing a reliable hardware foundation for long-term continuous synchronous telemetry.

[0080] It should be noted that the microneedle electrode sub-assembly adopts a modular design concept. Its core acquisition unit (barbed microneedle electrode) maintains a unified standard in material selection, structural design and manufacturing process to ensure the consistency of different electrophysiological signal acquisition. The integrated configuration of the microneedle electrode sub-assembly with the fixation sub-assembly and wearable sub-assembly is customized according to the specific monitoring purpose (ECG, EEG or EMG, etc.) and the species and body size differences of the target object, so as to meet the personalized needs of different application scenarios while ensuring the universality of the core components.

[0081] Furthermore, such as Figure 4 , Figure 5 As shown in (a) to (d), the fixing sub-assembly includes a first fixing group 21 and a second fixing group 22; The first fixing group 21 includes a fixing patch (not shown in the figure), a fixing gel 211 and a base 212. The fixing patch adheres and fixes the first microneedle group to the first preset point of the target object. The fixing gel is used to fix the electrodes of the first microneedle group to the base. The first microneedle group is arranged in a single-lead or multi-lead manner. The second fixing group 22 includes a base 221 and a magnetic suction group 222. The magnetic suction group 222 is used to magnetically fix the base 221. The base 221 is provided with arrayed positioning holes 2211. The electrodes of the second microneedle group are embedded in the positioning holes 2211. The base 221 is configured at the second preset point of the target object.

[0082] The fixation sub-assembly adopts a differentiated dual-group fixation strategy. It is designed with a first fixation group and a second fixation group to address the differences in the characteristics of different electrophysiological signal acquisition sites, signal amplitude characteristics, and electrode arrangement, thereby achieving precise adaptation and stable fixation for different preset points.

[0083] In one embodiment, the fixation patch in the first fixation group uses a medical-grade adhesive material (such as a pressure-sensitive adhesive patch) to adhere and fix the first microneedle group to the skin surface of the target object. The first preset point can correspond to the anatomical location for ECG acquisition. Because the ECG signal amplitude is relatively large (millivolt level) and the heart position is relatively fixed, the first microneedle group is arranged in a single-lead or multi-lead configuration, and stable signal acquisition can be achieved through the fixation patch. The single-lead configuration uses two electrodes to form one ECG signal pathway for baseline heart rate monitoring; the multi-lead configuration uses multiple electrodes to form a vector loop for standard ECG vector analysis.

[0084] The fixation gel, which can be a conductive polymer gel or a self-healing hydrogel, is filled between the base and the skin contact surface of the target object. It secures the electrodes of the first microneedle assembly to the base, fills the gaps left after the microneedle electrodes penetrate, reduces contact resistance, and provides insulation. The base supports the electrode array of the first microneedle assembly, providing mechanical support and structural rigidity. The base can be made of medical-grade resin or flexible polymer material and custom-molded using 3D printing to fit the skin contact surface of the target object.

[0085] The second preset point is usually located on the head of the target subject and can correspond to the anatomical location of the EEG acquisition (such as the prefrontal lobe, parietal lobe, hippocampus, etc.). Because the amplitude of the EEG signal is very weak (microvolt level) and the accuracy of the electrode position is extremely high, the second fixed group adopts a rigid positioning strategy to ensure that the spatial position of the multi-channel EEG electrodes is precise and controllable.

[0086] In one embodiment, the base in the second fixation group is customized using 3D printing technology, based on a three-dimensional model of the target object's skull (obtained through CT scans or a standard skull model library), and made of lightweight, high-strength materials with a thickness of 1-2 mm.

[0087] The 3D printing technologies used include two strategies: Fused Deposition Modeling (FDM) and photopolymerization. The FDM strategy uses carbon fiber reinforced PLA or PETG filaments, with a layer height of 0.1-0.2 mm, a nozzle diameter of 0.2-0.4 mm, a printing speed of 30-50 mm / s, and a heated bed temperature of 50-70℃. Post-printing processing includes removing supports and sanding off support marks. The photopolymerization strategy uses biocompatible photosensitive resin, with a layer thickness of 0.025-0.05 mm, an exposure time of 2-8 s / layer, and a printing equipment pixel size ≤ 50 μm and Z-axis repeatability within ±0.01 mm. Post-printing processing includes isopropyl alcohol cleaning of residual resin, UV secondary curing to improve strength, and removal and sanding off support marks.

[0088] The substrate features an array of positioning holes precisely arranged according to a pre-defined electrode array topology. The electrodes of the second microneedle group are embedded in these positioning holes, with the hole diameter slightly larger than the electrode's outer diameter to ensure mechanical stability and prevent lateral displacement after insertion. The positioning holes are designed to accommodate the needle portion of the electrode, facilitating wire routing along internal slots and subsequent lead-out. The electrodes penetrate the substrate through the positioning holes, with a 1-3 mm needle tip protruding inwards (towards the target scalp) for close contact / insertion into the scalp to acquire EEG signals. Electrodes are positioned in key brain regions such as the prefrontal cortex, parietal lobe, and hippocampus, supporting 1–8 channel EEG signal acquisition and capturing core EEG characteristics such as δ / θ / α / β / γ band power and epileptic seizure waves.

[0089] The magnetic attachment can consist of several (e.g., 3-4) small neodymium iron boron magnets fixed to the skull of the target subject with bone cement, and several small neodymium iron boron magnets embedded in the inner side of the base. The base after embedding the electrodes and magnets forms an "electrode cap" that matches the curvature of the skull of the target subject to achieve conformal attachment, ensuring a tight fit without lifting or gaps.

[0090] The fixation sub-assembly resolves the technical contradiction in electrophysiological monitoring between "flexible adhesive fixation for ECG and rigid, precise positioning for EEG." The first fixation group employs a flexible fixation method to adapt to the elastic deformation of the skin and the movement needs of the target subject; the second fixation group uses a rigid fixation method of "positioning holes + conformal base + magnetic attraction" to ensure multi-channel precise positioning of microvolt-level EEG signals from the head. The integration of the magnetic attraction structure enables unified and convenient assembly and disassembly operations, providing a stable physical basis for synchronous acquisition of dual-mode signals.

[0091] Furthermore, such as Figure 6 As shown, the wearable sub-component 3 includes a main wearable structure, which is detachably connected to the microneedle electrode sub-component, the fixing sub-component, and the signal acquisition component. The main wearable structure includes a first part 31 and a second part 32. The first part 31 is adapted to a first preset point of the target object, and the second part 32 is adapted to a second preset point of the target object. The first component 31 is used to integrate the first microneedle group, the first fixing group, and the signal acquisition component; The second component 32 is used to integrate the second microneedle group, the second fixing group, and the signal acquisition component.

