2.1 Design, Preparation, and Application of ITO Hydrogel Electrode
2.1 ITO水凝胶电极的设计、制备及应用

The initial development phase of the hydrogel electrode involves preparing a hydrogel matrix that embeds conductive particles. This is achieved by annealing the initial ITO sol—a precursor liquid devoid of crystalline ITO—facilitating the crystallization of ITO particles within the sol matrix, as illustrated in Figure 1a. During this critical process, the concentration of hydrogel precursors remains unchanged while the In³⁺ ions in indium oxide are replaced with Sn⁴⁺ ions, a change brought about by increasing the ITO sol concentration. The reduced indium oxide content and the introduction of tin ions result in numerous oxygen vacancies and an abundance of free electrons, essential for electrical conduction. A water-soluble sodium alginate gel is merged with the ITO semiconductor to form the hydrogel, yielding sodium alginate (C₆H₇O₆Na)_n hydrogel. This mixture is then poured into a mold and cooled, setting the stage for forming a hydrogel precursor specifically designed for EEG signal acquisition. A positive charge group containing calcium salt was added to the hydrogel precursor solution's network structure to strengthen the self-fit fit between the brain electrode's shape and the scalp structure and guarantee the stability of the electrode structure. This allowed the hydrogel electrode for EEG detection to be successfully prepared and solidified (Figure 1a). Furthermore, sodium alginate hydrogels' 3D network structure has a high porosity, which enhances the hydrogels' ability to transport electrons and increases their cyclic stability.
水凝胶电极的初始开发阶段涉及制备嵌入导电颗粒的水凝胶基质。这是通过退火初始ITO溶胶（一种不含结晶ITO的前体液体）来实现的，从而促进溶胶基质内ITO颗粒的结晶，如图1a所示。在这个关键过程中，水凝胶前体的浓度保持不变，而氧化铟中的In ³⁺ 离子被Sn ⁴⁺ 离子取代，这是通过增加ITO溶胶浓度带来的变化。氧化铟含量的降低和锡离子的引入导致了大量的氧空位和大量的自由电子，这对电传导至关重要。水溶性海藻酸钠凝胶与ITO半导体合并形成水凝胶，产生海藻酸钠（C ₆ H ₇ O ₆ Na） _n 水凝胶。然后将该混合物倒入模具中并冷却，为形成专门设计用于脑电图信号采集的水凝胶前体奠定了基础。在水凝胶前体溶液的网络结构中加入含有钙盐的正电荷基团，以加强脑电极形状与头皮结构之间的自拟合，保证电极结构的稳定性。这使得用于脑电图检测的水凝胶电极得以成功制备和固化（图1a）。此外，海藻酸钠水凝胶的三维网络结构具有较高的孔隙率，增强了水凝胶的电子传输能力，提高了其循环稳定性。

The hydrogel electrode's considerable flexibility allows it to be shaped into various shapes to fit different conductive caps. Figure 1b shows a schematic diagram of the head identified in this experiment utilizing conductive caps. By donning an elastic cap to help the hydrogel electrode suit the scalp surface, pressure is given to the EEG cap's surface. After cooling and solidifying, the hydrogel EEG electrode is first reheated to room temperature (20 °C). After that, it is carefully positioned onto the conductive cap's electrode sheet. Subsequently, as shown in Figure 1b, the test subject's head is firmly fitted with the conductive cap that contains the hydrogel electrode. The conduction scenario illustrates the individuals' circumstances in the particular experimental practice and highlights the significance of the elastic cap pushing the hydrogel electrode (Extended Data Figure S1a,b, Supporting Information). And you can also see this very intuitively from the statistical diagram, where all the nonconducting electrodes experience a sharp drop in impedance when subjected to the pressure of the elastic cap (Extended Data Figure S1c, Supporting Information). As soon as the hydrogel electrode and the scalp surface come into close contact, the electrodes' acquisition effect on EEG signals begins. The tight fitting of the electrode on the scalp surface is a result of the high-strength 3D structure of sodium alginate, which is heavily dependent on the hydrogel electrode's adaptive morphologically stable structure (Figure 1b). The stability of the hydrogel's conductivity is critically dependent on its homogeneity. By uniformly grinding ITO powder and incorporating it into the polymer network, we can enhance the distribution of charged particles within the hydrogel. This refinement in the hydrogel's composition bolsters the precision of the hydrogel electrode, optimizing it for accurate EEG signal transmission, as demonstrated in Figure 1a. Consequently, the adaptable deformed homogeneous electrode can successfully create a connection between the device and the brain's synaptic electric field. It can also send the electrical signal produced by the EEG test through the electrode sheet to the device, where it can be output for further processing and analysis. Because sodium alginate is soft and robust, the electrode can endure significant deformation and be readily corrected after loading.
水凝胶电极具有相当大的柔韧性，可以将其塑造成各种形状，以适应不同的导电帽。图1b显示了本实验中利用导电帽确定的头部示意图。通过戴上弹性帽来帮助水凝胶电极贴合头皮表面，对脑电图帽的表面施加压力。冷却和凝固后，首先将水凝胶脑电极重新加热至室温（20°C）。之后，将其小心地定位在导电帽的电极片上。随后，如图1b所示，测试对象的头部牢固地安装在包含水凝胶电极的导电帽上。传导场景说明了特定实验实践中个人的情况，并强调了推动水凝胶电极的弹性帽的重要性（扩展数据图S1a，b，支持信息）。您还可以从统计图中非常直观地看到这一点，其中所有非导电电极在承受弹性帽的压力时都会经历阻抗的急剧下降（扩展数据图S1c，支持信息）。一旦水凝胶电极和头皮表面紧密接触，电极对脑电图信号的采集作用就开始了。电极在头皮表面的紧密配合是海藻酸钠高强度 3D 结构的结果，这在很大程度上依赖于水凝胶电极的适应性形态稳定结构（图 1b）。水凝胶电导率的稳定性在很大程度上取决于其均匀性。通过均匀研磨ITO粉末并将其掺入聚合物网络中，我们可以增强水凝胶中带电颗粒的分布。 水凝胶成分的这种改进提高了水凝胶电极的精度，优化了水凝胶电极的精确脑电图信号传输，如图1a所示。因此，适应性强的变形均质电极可以成功地在设备和大脑的突触电场之间建立连接。它还可以通过电极片将脑电图测试产生的电信号发送到设备，在那里可以输出以进行进一步的处理和分析。由于海藻酸钠柔软而坚固，因此电极可以承受显着变形，并且在加载后易于校正。