[0092] The wearable sub-component provides the functions of "integration" and "overall wearable fixation", integrating components such as electrodes, fixation structure, and acquisition circuits in the signal acquisition component into a carrier that can be worn on the target object as a whole. This solves the problem of how to fix the entire device on the target object without affecting its free movement, and realizes the lightweight and modularization of the device.

[0093] In one embodiment, the wearable sub-assembly includes a main wearable structure. The main wearable structure is detachably connected to the microneedle electrode sub-assembly, the fixing sub-assembly, and the signal acquisition assembly using magnetic, snap-fit, or plug-in connections, facilitating quick replacement of components according to experimental needs and enabling modular maintenance of the device.

[0094] In a preferred embodiment, the main wearable structure includes a first part and a second part, forming a split wearable structure.

[0095] The first component adapts to the first preset point on the target object, and has a vest-like or shoulder strap structure, spanning the back of the target object and extending to both sides of the sternum. The first component carries the first micro-needle group fixed by the first fixing group, and also integrates the micro PCB circuit board and power supply battery in the signal acquisition component, forming an integrated signal acquisition and processing unit.

[0096] For example, the first component is a 3D-printed support structure or elastic fabric. Flexible polymer materials (such as medical-grade silicone or thermoplastic polyurethane, TPU) are used during 3D printing to conform to the contours of the chest and abdomen and allow for expansion during breathing. A micro-PCB circuit board and a power supply battery are encapsulated within the first component. The support structure or elastic fabric is then secured to the target object's back using elastic Velcro, straps, or similar methods. The connecting wires for the electrodes are flexible flat cables (FPC / FFC), routed along the skin surface with redundant length to accommodate the target object's stretching and curling movements.

[0097] The second component adapts to the second preset points on the target object, taking the form of a headgear or helmet, covering the top and back of the head. The second component docks with the base of the second fixation group, stably fixing the second microneedle group to the head. The second component also integrates the acquisition circuit module corresponding to the EEG channel in the signal acquisition assembly, forming a complete head-wearable unit. The second component is made of rigid, lightweight resin (such as photosensitive resin or polylactic acid PLA), custom-manufactured via 3D printing, with its inner surface contour matching the skull surface to ensure uniform contact between the electrodes and the scalp.

[0098] The first and second components can each integrate a wireless transmission module to independently transmit monitoring information, eliminating the constraints of wires.

[0099] With its separate design consisting of the first and second components, the ECG acquisition unit and the EEG acquisition unit can be installed and debugged independently, and can also be flexibly replaced according to experimental needs, which significantly improves the adaptability and ease of maintenance of the device.

[0100] Furthermore, the signal acquisition component includes a first acquisition module, a second acquisition module, and a clock trigger module; Synchronously acquire electrophysiological signals of the target object at different preset locations, specifically including: At the beginning of each sampling period, the clock trigger module simultaneously sends pulse signals to the first acquisition module and the second acquisition module; In response to the pulse signal, the first acquisition module and the second acquisition module start sampling at the same clock edge, latch the electrophysiological signal in real time, and attach the timestamp generated by the clock trigger module to give the first electrophysiological signal and the second electrophysiological signal.

[0101] The signal acquisition component adopts a parallel sampling architecture driven by a unified clock source, and achieves synchronous acquisition of the first electrophysiological signal and the second electrophysiological signal through strict timing control at the hardware level.

[0102] The first acquisition module is used to acquire electrophysiological signals at a first preset point to form a first electrophysiological signal. The second acquisition module is used to acquire electrophysiological signals at a second preset point to form a second electrophysiological signal.

[0103] In one specific implementation, the first and second acquisition modules utilize two 24-bit, 8-channel analog front-end chips, providing a total of 16 channels of parallel acquisition capability. Channel allocation is flexibly configurable: for example, channels 1-5 are used for ECG acquisition (single-ended or differential connection), and the remaining channels are used for EEG acquisition (differential connection, supporting bipolar or unipolar reference leads). Each chip integrates a programmable gain amplifier (adjustable from 1 to 12 times), a high input impedance buffer (>1 GΩ), and a 24-bit Δ-Σ analog-to-digital converter. The sampling rate supports software configuration from 250 sps to 32 ksps, with input reference integral noise as low as 0.47 μVrms in the 0.5–100 Hz frequency band. The chip incorporates a right leg drive circuit and a 50 / 60 Hz power frequency notch filter with a common-mode rejection ratio ≥ 110 dB, effectively suppressing power frequency and EMG interference.

[0104] The clock trigger module is responsible for generating a high-precision synchronization trigger signal, and also handles data buffering and packet assembly, hardware timestamp generation, power management, and acquisition parameter adjustment. In one specific implementation, the clock trigger module can be implemented using a low-power microcontroller or a programmable logic device (FPGA / CPLD), integrating a high-precision timer (resolution better than 0.1 ms) to provide a unified synchronization clock signal for the first and second acquisition modules, ensuring that the acquisition start times of the two signals are completely consistent, thus solving the core problem of multi-modal signal synchronization loss at the hardware level. For example, a low-power dual-core microcontroller with a maximum clock frequency of 128 MHz and support for dynamic frequency adjustment can be used. Its internal resources include: 256 KB SRAM, an external 128 Mbit QSPI PSRAM for large data buffering; an integrated Bluetooth 5.4 low-power transceiver (supporting 2M PHY and long-range mode) for low-channel-count short-range transmission; multiple SPI / I2C interfaces for reading analog front-end chips and controlling external wireless transceivers; a high-precision timer (resolution better than 0.1 ms) for generating sampling start timestamps for each data packet; and a power management section integrating a programmable multi-output power management unit that can independently shut down unused sensor channels and wireless modules, with standby power consumption controlled in the microamp level (typical value <5 μA).

[0105] At the start of each sampling period, the clock trigger module simultaneously sends pulse signals to both the first and second acquisition modules via a dedicated hardware trigger line. The pulse signals are narrow-pulse-width digital pulses connected to the sampling trigger pins (such as CONVST or START pins) in both acquisition modules.

[0106] In response to a pulse signal, at the same clock edge of the pulse signal (such as the same rising edge or the same falling edge), the sampling circuits inside the first acquisition module and the second acquisition module close simultaneously, capturing the signal amplitude at the same physical moment, and latching the sampled value into their respective data output registers in real time.

[0107] Meanwhile, the clock trigger module generates a timestamp in each sampling period, specifically generated by its internal 64-bit running counter. The timestamp is then appended to the header of the data frames of the electrophysiological signals acquired by the first and second acquisition modules in the corresponding sampling periods, forming time-stamped first and second electrophysiological signals, i.e., dual-mode electrophysiological signals.

[0108] By adopting a parallel acquisition and unified clock triggering architecture, the synchronous acquisition of dual-mode electrophysiological signals is achieved at the hardware level, with the synchronization error controlled to less than 1ms, meeting the high-precision requirements of signal timing consistency in heart-brain correlation research.

[0109] In one extended embodiment, the signal acquisition component further includes a wireless transmission module, which works in conjunction with the first acquisition module, the second acquisition module, and the clock trigger module to form a low-power multi-channel electrophysiological signal synchronous acquisition circuit.

[0110] The wireless transmission unit is used to transmit preprocessed monitoring information to the signal analysis component in real time, and to receive control commands (such as sampling rate switching and channel enable adjustment) issued by the signal analysis component. Two optional transmission schemes are provided based on the number of channels and synchronization accuracy requirements.

[0111] When the number of acquisition channels is ≤4 and the sampling rate is 1-2 kHz, data transmission is directly performed using the Bluetooth Low Energy transceiver integrated within the clock trigger module, operating in 2M PHY mode. This scheme employs a block timestamp strategy: for every M samples acquired (M being 50-100), the clock trigger module reads a high-precision timer to obtain a 64-bit start sampling time, packages this data along with all sample data within the block into a single data frame, and transmits it via Bluetooth. After parsing the frame header, the receiving end can accurately reconstruct the sampling time of each sample based on the start timestamp and a fixed sampling interval, with a time synchronization error less than one sampling period (e.g., error ≤0.5 ms at a 2kHz sampling rate). The overall power consumption of the circuit using this scheme can be controlled to within 50 mW.

[0112] When the number of acquisition channels is ≥8 and the sampling rate is ≥2 kHz, an external Sub-1GHz wireless transceiver (supporting 433 / 868 / 915 MHz ISM bands) is used. The clock trigger module communicates with the transceiver via an SPI interface, using a custom point-to-point protocol, with an air rate configured at 250-500 kbps. In this scheme, each data packet simultaneously carries a continuous block of samples from multiple channels (such as ECG and EEG), and a 64-bit hardware start timestamp is appended to the packet header. Since Sub-1GHz transmission has no connection gap, the data packet transmission interval is precisely controlled by the clock trigger module, and the receiver can achieve microsecond-level synchronization alignment based on the timestamp. Taking 8 channels and a 2kHz sampling rate as an example, the raw data rate is approximately 384 kbps. With a 500 kbps air rate, the transmit power consumption is approximately 30-60 mW, and the overall circuit power consumption can still be maintained below 100 mW.

[0113] By designing a low-power, multi-channel ECG and cerebral electrophysiological synchronous acquisition circuit triggered by a unified clock, hardware synchronization of dual-mode electrophysiological signals is achieved, solving the core problem of lack of synchronization of multimodal signals. The signal acquisition component has low overall power consumption, enabling long-term continuous monitoring and resolving the contradiction between multi-channel monitoring and long-term power consumption.

[0114] Furthermore, the signal acquisition component also includes a dual-mode preprocessing module; Preprocessing of electrophysiological signals specifically includes: Obtain control commands and determine the preprocessing mode; Based on the preprocessing mode, perform either the first preprocessing or the second preprocessing; The first preprocessing step specifically includes the following steps: The first and second electrophysiological signals are filtered and compressed respectively to provide the first monitoring information; The second preprocessing step includes the following steps: The first and second electrophysiological signals were filtered respectively to eliminate motion artifacts and extract physiological features, thus providing the second monitoring information.

[0115] It is understandable that the monitoring information refers to the data packet output after preprocessing by the signal acquisition component, including the first monitoring information and the second monitoring information. The first monitoring information is the raw waveform data after filtering and compression; the second monitoring information is the quantified physiological parameters after filtering, artifact removal, and feature extraction.

[0116] The dual-mode preprocessing module is used to preprocess electrophysiological signals and provide monitoring information. It receives electrophysiological signals from the first and second acquisition modules, determines the applicable preprocessing mode based on acquired control commands, and executes the corresponding signal processing flow based on the determined preprocessing mode. The first preprocessing performs filtering and compression on the first and second electrophysiological signals respectively, providing first monitoring information, suitable for scenarios requiring long-term continuous monitoring and with strict limitations on data transmission bandwidth. The second preprocessing performs filtering, motion artifact removal, and physiological feature extraction on the first and second electrophysiological signals respectively, providing second monitoring information, suitable for scenarios where the target object is in a free-moving state and the signal is significantly affected by motion interference. The two preprocessing modes can be flexibly switched according to experimental needs, or the signal analysis component can automatically issue control commands for remote switching based on signal quality assessment results.

[0117] In the first preprocessing mode, the dual-mode preprocessing module first performs digital filtering on the first and second electrophysiological signals to remove high-frequency noise and power frequency interference. Then, the filtered signals are compressed to reduce the data volume for efficient transmission by the low-power wireless transmission module. The compressed signal is packaged together with the corresponding timestamp information to form the first monitoring information, which is then sent to the signal analysis component for decompression and in-depth analysis.

[0118] In the second preprocessing mode, the dual-mode preprocessing module first performs digital filtering on the first and second electrophysiological signals respectively; then, motion artifact removal is performed to suppress motion interference introduced by the target object in a free-moving state; finally, physiological feature extraction is performed to extract key physiological feature parameters from the artifact-removed signal. The processed feature parameters are packaged together with the corresponding timestamp information to form the second monitoring information.

[0119] When the target object is in a free-moving state, motion artifacts mainly come from electrode displacement (scratching, running, and head shaking cause micro-movements at the contact point between the microneedle electrode and the skin), baseline drift (muscle contractions such as chewing and breathing cause soft tissue deformation, leading to electrode displacement), and high-frequency noise (sudden electromyographic interference caused by violent activities such as convulsions during epileptic seizures). These artifacts manifest as amplitude jumps in the time domain and overlap with physiological signals in the frequency domain, making them difficult to remove effectively using traditional filtering methods.

[0120] Further, motion artifact removal is performed, specifically including the following steps: The system performs time-frequency domain feature analysis using a pre-defined neural network model, identifies motion artifact segments and non-motion artifact segments, and generates mask values ​​for motion artifact segments. Based on the mask value, an adaptive elimination strategy is matched to correct the motion artifact clips; A sliding window overlap fusion mechanism is used to stitch non-motion artifact segments with corrected motion artifact segments to complete motion artifact elimination.

[0121] In one implementation, motion artifact removal is achieved using a lightweight neural network model. Specifically, a dual-mode preprocessing module performs joint time-frequency feature analysis on the filtered electrophysiological signal using a pre-defined neural network model. The neural network model employs a lightweight convolutional neural network trained using knowledge distillation technology. The input is raw ECG or EEG segments (window length 1-5 seconds, sampling rate 1-2 kHz), and the model output is a binary mask used to label valid signal segments (non-motion artifact segments) and motion artifact segments. This model can achieve a single inference time of less than 30 ms on an embedded processor, enabling real-time labeling and filtering of artifact segments, and improving the signal-to-noise ratio of the processed signal by 10-20 dB.