2.2 Principles of EEG Testing
2.2 脑电图测试的原理

The brain wave is a human bioelectrical signal, as is well known. It is necessary to fully comprehend the mechanism and underlying principles of bioelectricity's activity on the human body to collect it. In contrast to regular current, bioelectricity in the human body is produced by the movement of ions inside and outside cells. This electrical signal that we can extract from the brain comes from the gray matter structure located in the outermost layer of the cerebral cortex (Figure 1c). This structure is primarily responsible for controlling human behavior and emotional processing. It also shows the brain maps for the lifespan of an individual.^[42] The primary represented areas are the insula, anterior temporal cortex, medial prefrontal, orbitofrontal, ventromedial prefrontal cortex, and anterior/posterior cingulate cortex. The permeability of the cell membrane to particular ions varies in response to stimulation of brain neuron cells, and the osmotic pressure both inside and outside the cell membrane promotes the flow of sodium ions into and out of the membrane. This occurrence causes a positive potential difference between the interior and outside of the neuron, breaking its resting potential and causing the cell to create an action potential. The synaptic vesicles containing neurotransmitters are then released into the synaptic gap and then accepted by the postsynaptic membrane, changing the neuronal potential one by one. This process occurs when the action potential is then transported along the axon to the axon terminal.
众所周知，脑电波是人类的生物电信号。要充分了解生物电对人体活性的机理和基本原理，才能收集到生物电。与常规电流相比，人体中的生物电是由细胞内外离子的运动产生的。我们可以从大脑中提取的电信号来自位于大脑皮层最外层的灰质结构（图1c）。这种结构主要负责控制人类行为和情绪处理。它还显示了个人寿命的大脑图谱。 ⁴² 主要代表区域是岛叶、前颞叶皮层、内侧前额叶、眶额叶、腹内侧前额叶皮层和前/后扣带皮层。细胞膜对特定离子的渗透性随着脑神经元细胞的刺激而变化，细胞膜内外的渗透压促进钠离子流入和流出膜。这种情况会导致神经元内部和外部之间的正电位差，破坏其静息电位并导致细胞产生动作电位。然后，含有神经递质的突触囊泡被释放到突触间隙中，然后被突触后膜接受，逐个改变神经元电位。当动作电位沿着轴突传输到轴突末端时，就会发生此过程。

Two crucial requirements must be met for collecting EEG signals from synapses in the gray matter: first, neurons must be synchronized; second, the direction of the electric field activity created by neurons must remain constant. One of the fundamental ideas behind EEG data gathering is the electric field effect. In other words, the action potential causes current to flow, which in turn causes a magnetic field. This magnetic field then influences the action potential of nearby neurons, activating several neurons at once. When several neurons' dendrites generate an electric field simultaneously and in the same direction, it is known as an open field^[43] and is detectable by EEG equipment (Figure 1c). The typical open electric field structure found in the third and fifth layers of gray matter cells helps to detect electrical signals from which to plot electroencephalograms. The synchronous activity of each neuron produces a constant electric field in the open electric field, and the electric fields in the same direction are superimposed simultaneously. The EEG detector on the scalp can record the electrical impulses from the brain to a certain extent by capturing this superimposed electric field. If the activity of neurons is chaotic, a closed electric field is formed, which means that even if the first necessary condition is met, synchronous discharge occurs, and it is impossible to record the potential of long-distance neuron activity across different directions. Because different neurons have electric fields in different directions, these fields will cancel each other out, and the vector sum will approach 0.
从灰质突触收集脑电信号必须满足两个关键要求：首先，神经元必须同步;其次，神经元产生的电场活动方向必须保持不变。脑电图数据收集背后的基本思想之一是电场效应。换句话说，动作电位导致电流流动，进而产生磁场。然后，该磁场会影响附近神经元的动作电位，同时激活多个神经元。当几个神经元的树突同时以相同的方向产生电场时，它被称为开放场 ⁴³ ，并且可以通过脑电图设备检测到（图1c）。在灰质细胞的第三层和第五层中发现的典型开放电场结构有助于检测电信号，从中绘制脑电图。每个神经元的同步活动在开放电场中产生恒定的电场，同一方向的电场同时叠加。头皮上的脑电图检测器可以通过捕捉这种叠加的电场，在一定程度上记录来自大脑的电脉冲。如果神经元的活动是混沌的，就会形成一个封闭的电场，这意味着即使满足第一个必要条件，也会发生同步放电，并且不可能记录到不同方向的长距离神经元活动的潜力。因为不同的神经元在不同的方向上具有电场，这些场会相互抵消，向量和将趋近于0。