[0122] Based on the time-frequency domain feature analysis results, the dual-mode preprocessing module identifies motion artifact segments and non-motion artifact segments, and generates mask values ​​corresponding to each segment. The mask values ​​are used to quantify the degree to which a signal segment is contaminated by motion artifacts.

[0123] Next, the dual-mode preprocessing module performs hierarchical correction on motion artifact segments based on a mask value matching adaptive elimination strategy. Specifically, the adaptive elimination strategy includes: for lightly contaminated segments, adaptive notch filtering is used for suppression; for moderately contaminated segments, weighted fusion or interpolation reconstruction is used for repair; and for heavily contaminated segments, waveform replacement or generative reconstruction is used for recovery.

[0124] After completing the correction of each level of artifact segments, the dual-mode preprocessing module uses a cosine window weighting method to smoothly transition the signal segments of adjacent processing windows, eliminate boundary discontinuities, and obtain a continuous and seamless artifact-removed signal stream.

[0125] After motion artifact removal is completed, the dual-mode preprocessing module extracts physiological features from the processed signal. The extracted physiological features include time-domain morphological features (such as R-wave peak value, RR interval, and QRS width) and frequency-domain energy features (such as power and spectral entropy in each frequency band), which are used by the signal analysis component for in-depth analysis.

[0126] This invention employs a dual strategy of hardware fixation optimization and algorithm elimination to achieve precise elimination of motion artifacts: by reducing electrode displacement caused by mouse and rat movements through a lightweight fixation structure, the generation of motion artifacts is reduced from the source; the lightweight knowledge distillation machine learning model is equipped to perform feature recognition on the collected electrophysiological signals, accurately distinguish between effective signals and motion artifact signals, achieve rapid artifact elimination, and improve the signal-to-noise ratio by ≥30dB.

[0127] It should be noted that the dual-mode preprocessing module's functionality can be implemented by a separate edge computing processor, forming an intelligent acquisition mode. The edge computing processor employs a multi-core microcontroller architecture, with a built-in neural network accelerator and digital signal processor. It communicates with the clock trigger module via a high-speed interface, performing real-time preprocessing on the raw signal locally (such as filtering, feature extraction, and emergency event precursor pattern recognition). Only the warning tag, compressed features, or key segments are transmitted to the remote end via a low-power wireless transmission module, significantly reducing wireless transmission power consumption. This mode is suitable for applications requiring real-time warnings of events such as epileptic seizures and arrhythmias, and event records can be retained locally even if the wireless connection is temporarily interrupted.

[0128] Furthermore, the signal analysis component includes an information recognition module and a feature analysis module; The information identification module identifies and processes the monitoring information to form information to be analyzed; The feature analysis module performs feature analysis on the information to be analyzed and provides the telemetry results of the dual-mode electrophysiological signal; The monitoring information is identified and processed to form information to be analyzed, which specifically includes the following steps: The first monitoring information is decompressed, motion artifacts are eliminated, and the first spatiotemporal alignment is performed. Then, feature extraction is performed to form the information to be analyzed; and / or, The second monitoring information is spatiotemporally aligned to form the information to be analyzed.

[0129] Specifically, the information identification module calls the corresponding processing path based on the type of monitoring information (first monitoring information or second monitoring information).

[0130] For the first monitoring information, the information identification module sequentially performs decompression, motion artifact removal, first spatiotemporal alignment, and feature extraction to form the information to be analyzed. Specifically, the decompression operation restores the compressed signal data from the first preprocessing; motion artifact removal uses the same neural network model and adaptive elimination strategy as in the second preprocessing to remove motion interference; the first spatiotemporal alignment performs precise time alignment of the dual-mode electrophysiological signals at the signal level; and feature extraction extracts time-domain and frequency-domain feature parameters from the aligned signal.

[0131] For the second monitoring information, the information recognition module performs a second spatiotemporal alignment to form the information to be analyzed. The second spatiotemporal alignment performs nonlinear time alignment on the dual-mode electrophysiological signals at the feature level, which is suitable for signals that have completed front-end artifact elimination and feature extraction, in order to adapt to the local rate differences and nonlinear time distortions that exist in the physiological conduction process of ECG and EEG features.

[0132] Furthermore, the first spatiotemporal alignment specifically includes the following steps: Based on the sampling window length, the first monitoring information is segmented and processed to give signal segment pairs; Perform cross-correlation analysis on all signal segment pairs and give the first cross-correlation coefficient; Remove signal segment pairs whose first cross-correlation coefficient is lower than a first preset threshold, and give a coarsely registered signal segment group; For all signal segment pairs in the coarsely registered signal segment group, interpolation upsampling is performed, followed by cross-correlation analysis to give the second cross-correlation coefficient; Retain signal segment pairs whose second cross-correlation coefficient is higher than the second preset threshold, and give a finely registered signal segment group; For all signal segment pairs in the finely registered signal segment group, extract the first feature point and the second feature point; Calculate the time difference between each first feature point and its corresponding second feature point, and then average the results to obtain the first translation amount; Based on the first translation amount, the signal segment is time-axis shifted using an interpolation algorithm to give aligned electrophysiological signal pairs.

[0133] Understandably, although hardware-level acquisition synchronization is achieved through a unified clock trigger, minor timing offsets (mainly packet-level latency jitter) may still be introduced during wireless transmission, data packetization, and reception. The first spatiotemporal alignment further eliminates the residual time offsets introduced by the above-mentioned processes through software-level post-calibration.

[0134] Specifically, based on a preset sampling window length, the first electrophysiological signal (ECG signal) and the second electrophysiological signal (EEG signal) in the first monitoring information are segmented and processed to give several signal segment pairs. Each signal segment pair contains ECG and EEG segments within the same time window.

[0135] Next, cross-correlation analysis was performed on all signal segment pairs using synchronization markers in ECG R waves and EEG to calculate the first cross-correlation coefficient between each segment pair and estimate the residual clock drift. Signal segment pairs with a first cross-correlation coefficient lower than a first preset threshold (indicating insufficient correlation or potential severe interference) were removed, and the remaining signal segment pairs formed a coarsely registered signal segment group.

[0136] Next, all signal segment pairs in the coarse registration signal segment group are interpolated and upsampled to improve temporal resolution, followed by another cross-correlation analysis to obtain a second cross-correlation coefficient. Signal segment pairs with a second cross-correlation coefficient higher than a second preset threshold are retained to form the fine registration signal segment group.