2.3 Conductive Mechanism of Hydrogels/Basic Characterization of Hydrogel Electrode
2.3 水凝胶的导电机理/水凝胶电极的基本表征

We then looked more closely at the variables influencing the electrode's conductivity. Following complete grinding and annealing, morphological analysis revealed that ITO particles formed a uniformly arranged and sized polymer particle structure (Figure 2a). After the homogenous particles were combined and allowed to stand in sodium alginate hydrogel, their 3D network structure was ascertained (Figure 2c). A significant amount of ITO particles are uniformly wrapped in the 3D structure created by the hydrogel as a result of the gel electrode precursor being fully stirred and allowed to stand. This structure carries the conductive particles on the surface and contacts the connected skin. Prior theoretical research anticipated that particles with varying concentrations of indium tin oxide would exhibit ITO diffraction peaks of varying heights after the same annealing method for ITO sol with ordinary concentration and high concentration (4–6 times). High-concentration ITO particles with more diffraction peaks and an increased indium tin oxide content would be added to crystallize following the actual measurement (Figure 2b). The more crystallized ITO particles that can be created by raising the annealing temperature and employing an inert gas as the annealing environment, the more charge carriers that can be formed. The inert environment facilitates the formation of oxygen vacancies and desorption of adsorbed oxygen, which also increases carrier concentration and particle mobility and enhances electrical conductivity.^[46
然后，我们更仔细地研究了影响电极电导率的变量。在完全研磨和退火后，形貌分析显示ITO颗粒形成了均匀排列和尺寸的聚合物颗粒结构（图2a）。将均匀颗粒组合并置于海藻酸钠水凝胶中后，确定其3D网络结构（图2c）。由于凝胶电极前驱体被充分搅拌并允许静置，大量的ITO颗粒被均匀地包裹在水凝胶产生的3D结构中。这种结构在表面上携带导电颗粒并接触连接的皮肤。先前的理论研究预计，在常浓度和高浓度（4-6倍）的ITO溶胶中，采用相同的退火方法后，具有不同浓度的氧化铟锡颗粒将表现出不同高度的ITO衍射峰。在实际测量后，将添加具有更多衍射峰和增加氧化铟锡含量的高浓度ITO颗粒进行结晶（图2b）。通过提高退火温度并使用惰性气体作为退火环境可以产生的结晶ITO颗粒越多，可以形成的电荷载流子就越多。惰性环境有利于氧空位的形成和吸附氧的解吸，这也增加了载流子的浓度和颗粒的迁移率，增强了导电性。 ^{46 ]}

The hydrogel electrode's conductivity is determined by carefully choosing how these conductive particles are prepared (Figure 2e). The hydrogel electrodes produced by physical mixing are comparable to the conductivity provided by conductive pastes widely used for EEG testing. The hydrate's microcrystalline structure enhances the hydrogel's mechanical characteristics and ionic conductivity. Generally, ITO and sodium alginate have diffraction peaks at 21.48°, 30.54°, 35.46°, 51.02°, 60.62° (JCPDS card number 06-0416), 14.0°, 23.0°, respectively. However, some corresponding diffraction peaks were not observed in the XRD patterns of composite hydrogels mixed with ITO particles of different qualities (1%−8%) (Figure 2d). This may be due to the effective chemical cross-linking between ITO and sodium alginate by the free electrons carried by Sn⁴⁺, oxygen vacancy, and carboxyl group during the gel mixing process, which ensures the uniform mixing of the two and prevents the crystallization behavior of ITO and sodium alginate respectively, thus forming a partial amorphous composite aerogel.^[47] The diffraction peak can determine that the optimal concentration of the electrode precursor liquid is a conductive particle mass ratio of 2%. By observing the I–V and R–I curves, the hydrogel electrode cured at the ideal concentration of 2% showed consistent electrical conductivity in the test (Figure 2e,f).
水凝胶电极的电导率是通过仔细选择这些导电颗粒的制备方式来确定的（图2e）。通过物理混合产生的水凝胶电极与广泛用于脑电图测试的导电浆提供的电导率相当。水合物的微晶结构增强了水凝胶的机械特性和离子电导率。通常，ITO和海藻酸钠的衍射峰分别为21.48°、30.54°、35.46°、51.02°、60.62°（JCPDS卡号06-0416）、14.0°、23.0°。然而，在混合了不同质量（1%−8%）的ITO颗粒的复合水凝胶的XRD图谱中未观察到一些相应的衍射峰（图2d）。这可能是由于在凝胶混合过程中，Sn ⁴⁺ 、氧空位和羧基携带的自由电子在ITO和海藻酸钠之间发生了有效的化学交联，从而保证了两者的均匀混合，并防止了ITO和海藻酸钠的结晶行为，从而形成了部分无定形复合气凝胶。 ⁴⁷ 衍射峰可以确定电极前驱体液体的最佳浓度为导电颗粒质量比为2%。通过观察I-V和R-I曲线，在2%的理想浓度下固化的水凝胶电极在测试中显示出一致的电导率（图2e，f）。