[0137] Then, feature point matching is performed. That is, for all signal segment pairs in the finely registered signal segment group, the first feature point (corresponding to ECG signal feature points, such as R-wave peak points) and the second feature point (corresponding to EEG signal feature points, such as synchronization marker points) are extracted.

[0138] Then, the time difference between each first feature point and its corresponding second feature point is calculated and averaged to obtain the first translation. The first translation represents the global time offset between the electrocardiogram (ECG) signal and the electroencephalogram (EEG) signal.

[0139] Finally, based on the first translation, the signal segments are translated along the time axis using an interpolation algorithm (such as cubic spline interpolation) to adjust the time axes of the two signals to the same reference, resulting in a time-aligned ECG and EEG signal pair, which is then sent to the feature analysis module as the information to be analyzed.

[0140] Furthermore, such as Figure 7 As shown, the second spatiotemporal alignment specifically includes the following steps: Determine the first feature point sequence and the second feature point sequence in the second monitoring information; Electrophysiological events are identified from the first feature point sequence, and multiple electrophysiological event windows are provided. Based on each electrophysiological event window, the first feature point sequence and the second feature point sequence are divided into segments, and the first feature point sub-sequence and the second feature point sequence corresponding to each electrophysiological event window are given. Within each electrophysiological event window, a corresponding electrophysiological coupling weighted cost matrix is ​​constructed based on the first feature point subsequence and the second feature point subsequence. Based on the electrophysiological coupling weighted cost matrix, a dynamic time warping algorithm is used to search for the optimal alignment path within the corresponding electrophysiological event window; Based on the optimal alignment path within each electrophysiological event window, the corresponding local time offset sequence is determined, bidirectional coupling consistency verification is performed, and the verified local time offset sequence is given. Based on the validated local time offset sequence, the second feature point subsequence within each electrophysiological event window is reconstructed using timestamps, and aligned feature point pairs are given.

[0141] In a specific example, the feature point sequence is first determined, that is, the feature point sequence corresponding to the first electrophysiological signal and the feature point sequence corresponding to the second electrophysiological signal in the second monitoring information are determined. For example, the first feature point sequence consists of discrete feature points such as the peak time of the R wave detected and marked in the electrocardiogram signal, and each feature point is accompanied by an original timestamp generated by the clock trigger module; the second feature point sequence consists of feature points such as the time stamps corresponding to the power extrema of the δ, θ, α, β, and γ frequency bands extracted from the electroencephalogram signal, and the onset time of the epileptic spike burst, and also has its own original timestamp.

[0142] Next, based on the inherent periodicity of the first electrophysiological signal, electrophysiological event identification is performed on the first feature point sequence. Here, an electrophysiological event refers to a characteristic event in the electrophysiological signal that has clear physiological significance and can be detected and identified, such as the peak value of the R wave in ECG or the burst of spike waves in EEG. For example, using adjacent R wave peak times as boundaries, the continuous ECG feature sequence is divided into multiple RR interval event windows. Each event window corresponds to a complete cardiac cycle, and its duration changes dynamically with the real-time heart rate of the target object.

[0143] Then, based on each electrophysiological event window, the first feature point sequence and the second feature point sequence are segmented, and the first feature point subsequence and the second feature point subsequence corresponding to each electrophysiological event window are given. The truncation range of the second feature point subsequence is determined by the time boundary of the corresponding electrophysiological event window, that is, the second feature points falling within this time range are selected.

[0144] Subsequently, an electrophysiological coupling weighted cost matrix is ​​constructed for each electrophysiological event window. The rows of the matrix correspond to the feature points (i.e., ECG feature points) in the first feature point subsequence within the electrophysiological event window, and the columns correspond to the feature points (i.e., EEG feature points) in the second feature point subsequence. The numerical value of each element in the matrix represents the matching cost between the corresponding pair of feature points, which is jointly determined based on the physiological correlation strength and temporal distance between the ECG and EEG feature points. For example, feature point pairs with known strong physiological coupling relationships, such as the ECG R-wave peak and the EEG gamma-band power burst point, are assigned lower cost values; feature point pairs with weaker or unknown correlations are assigned higher cost values. The temporal distance weight is normalized according to the typical range of ECG and EEG electrophysiological conduction delays. Therefore, elements with low matching costs represent feature point pairs at that location that are more likely to have a true physiological coupling relationship. This weighting method allows the dynamic time-warped search process to prioritize physiologically reasonable feature point matching relationships, avoiding physiological distortion caused by simply aligning based on numerical similarity.

[0145] Therefore, the Dynamic Time Warping (DTW) algorithm is used to search for the optimal alignment path within each electrophysiological event window.

[0146] Furthermore, based on the electrophysiological coupling weighted cost matrix, a dynamic time warping algorithm is used to search for the optimal alignment path within the corresponding electrophysiological event window, specifically including the following steps: Determine the search start and end boundaries corresponding to the electrophysiological event window; Based on the electrophysiological coupling weighted cost matrix, the initial cumulative cost of the search starting point is determined, and the corresponding initial cumulative cost matrix is ​​given. Within the electrophysiological event window, combining monotonicity and continuity constraints, a dynamic programming algorithm is used to recursively calculate the cumulative cost point by point and update the corresponding initial cumulative cost matrix, thus providing the cumulative cost matrix. Based on the cumulative cost matrix, the position of the matrix element with the minimum cumulative cost is selected as the path endpoint on the termination boundary; Starting from the end of the path, and combining the preceding direction information, we backtrack to the starting point of the search to obtain the optimal alignment path.

[0147] Specifically, for each electrophysiological event window, the search starting point is the pair of feature points formed by the first feature point of the first feature point subsequence and the first feature point of the second feature point subsequence within its time range; the last row and last column, determined by the dimension of the corresponding electrophysiological coupling weighted cost matrix, serve as the termination boundary. The position of the search starting point in the matrix is ​​denoted as (1,1), and the termination boundary includes all matrix elements in the last row (m,1)~(m,n) and the last column (1,n)~(m,n), where m and n are the lengths of the first and second feature point subsequences within the window, respectively. This dimension is consistent with the electrophysiological coupling weighted cost matrix and the cumulative cost matrix. The termination boundary limits the search range of the dynamic programming recursive calculation, preventing the recursive process from exceeding the matrix dimension corresponding to the electrophysiological event window, and restricting the candidate position of the path endpoint to the end region of the window.

[0148] The initial cumulative cost value of the search starting point (1,1) is set as its corresponding matching cost value; the cumulative cost matrix positions outside the search range are set to preset maxima (e.g., infinity or sufficiently large positive numbers) to prevent the dynamic programming recursion process from exceeding the constraint boundaries; the remaining positions are set to preset maxima (and gradually updated to the actual cumulative cost value in subsequent recursion processes). Thus, the initial cumulative cost matrix is ​​given. The cumulative cost value represents the cost of the path with the minimum total matching cost among all possible paths from the search starting point to the current matrix element (i.e., the current feature point pair), reflecting the optimal alignment cost from the beginning of the sequence to the current position. The smaller the value, the better the alignment path to that position.