2.4 Process of Improving the Performance of Electrodes
2.4 提高电极性能的过程

The adhesion established on various scalp and instrument surfaces by a universal water-based conductive paste for EEG will help the instrument collect EEG signals. However, it is still a big challenge from the perspective of preliminary preparation for experimental detection, post-cleaning, and tester comfort. The preparation process of the hydrogel electrode is simple, and it is convenient to fill and remove. The adhesion degree of the ITO hydrogel electrode and conductive paste is affected by the hydrophilicity of the material itself. The comparison experiment reveals a disparity in hydrophilicity between the cured spherical hydrogel electrode and the conductive paste. Experimental results demonstrate that the contact angle between the spherical electrode and the carrier sheet is 11.2°, while it measures 29.3° for the conductive paste, making it approximately 2.61 times greater than that of the spherical electrode (Extended Data Figure S2a,b, Supporting Information). It can be concluded that the wettability of the pellet electrode is better than that of the conductive paste, and the different hydrophilicity will affect the convenience of subsequent cleaning of different conductive methods. After the precursor solution is cured, a sodium alginate hydrogel electrode containing Ca²⁺ ions on the surface is formed, which will not stick to the tester's scalp and the instrument, simplifying the process before and after the test. An electroencephalogram is a charted representation of the brain's spontaneous electrical activity captured through the interface between a conductive substance and the scalp. Incorporating ITO particles into the hydrogel electrode enhances its electrical conductivity and stability. Moreover, the adaptable nature of the hydrogel's structure guarantees consistent and firm contact with the scalp, ensuring reliable signal acquisition. This mechanism allows hydrogel electrodes of different mixed concentrations (1%−8%) to transmit electrical signals from almost any untreated scalp.
用于脑电图的通用水基导电膏在各种头皮和器械表面上建立的粘附力将有助于器械收集脑电图信号。然而，从实验检测、后清洗和测试人员舒适度的前期准备来看，这仍然是一个很大的挑战。水凝胶电极的制备工艺简单，填充和取出方便。ITO水凝胶电极和导电浆料的粘附度受材料本身的亲水性影响。对比实验表明，固化的球形水凝胶电极与导电浆料的亲水性存在差异。实验结果表明，球形电极与载体片的接触角为11.2°，而导电浆料的接触角为29.3°，约为球形电极的2.61倍（扩展数据图S2a，b，支持信息）。可以得出结论，颗粒电极的润湿性优于导电浆料，不同的亲水性会影响后续不同导电方法清洗的便利性。前驱体溶液固化后，表面形成含有Ca ²⁺ 离子的海藻酸钠水凝胶电极，不会粘在测试人员的头皮和仪器上，简化了测试前后的流程。脑电图是通过导电物质和头皮之间的界面捕获的大脑自发电活动的图表表示。将ITO颗粒掺入水凝胶电极中可增强其导电性和稳定性。 此外，水凝胶结构的适应性保证了与头皮的一致和牢固的接触，确保了可靠的信号采集。这种机制允许不同混合浓度（1%−8%）的水凝胶电极从几乎任何未经处理的头皮传输电信号。