[0149] Subsequently, within the electrophysiological event window, combining monotonicity and continuity constraints, and employing a preset step size mode, a dynamic programming algorithm is used to recursively calculate the cumulative cost point by point. Specifically, for each element (i,j) of the cumulative cost matrix, its cumulative cost is obtained by adding its corresponding matching cost to the minimum cumulative cost from the three preceding directions (directly above (i-1,j), directly to the left (i,j-1), and the upper left diagonal (i-1,j-1)). The recursive process follows monotonicity constraints (only extending to the right, down, or lower right, backtracking is not allowed) and continuity constraints (only moving one unit per step, skipping points is not allowed). The preset step size mode can adopt a symmetrical step size mode, that is, the proportional weights of diagonal stepping, horizontal stepping, and vertical stepping are equal, so that the ECG feature points and EEG feature points establish a correspondence at an approximately 1:1 rate in the alignment path, avoiding time axis distortion caused by excessive stretching in one direction.

[0150] While recursively calculating the cumulative cost point by point, the direction of the minimum cumulative cost source selected by each matrix element (directly above, directly to the left, or diagonally) is recorded as the precursor direction information for path backtracking.

[0151] After completing the recursive calculation, the cumulative cost matrix is ​​obtained. At the termination boundary (last row and last column), the cumulative cost of all boundary elements is traversed, and the matrix element with the smallest cumulative cost is selected as the path endpoint. This open endpoint strategy allows the optimal alignment path to choose the most reasonable alignment termination position at the end of the window based on the actual physiological coupling strength, rather than forcibly aligning to the last feature point of the subsequence, thus avoiding alignment distortion caused by hard pairing at the end.

[0152] Next, starting from the selected path endpoint, and combining the preceding direction information recorded during the recursion process, the system backtracks backward along the source direction with the minimum cumulative cost, gradually tracing back to the search starting point. The sequence of cumulative cost matrix elements passed during the backtracking process constitutes the optimal alignment path within the electrophysiological event window.

[0153] The above steps are performed sequentially for each electrophysiological event window to obtain the optimal alignment path for each window. The optimal alignment path represents the optimal temporal mapping relationship between the first feature point subsequence and the second feature point subsequence within the electrophysiological event window. Each matrix element on the path corresponds to a pair of matching feature points, and each pair of matching feature points corresponds to a temporal association of neural cardiac conduction, reflecting the true physiological delay of ECG and EEG signal interaction.

[0154] Next, local time offset sequences are determined from the optimal alignment path. Specifically, the time difference between each pair of matched feature points is calculated, resulting in a series of offset values ​​reflecting the actual neural conduction delay in the heart. Then, these offset values ​​undergo bidirectional coupling consistency verification. For example, local time offset sequences are determined based on the optimal alignment path in both the ECG→EEG and EEG→ECG directions, resulting in two sets of local time offset sequences. The offset sequences in these two directions are compared, and abnormal offset values ​​with differences exceeding a preset threshold are removed. This bidirectional verification mechanism utilizes the physiological prior of a bidirectional neural regulatory pathway between the heart and brain, ensuring that the retained offset values ​​reflect the true physiological conduction delay.

[0155] Finally, the verified local time offset sequence is traversed, and the matching index on the optimal alignment path within the electrophysiological event window is combined to re-mark the original timestamps of the EEG feature points with the corresponding ECG feature point timestamps plus the local delay value (i.e., offset value) specific to the pairing. For the time gaps between matching points, linear interpolation or cubic spline interpolation methods are used to obtain EEG feature points that are strictly aligned with the ECG feature point time grid while maintaining the continuity of physiological trends.

[0156] After the above processing, the ECG and EEG feature points that were originally asynchronous on the original timestamps are mapped onto a unified logical time axis, forming feature point pairs with precise timestamp correspondence. These pairs are then sent to the feature analysis module as information to be analyzed and can be used for ECG-EEG correlation analysis or risk stratification assessment, etc.

[0157] Furthermore, the signal analysis component is also used to perform feature analysis based on a preset physiological feature processing model, and to provide telemetry results of dual-mode electrophysiological signals.

[0158] Feature analysis is performed based on a pre-defined physiological feature processing model to provide telemetry results of dual-mode electrophysiological signals. The specific steps include the following: Extract the first and second physiological characteristics from the information to be analyzed; Based on the first and second physiological characteristics, a correlation analysis is performed to determine the physiological coupling and regulatory relationships between the first and second physiological characteristics, and the results of the correlation analysis are given. Based on the correlation analysis results, the telemetry results of the dual-mode electrophysiological signals are presented.

[0159] In one implementation, the feature analysis module calls the corresponding physiological feature processing model for a specific disease model, performs correlation analysis and risk assessment on electrocardiographic and electroencephalographic features, and outputs telemetry results including the cardio-brain coupling coefficient, physiological state assessment results, and risk warning information.

[0160] For example, for spontaneous hypertension models (such as SHR rats), the feature analysis module extracts electrocardiographic and electroencephalographic features, establishes a cardiac-brain electrophysiological correlation analysis model, and studies the brain-heart regulation mechanism of neurogenic hypertension.

[0161] Among them, cardiac electrophysiological characteristics include time-domain heart rate variability indicators (such as RMSSD, SDNN) and frequency-domain indicators (such as LF, HF, LF / HF ratio), QT interval, etc.; brain electrophysiological characteristics include the absolute and relative power of each frequency band (δ, θ, α, β, γ bands), with particular attention to cortical rhythms related to sympathetic activity (such as β / γ ratio).

[0162] The cardiac-encephalogram (CEG) correlation analysis employs sparse canonical correlation analysis or lightweight graph neural networks to construct lag cross-correlation functions between EEG power fluctuations and heart rate and heart rate variability indicators. It then calculates the degree of influence of EEG changes on heart rate and the regulatory effect of heart rate changes on subsequent EEG activity, forming cardiac-brain coupling characteristics to reveal the mutual regulatory relationship between autonomic nervous activity of the brain and cardiac function.

[0163] The aforementioned electrocardiographic, electroencephalographic, and cardioencephalometric characteristics are input into a pre-trained classifier model to assess whether the current target object is in a state of stress and the level of stress intensity. When the stress level score exceeds a preset threshold, an early warning is triggered.

[0164] For example, for hereditary epilepsy models (such as GAERS rats), the feature analysis module performs epileptic seizure detection and stratified assessment of sudden epileptic death risk (SUDEP).

[0165] Seizure detection employs a lightweight neural network model to process multi-channel EEG segments in real time, identifying ictal and non-ictal periods and marking the seizure onset time. Seizure wave characteristics include spikes, sharp waves, slow waves, and rhythmic high-frequency bursts.

[0166] High-risk arrhythmia identification involves real-time monitoring of abnormal RR intervals, QT interval prolongation, and conduction block in electrocardiogram (ECG) signals, identifying precursory features of malignant arrhythmias. A machine learning model is used to fuse ECG features with epileptic seizure states to determine whether epileptic seizures are associated with a risk of malignant arrhythmias.

[0167] The risk assessment model inputs the intensity of epileptic seizures from EEG, the degree of decrease in heart rate variability from ECG, QT interval prolongation, and optional postural data (such as atonic or post-tetanic falls) to output a sudden death risk level (low / medium / high). When a high-risk state persists for more than a preset time, a remote alarm is automatically sent.

[0168] The analysis results of the two physiological feature processing models can be output separately or displayed together as needed, providing standardized analysis tools and data support for the study of the mechanisms of neurogenic cardiovascular diseases and epilepsy and other heart-brain related diseases.

[0169] To address the issues of poor compatibility and low processing efficiency caused by the generalized design of existing monitoring equipment with specific pathological models, this invention achieves deep integration and application of electrodes, circuits, wearable systems, and pathological models. It develops a customized acquisition mode and analysis framework that can accurately capture characteristic electrophysiological signals. Furthermore, it achieves motion artifact elimination, spatiotemporal signal alignment, feature extraction, and risk warning through a lightweight machine learning model, thereby improving analysis and processing efficiency and providing direct and reliable data support for disease mechanism research.

[0170] Secondly, such as Figure 8 As shown, the present invention also provides a method for synchronous acquisition and telemetry of dual-mode electrophysiological signals, using the aforementioned synchronous acquisition and telemetry device for dual-mode electrophysiological signals, specifically including the following steps: The microneedle electrode sub-assembly is fixed to a preset point on the target object by a fixing sub-assembly, and the microneedle electrode sub-assembly, fixing sub-assembly, and signal acquisition sub-assembly are integrated and worn on the target object by a wearable sub-assembly; The signal acquisition component is activated, and the electrophysiological signals of the target object at different preset points are acquired synchronously through the electrode component. The electrophysiological signals are then preprocessed to provide monitoring information. The monitoring information is received and analyzed by the signal analysis component, and the telemetry results of the dual-mode electrophysiological signal are given.

[0171] The technical solution of this invention has the technical characteristics of being minimally invasive, lightweight, wireless, and synchronous. It solves the problems of existing equipment having single functions and poor adaptability, significantly improves the authenticity of data collection and the welfare of target subjects, provides standardized scientific research tools for basic research in the fields of neurogenic cardiovascular diseases and epilepsy, and promotes the industrialization and transformation of medical-engineering interdisciplinary technologies.

[0172] The synchronous acquisition and telemetry device of the present invention was configured on experimental mice (such as C57BL / 6) to perform synchronous acquisition and telemetry of electrocardiogram (ECG) and electroencephalogram (EEG) signals. The specific operation is as follows: The first microneedle assembly (single-lead dual-electrode) was fixed to the first preset point on the back of the mouse using the first fixation assembly. The fixation patch of the first fixation assembly adhered and fixed the first microneedle assembly to the skin on the back, completing the stable installation of the ECG acquisition electrode.

[0173] The second microneedle assembly (multi-channel probe electrode) is fixed to the base of the second fixation assembly. The base is a 3D-printed customized electrode cap that conformally matches the curvature of the mouse skull. Each electrode of the second microneedle assembly is embedded in the positioning holes arranged in an array on the base, and the second magnetic fixation structure magnetically fixes the base to the second preset point on the mouse head.

[0174] The first and second microneedle groups were connected to the signal acquisition component via wires. The signal acquisition component was integrated into the first and second parts (corresponding to the mouse's back) of the wearable sub-component, which were detachably connected and worn securely on the mouse. After electrode installation, the mouse could move freely without any obvious behavioral abnormalities.

[0175] The signal acquisition component is activated to begin acquisition. For electrocardiogram (ECG) signals, clear R waves can be acquired and core ECG indicators such as heart rate variability (HRV) and autonomic nervous system index can be calculated. For electroencephalogram (EEG) signals, parallel acquisition of at least four channels of EEG can be achieved, covering the frontal, temporal, parietal, and occipital lobes of the brain, to extract core EEG features such as power in the δ / θ / α / β / γ frequency bands.

[0176] Ultimately, synchronous acquisition of electrocardiogram (ECG) and electroencephalogram (EEG) signals was achieved. The ECG signal acquisition results are as follows: Figure 9 As shown, the EEG signal acquisition results are as follows: Figure 10 As shown.

[0177] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and modifications.

Claims

1. A synchronous acquisition and telemetry device for dual-mode electrophysiological signals, characterized in that, This includes electrode assemblies for signal connection, signal acquisition assemblies, and signal analysis assemblies; The electrode assembly includes a microneedle electrode sub-assembly, a fixation sub-assembly, and a wearable sub-assembly. The microneedle electrode sub-assembly consists of multiple microneedle groups, including a first microneedle group and a second microneedle group. The first microneedle group and the second microneedle group are used to collect electrophysiological signals at different preset points on the target object. The fixation sub-assembly is used to fix the microneedle electrode sub-assembly at the preset points. The wearable sub-component is used to integrate the microneedle electrode sub-component, the fixation sub-component, and the signal acquisition component, and is worn and fixed to the target object; The signal acquisition component is used to synchronously acquire electrophysiological signals of the target object at different preset points through the electrode assembly, and to preprocess the electrophysiological signals to provide monitoring information; The signal analysis component is used to receive and analyze monitoring information and provide telemetry results of dual-mode electrophysiological signals.

2. The synchronous acquisition and telemetry device for dual-mode electrophysiological signals as described in claim 1, characterized in that, The microneedle assembly includes electrodes, microtubes, and wires; The electrode has a barbed front end and a wire connection at the rear end. A microtube is fitted over the rear end of the electrode. The electrode substrate is stainless steel or titanium alloy, and the electrode surface is coated with a plating layer, which is at least one of elemental gold, elemental platinum, and platinum-iridium alloy. The microtube is made of an insulating polymer.

3. The synchronous acquisition and telemetry device for dual-mode electrophysiological signals as described in claim 2, characterized in that, The electrode is prepared by the following steps: S1, the electrode substrate is pre-cut to obtain the first needle body; S2, based on the type of electrode substrate, a corresponding processing technology is adapted to process a three-dimensional barb structure on the first needle body to obtain the second needle body; S3, based on the material of the coating, a corresponding coating process is applied to prepare a coating on the surface of the second needle body to obtain an electrode.