To produce a reproducible EEG test setup, the conductive cap has a standardized set of electrode positions on the skull, such as the International 10/20 system (Figure 3a, left).^[48] The percentages 10% and 20% denote the separation between neighboring electrodes, equivalent to 10% or 20% of the total distance between the left and right sides of the front and rear of the skull. The four anatomical markers Nasion (the point between the forehead and the nose), Inion (the lowest point beneath the skull) (lz), left anterior ear point (A1), and right anterior ear point (A2) are used to ensure that the electrodes are positioned correctly. The letters F, T, C, P, O, and Z denote the frontal, temporal, central, parietal, and occipital lobes, respectively, and the electrodes are positioned on the midline. Even numbers indicate the electrodes in the right hemisphere, whereas odd numbers show those in the left hemisphere. There are no two electrode locations, Fpz and Oz, in the 19-electrode system. Potential differences were gathered by the scalp electrodes and entered into the computer. The program acquired the EEG signal of each hydrogel electrode and its resistance value (Figure 3a, right).
为了产生可重复的脑电图测试设置，导电帽在颅骨上有一组标准化的电极位置，例如国际 10/20 系统（图 3a，左）。 ⁴⁸ 百分比 10% 和 20% 表示相邻电极之间的距离，相当于颅骨前后左右两侧总距离的 10% 或 20%。四个解剖标记 Nasion（前额和鼻子之间的点）、Inion（颅骨下方的最低点）（lz）、左前耳点 （A1） 和右前耳点 （A2） 用于确保电极正确定位。字母 F、T、C、P、O 和 Z 分别表示额叶、颞叶、中央叶、顶叶和枕叶，电极位于中线。偶数表示右半球的电极，而奇数表示左半球的电极。在 19 个电极系统中没有两个电极位置，即 Fpz 和 Oz。电位差由头皮电极收集并输入计算机。该程序采集了每个水凝胶电极的脑电图信号及其电阻值（图3a，右）。

We tested the electrode resistance of several hydrogels with varying mass ratios combined with conductive powder to assess the hydrogel electrode's conductivity. We then counted the various electrode resistance ranges under numerous tests (Figure 3f,g). Because the mixing mass ratio of these hydrogel electrodes is reasonably reasonable, the conductivity is high, and all 19 conductive electrodes can be guaranteed (Figure 3d,e) for hydrogel electrodes mixed with an appropriate concentration of ITO, such as Sample 1 (2% mass ratio electrode) and Sample 2 (4% mass ratio electrode). By guaranteeing the concentration of conductive particles, the structural integrity may be preserved under adaptive deformation throughout the test procedure, and the electrode's conductivity can be significantly enhanced. The conductivity of hydrogel electrodes mixed with excessively high or low ITO, like Samples 4 (8% mass ratio electrode) and 3 (1% mass ratio electrode), is greatly decreased because of the low concentration of conductive particles, or the hydrogel surface is harmed during the adaptive deformation process before the test is finished (Figure 3b,c). After conducting numerous tests, it was discovered that, out of the 21 cases, the average resistance impedance of Sample 1 hydrogel electrodes needed for each test is less than 15 KΩ (ohms), with the most significant average proportion of 0–10 KΩ electrodes (Figure 3g). These promising electrical conductivity findings highlight the potential of hydrogel pellet electrodes for real-world uses, particularly in the biomedical industry.^[49
我们测试了几种具有不同质量比的水凝胶与导电粉末结合的电极电阻，以评估水凝胶电极的电导率。然后，我们在多次测试中计算了各种电极电阻范围（图3f，g）。由于这些水凝胶电极的混合质量比合理，电导率高，对于混合了适当浓度ITO的水凝胶电极，如样品1（质量比为2%）和样品2（质量比为4%），可以保证所有19个导电电极（图3d，e）。通过保证导电颗粒的浓度，可以在整个测试过程中保持结构完整性，并显着提高电极的导电性。由于导电颗粒浓度低，或在测试完成前的自适应变形过程中，水凝胶表面在自适应变形过程中受到损害，因此与过高或过低的ITO混合的水凝胶电极的电导率大大降低（图3b，c）。在进行了大量测试后，发现在 21 个案例中，每次测试所需的样品 1 水凝胶电极的平均电阻抗小于 15 KΩ（欧姆），其中 0–10 KΩ 电极的平均比例最显着（图 3g）。这些有希望的电导率研究结果凸显了水凝胶颗粒电极在实际用途中的潜力，特别是在生物医学行业。 ^{49 ]}

Furthermore, the degree of hydrogel solidification impacted the hydrogel's conductivity in the ITO hydrogel electrode. The porous structure was caused by the uncured hydrogel material gradually degrading during the investigation (Figure 2c). The findings of the experiment indicate that the ideal concentration of specific curing liquid is 1:20. Additionally, the effects of varying curing liquid concentrations on the conductivity of Sample 1—the ideal specific hydrogel electrode—are contrasted (Extended Data Figure S3a–d, Supporting Information). To preliminarily evaluate the biocompatibility of the cytotoxicity of the hydrogel electrode on the cells was detected by CCK-8 viability assay, and from Extended Data Figure S4 (Supporting Information), it can be seen that when the mass of the ball was increased up to 1.5 mg, the ball was virtually free of cytotoxicity to the cells.
此外，水凝胶的凝固程度会影响水凝胶在ITO水凝胶电极中的电导率。多孔结构是由未固化的水凝胶材料在研究过程中逐渐降解引起的（图2c）。实验结果表明，比固化液的理想浓度为1：20。此外，还对比了不同固化液浓度对样品 1（理想的特定水凝胶电极）电导率的影响（扩展数据图 S3a–d，支持信息）。为了初步评估水凝胶电极对细胞的细胞毒性的生物相容性，通过CCK-8活力测定法检测，从扩展数据图S4（支持信息）可以看出，当球的质量增加到1.5mg时，球对细胞几乎没有细胞毒性。

2.5 Application of Hydrogel Electrode in EEG Detection
2.5 水凝胶电极在脑电检测中的应用

The test individuals' EEG signals were tracked using an EEG monitoring procedure and captured as high-time resolution EEG data. It is important for researching neural networks, neural engineering, neurology, and cross-dataset learning clinical applications.^[50] However, EEG monitoring methods usually require a long processing time before and after termination. In the case of high-density electrodes, there are various noise sources called artifacts, such as the potential difference of muscle activity near the acquisition site, eye movement potential difference, blink potential difference, etc.^[51] whose production interferes with the recording of EEG electrical signals. Long monitoring procedures can cause discomfort to patients or make testing more difficult. Therefore, biomedical electrode devices that can be operated quickly and have high comfort for monitoring brain waves, such as hydrogel electrodes, are necessary to monitor EEG signals rapidly in patients at any time. Essential factors for precise EEG signal monitoring with conductive electrodes involve the accurate and effortless transmission of the target's EEG signals. However, conventional EEG monitoring techniques, such as using conductive paste, compromise patient comfort and are prone to introducing noise into the signal transmission from bodily movements or involuntary patient activities. To demonstrate the practical application of the hydrogel electrode, the hydrogel electrode is fixed between the scalp and the conductive cap through the loading of the electrode sheet to monitor the brain waves of the test subjects continuously in real-time.
使用脑电图监测程序跟踪测试个体的脑电图信号，并将其捕获为高时间分辨率的脑电图数据。它对于研究神经网络、神经工程、神经病学和跨数据集学习临床应用非常重要。 ⁵⁰ 然而，脑电图监测方法在终止前后通常需要较长的处理时间。在高密度电极的情况下，存在各种称为伪影的噪声源，例如采集部位附近肌肉活动的电位差、眼球运动电位差、眨眼电位差等， ⁵¹ 其产生会干扰脑电信号的记录。长时间的监测程序会给患者带来不适或使测试更加困难。因此，水凝胶电极等能够快速操作且对监测脑电波具有高舒适性的生物医学电极设备，对于随时快速监测患者的脑电信号是必要的。使用导电电极进行精确脑电图信号监测的基本因素包括准确、轻松地传输目标的脑电信号。然而，传统的脑电图监测技术（例如使用导电膏）会损害患者的舒适度，并且容易在身体运动或患者不自主活动的信号传输中引入噪声。为了演示水凝胶电极的实际应用，水凝胶电极通过电极片的加载固定在头皮和导电帽之间，以连续实时监测被测对象的脑电波。

Another significance of hydrogel elastic electrodes is that they provide load-carrying capacity without affecting performance or structural integrity.^[52] To evaluate its usefulness, we compared the acquisition of different EEG signals with a conductive paste and a conductive hydrogel electrode (Sample 1). Our objective was to expedite the evaluation of hydrogel electrodes as potential replacements for conductive paste. Therefore, we enlisted a cohort of participants to record EEG signals under two conditions—eyes open and eyes closed—first utilizing conductive paste and subsequently with hydrogel electrodes for a direct performance comparison. The signal measured by the conductive paste without initial signal processing usually has a low signal-to-noise ratio. Preprocessing steps such as filtering are usually used to minimize artifacts and unnecessary noise to improve the signal-to-noise ratio. However, care must be taken and visualized using independent component analysis (ICA) to avoid removing any meaningful and information-rich EEG components.^[53] After analysis, it is preliminarily concluded that both the hydrogel electrode and the conductive paste can be used for EEG acquisition, and the EEG time-domain graphs obtained by the two conductive methods are very similar in terms of peak value and fluctuation amplitude after pretreatment (Figure 4a–d). By comparing Figure 4c,d, it can be found that when the hydrogel electrode is used to collect brain waves, the signals collected by the conductive paste will show a larger and more standardized fluctuation amplitude, which shows that the collected EEG signals are more complete in detail and are not easy to be distorted.
水凝胶弹性电极的另一个意义是它们在不影响性能或结构完整性的情况下提供承载能力。 ⁵² 为了评估其有用性，我们比较了使用导电膏和导电水凝胶电极（样品 1）采集不同的 EEG 信号。我们的目标是加快水凝胶电极作为导电浆料潜在替代品的评估。因此，我们招募了一组参与者在两种条件下记录脑电图信号——睁开眼睛和闭上眼睛——首先使用导电膏，然后使用水凝胶电极进行直接性能比较。由导电浆测量的信号，未经初始信号处理，通常具有较低的信噪比。滤波等预处理步骤通常用于最大限度地减少伪影和不必要的噪声，以提高信噪比。但是，必须小心使用独立分量分析 （ICA） 进行可视化，以避免去除任何有意义且信息丰富的脑电图分量。 ⁵³ 经分析，初步认为水凝胶电极和导电浆均可用于脑电图采集，两种导电方法得到的脑电图在预处理后的峰值和波动幅度方面非常相似（图4a-d）。通过对比图4c，d可以发现，当使用水凝胶电极采集脑电波时，导电浆收集的信号会显示出更大、更标准化的波动幅度，这表明采集到的脑电信号在细节上更加完整，不易失真。