4. The synchronous acquisition and telemetry device for dual-mode electrophysiological signals as described in claim 3, characterized in that, In step S2, a three-dimensional barb structure is machined on the first needle body to obtain the second needle body, specifically including the following steps: The first needle body is ablated and cut using a femtosecond laser and a multi-axis rotating fixture. The femtosecond laser pulse width is controlled to be <500fs, the single pulse energy is 1-100μJ, and the repetition frequency is 10-100kHz. A three-dimensional barbed structure is formed to obtain the second needle body.

5. The synchronous acquisition and telemetry device for dual-mode electrophysiological signals as described in claim 3, characterized in that, In step S3, a coating is prepared on the surface of the second needle body, specifically including the following steps: The second needle body is subjected to a nickel underlayer treatment to obtain an electrode intermediate. In a pre-prepared plating solution, the electrode intermediate is subjected to pulse electroplating. The cathode current density is controlled at 0.5-2 A / dm², the pulse duty cycle at 10%-30%, the pulse frequency at 100-1000 Hz, the plating solution temperature at 45-55℃, and the pH value at 8-9. Mechanical stirring is used in conjunction with cathode movement to prepare a coating with a thickness of 0.5-2 μm. The pre-prepared plating solution includes gold ions, sodium sulfite, and a leveling agent and / or a wetting agent. The concentration of gold ions is 8-12 g / L, the concentration of sodium sulfite is 80-120 g / L, the concentration of the leveling agent is 10-100 mg / L, and the concentration of the wetting agent is 50-200 mg / L. The leveling agent is at least one of a nitrogen-containing heterocyclic compound or a thiourea derivative, and the wetting agent is at least one of a fatty alcohol polyoxyethylene ether or a sulfonate surfactant.

6. The synchronous acquisition and telemetry device for dual-mode electrophysiological signals as described in claim 2, characterized in that, The fixing sub-assembly includes a first fixing group and a second fixing group; The first fixing group includes a fixing patch, a fixing gel, and a base. The fixing patch adheres and fixes the first microneedle group to a first preset point on the target object. The fixing gel is used to fix the electrodes of the first microneedle group to the base. The first microneedle group is arranged in a single-lead or multi-lead manner. The second fixing group includes a base and a magnetic suction group. The magnetic suction group is used to magnetically fix the base. The base is provided with an array of positioning holes. The electrodes of the second microneedle group are embedded in the positioning holes. The base is positioned at the second preset point of the target object.

7. The synchronous acquisition and telemetry device for dual-mode electrophysiological signals as described in claim 1, characterized in that, Electrophysiological signals include a first electrophysiological signal and a second electrophysiological signal; The signal acquisition component includes a first acquisition module, a second acquisition module, and a clock trigger module; The electrophysiological signals of the target object are collected synchronously at different preset points, specifically including the following steps: At the beginning of each sampling period, the clock trigger module simultaneously sends pulse signals to the first acquisition module and the second acquisition module; In response to the pulse signal, the first acquisition module and the second acquisition module start sampling at the same clock edge, latch the electrophysiological signal in real time, and attach the timestamp generated by the clock trigger module to give the first electrophysiological signal and the second electrophysiological signal.

8. The synchronous acquisition and telemetry device for dual-mode electrophysiological signals as described in claim 7, characterized in that, The monitoring information includes primary monitoring information and secondary monitoring information; The signal analysis component includes an information recognition module and a feature analysis module; The information identification module identifies and processes the monitoring information to form information to be analyzed; The feature analysis module performs feature analysis on the information to be analyzed and provides the telemetry results of the dual-mode electrophysiological signal; The monitoring information is identified and processed to form information to be analyzed, which specifically includes the following steps: The first monitoring information is decompressed, motion artifacts are eliminated, and the first spatiotemporal alignment is performed. Then, feature extraction is performed to form the information to be analyzed; and / or, The second monitoring information is spatiotemporally aligned to form information to be analyzed. The second spatiotemporal alignment specifically includes the following steps: Determine the first feature point sequence and the second feature point sequence in the second monitoring information; Electrophysiological events are identified from the first feature point sequence, and multiple electrophysiological event windows are provided. Based on each electrophysiological event window, the first feature point sequence and the second feature point sequence are divided into segments, and the first feature point sub-sequence and the second feature point sequence corresponding to each electrophysiological event window are given. Within each electrophysiological event window, a corresponding electrophysiological coupling weighted cost matrix is ​​constructed based on the first feature point subsequence and the second feature point subsequence. Based on the electrophysiological coupling weighted cost matrix, a dynamic time warping algorithm is used to search for the optimal alignment path within the corresponding electrophysiological event window; Based on the optimal alignment path within each electrophysiological event window, the corresponding local time offset sequence is determined, bidirectional coupling consistency verification is performed, and the verified local time offset sequence is given. Based on the validated local time offset sequence, the second feature point subsequence within each electrophysiological event window is reconstructed using timestamps, and aligned feature point pairs are given.

9. The synchronous acquisition and telemetry device for dual-mode electrophysiological signals as described in claim 8, characterized in that, Based on the electrophysiological coupling weighted cost matrix, a dynamic time warping algorithm is used to search for the optimal alignment path within the corresponding electrophysiological event window. The specific steps include: Determine the search start and end boundaries corresponding to the electrophysiological event window; Based on the electrophysiological coupling weighted cost matrix, the initial cumulative cost of the search starting point is determined, and the corresponding initial cumulative cost matrix is ​​given. Within the electrophysiological event window, combining monotonicity and continuity constraints, a dynamic programming algorithm is used to recursively calculate the cumulative cost point by point and update the corresponding initial cumulative cost matrix, thus providing the cumulative cost matrix. Based on the cumulative cost matrix, the position of the matrix element with the minimum cumulative cost is selected as the path endpoint on the termination boundary; Starting from the end of the path, and combining the preceding direction information, we backtrack to the starting point of the search to obtain the optimal alignment path.

10. A method for synchronous acquisition and telemetry of dual-mode electrophysiological signals, characterized in that, The method using the synchronous acquisition and telemetry device for dual-mode electrophysiological signals as described in any one of claims 1 to 9 includes the following steps: The microneedle electrode sub-assembly is fixed to a preset point on the target object by a fixing sub-assembly, and the microneedle electrode sub-assembly, fixing sub-assembly, and signal acquisition sub-assembly are integrated and worn on the target object by a wearable sub-assembly; The signal acquisition component is activated, and the electrophysiological signals of the target object at different preset points are acquired synchronously through the electrode component. The electrophysiological signals are then preprocessed to provide monitoring information. The monitoring information is received and analyzed by the signal analysis component, and the telemetry results of the dual-mode electrophysiological signal are given.

Images