We require evidence that hydrogel electrodes may take the role of conductive paste in the initial waveform similarity comparison to gather brain waves. We can examine the EEG signal's frequency because it is recorded as a sine wave. The EEG signal under study typically ranges from 0.5 to 30 Hz. Once the relationship between the subject's mental motor state and the signals found on the scalp is observed, the signals are classified into bands, or rhythms, that correspond to particular ranges that are more prominent in particular mental states: α (8–13 Hz), β (13–30 Hz), θ (4–7 Hz), and δ (0.5–3.5 Hz).^[55] These rhythms' EEG signals can be retrieved using spectral analysis techniques like wavelet transform, etc., and the energy ratio of each rhythm's EEG signal can be computed to process and evaluate the EEG data. These four rhythms' waveforms may coexist in the same region of the brain, and because the brain engages in many thought processes at once, the EEG signals of each rhythm exhibit distinct intensity features. Therefore, using paste and hydrogel electrodes made with varying concentrations in both the open and closed eye conditions, we confirmed the similarity of accuracy in identifying the brain waves collected several times. By applying frequency domain analysis to the energy proportions of the four rhythmic waves, it was able to classify the collected brain waves (Extended Data Figure S5a, Supporting Information). The ideal hydrogel electrode concentration ratio and the presumptive outcomes can take the place of or surpass the conductive paste in terms of stable classification accuracy.
我们需要证据证明水凝胶电极可能在初始波形相似性比较中发挥导电浆的作用以收集脑电波。我们可以检查脑电图信号的频率，因为它被记录为正弦波。所研究的脑电图信号通常在 0.5 至 30 Hz 之间。一旦观察到受试者的精神运动状态与头皮上发现的信号之间的关系，信号就会被分类为对应于特定精神状态中更突出的特定范围的波段或节奏：α（8-13 Hz）、β （13-30 Hz）、θ （4-7 Hz） 和 δ （0.5-3.5 Hz）。 ⁵⁵ 这些节律的脑电信号可以使用小波变换等频谱分析技术进行检索，并且可以计算每个节律的脑电信号的能量比来处理和评估脑电图数据。这四种节律的波形可能共存于大脑的同一区域，并且由于大脑同时参与许多思维过程，因此每个节律的脑电信号表现出不同的强度特征。因此，使用在睁眼和闭眼条件下以不同浓度制成的糊状物和水凝胶电极，我们确认了识别多次收集的脑电波的准确性的相似性。通过对四个节奏波的能量比例进行频域分析，它能够对收集到的脑电波进行分类（扩展数据图S5a，支持信息）。理想的水凝胶电极浓度比和推定结果在稳定的分类精度方面可以取代或超过导电浆料。

We introduced an emotion recognition task based on the EEG signals collected to validate the electrodes' capability to consistently and accurately capture brain waves across various emotional states. This involved preprocessing the acquired EEG data after participants viewed videos designed to evoke a spectrum of emotions during the experiment. This step ensures the integrity and reliability of the electrical signal transmission, allowing for a comprehensive analysis of how different emotional states are reflected in the brain's electrical activity (Figure 5a–d). Following examination, the following four temporal domain features (mean value, first-order difference, second-order difference, and standard deviation) and four frequency domain features (α, β, θ, and δ, which represented the percentage of the total energy of the four signals) were chosen for accuracy study.^[56] Even after wearing for an extended period, there may be minimal variation in the power spectral density and amplitude of the alpha rhythm wave signal between the EEG signals recorded by the conductive paste and the hydrogel (sample 1) under conditions of anger (Figure 5e). When there is little to no difference between the two, more quantitative analysis is required. The primary classification principle of Support Vector Machine (SVM), a supervised machine learning technique, is choosing the best hyperplane that meets the following two requirements: It is possible to optimize the distance between samples on both sides while isolating the two types of samples. The kernel function can map the sample to the high-dimensional space for classification when the data cannot be separated in the low-dimensional space. The SVM method is ideal for hydrogel electrodes to finish recognizing and classifying eyes and emotions since it does not require many samples, handles linear indivisible data, and selects alternative kernel functions. Through the classifier's operation, the classification matrix was obtained. The hydrogel electrode's ideal ratio remained optimal even after all SVM processes (Figure 5g). The 152D classifier (8 parameters * 19 channels) was used to classify the features related to emotion recognition. With the ideal ratio electrode, the medium Gaussian SVM (SVM1) and exact Gaussian SVM (SVM2) had the highest eye-opening and eye-closing recognition rates (83.5% and 99.3%, respectively) and the most excellent stable emotion recognition accuracy among hydrogel electrodes and conductive paste.
我们引入了一项基于收集的脑电图信号的情绪识别任务，以验证电极在各种情绪状态下一致准确地捕获脑电波的能力。这涉及在参与者观看旨在唤起实验期间一系列情绪的视频后对获取的脑电图数据进行预处理。这一步骤确保了电信号传输的完整性和可靠性，可以全面分析不同的情绪状态如何反映在大脑的电活动中（图5a-d）。经过检查，选择以下4个时域特征（平均值、一阶差、二阶差和标准差）和4个频域特征（α、β、θ和δ，它们代表了四个信号的总能量的百分比）进行精度研究。 ⁵⁶ 即使在长时间佩戴后，在愤怒条件下，导电浆和水凝胶（样品1）记录的脑电图信号之间的α节律波信号的功率谱密度和振幅也可能有最小的变化（图5e）。当两者之间几乎没有差异时，需要更多的定量分析。支持向量机 （SVM） 是一种监督机器学习技术，其主要分类原则是选择满足以下两个要求的最佳超平面： 在隔离两种类型的样本的同时，可以优化两侧样本之间的距离。当数据在低维空间中无法分离时，核函数可以将样本映射到高维空间进行分类。 SVM 方法非常适合水凝胶电极完成对眼睛和情绪的识别和分类，因为它不需要很多样品，可以处理线性不可分割的数据，并选择替代的核函数。通过分类器的操作，得到分类矩阵。即使在所有SVM工艺之后，水凝胶电极的理想比例仍保持最佳状态（图5g）。采用152D分类器（8个参数*19个通道）对情绪识别相关特征进行分类。在理想比例电极下，介质高斯支持向量机（SVM1）和精确高斯向量机（SVM2）在水凝胶电极和导电浆料中具有最高的睁眼和闭眼识别率（分别为83.5%和99.3%），稳定性识别精度最优异。

Furthermore, as indicated by the feature confusion matrix in the time-frequency domain following feature selection (Figure 5f), the figure illustrates that when the hydrogel electrode ratio is at its optimal ratio (sample 1), the subjects' mental activities are more concentrated and well-organized, and their neural electrical activities are more consistent. As a result, the EEG signals will contain a greater number of identical or similar features, which will facilitate identification. A comparison of the confusion matrix provided evidence of the hydrogel electrode's substitutability for the conductive paste in emotion recognition.
此外，如特征选择后时频域中的特征混淆矩阵所示（图5f），该图表明，当水凝胶电极比率处于最佳比例（样本1）时，受试者的心理活动更加集中和有条理，他们的神经电活动更加一致。因此，脑电图信号将包含更多相同或相似的特征，这将有助于识别。混淆基质的比较提供了水凝胶电极在情绪识别中对导电浆料的替代性的证据。

Conversely, using conductive paste during the testing led to discomfort among the subjects, resulting in more varied psychological responses. This discomfort correlated with increased divergence in their neuronal electrical activities, making the brain wave patterns more chaotic and less predictable. Consequently, the EEG signals exhibited unexpected and unpredictable characteristics, complicating the identification process and reducing accuracy. However, it is worth noting that feature selection effectively minimized the feature dimensions while preserving the accuracy of classification, thereby validating the substitution of conductive paste with an optimal ratio of hydrogel electrode. Furthermore, we applied the linear discriminant analysis to ascertain its viability, and the results were consistent with those obtained using the SVM classification method. This consistency underscores the feasibility and broad applicability of replacing conductive paste with hydrogel electrodes in EEG signal acquisition (Extended Data Figure S5b, Supporting Information).
相反，在测试过程中使用导电膏会导致受试者感到不适，从而导致更多样化的心理反应。这种不适与神经元电活动的发散增加有关，使脑电波模式更加混乱和难以预测。因此，脑电图信号表现出意想不到和不可预测的特征，使识别过程复杂化并降低准确性。然而，值得注意的是，特征选择有效地最小化了特征尺寸，同时保持了分类的准确性，从而验证了用最佳比例的水凝胶电极替代导电浆料。此外，我们应用线性判别分析来确定其可行性，结果与使用SVM分类方法获得的结果一致。这种一致性强调了在脑电图信号采集中用水凝胶电极代替导电膏的可行性和广泛适用性（扩展数据图S5b，支持信息）。

Moreover, considering the well-documented hydrophilic and biocompatible properties of sodium alginate components, we investigated the impact of hydrogel electrodes on the scalp comfort of test subjects. This exploration aimed to assess the technical performance of hydrogel electrodes and their practicality in user experience and safety during EEG signal collection. The results demonstrated that the hydrogel electrode was not sticky, the task of washing the test subjects' scalps was eliminated, and the test subjects' comfort was noticeably increased compared to the control group utilizing regular conductive paste as the electrode for the EEG test.
此外，考虑到海藻酸钠成分的亲水性和生物相容性，我们研究了水凝胶电极对测试对象头皮舒适度的影响。本研究旨在评估水凝胶电极的技术性能及其在脑电信号采集过程中的用户体验和安全性的实用性。结果表明，与使用常规导电膏作为电极的对照组相比，水凝胶电极不粘，消除了清洗测试对象头皮的任务，并且测试对象的舒适度明显提高。