Abstract
ECG monitoring in mice is an essential tool for assessing physiological health. However, acquiring high-quality ECG signals from freely moving and conscious mice has remained a significant technical challenge. This study presents a non-invasive wearable real-time ECG monitoring platform for mice, integrating hollow ultra-flexible electrodes with a soft wireless circuit system, enabling high-fidelity real-time ECG monitoring during free activity. Using microfabrication and transfer techniques, we developed ultra-flexible electrodes that are scalable for large-area manufacturing and customizable to specific requirements. These electrodes conform tightly to the surfaces of the mouse’s heart, abdomen, and limbs, enabling high-quality ECG signal acquisition. Even under 40% strain, the electrodes maintain stable mechanical and electrical performance. Experimental results demonstrate that the acquired ECG signals exhibit clear and distinguishable QRS waveforms, with amplitudes comparable to those obtained using traditional needle electrodes. In addition, we designed and developed a wearable soft wireless circuit system encapsulated in an “electronic backpack” mounted on the mouse’s back. This system features low-power wireless signal transmission, enabling real-time, long-term ECG signal acquisition. To validate system performance, we established a mouse myocardial infarction model and used the platform for ECG signal monitoring and analysis. The results showed typical features such as ST-segment elevation and baseline abnormalities, which are highly consistent with clinical observations. The proposed non-invasive wearable ECG monitoring platform provides an innovative, cost-effective solution for high-efficiency physiological signal acquisition and drug evaluation in mice, offering broad application potential and significant value for further development.
Similar content being viewed by others
Introduction
Laboratory mice are widely regarded as the preferred animal model in biomedical research, playing a crucial role in both medical investigations and pharmacological drug development. Physiological parameters, including electrocardiogram (ECG), electromyography (EMG), and body temperature, are essential indicators for assessing physiological states, pharmacokinetics, and drug efficacy in these models. Conventional ECG monitoring techniques typically rely on the insertion of needle electrodes into anesthetized mice1,2,3. However, such invasive approaches pose significant limitations, particularly in their inability to achieve continuous real-time monitoring. Moreover, the use of anesthetics has been shown to profoundly suppress cardiac activity and alter the pharmacodynamics of cardiovascular drugs, introducing substantial variability and potential bias into experimental outcomes4,5. Additionally, repeated needle electrode insertions often result in localized inflammation and tissue damage at the puncture site, further compromising the integrity and reproducibility of the collected data. These limitations highlight an urgent need for non-invasive, real-time, and minimally disruptive ECG monitoring technologies to ensure the accuracy, consistency, and reliability of physiological data in preclinical studies using mouse models.
ECG monitoring in awake mice plays a crucial role in advancing research in cardiac electrophysiology, cardiovascular pharmacology, toxicology, and drug development. Recently developed implantable ECG modules have addressed some limitations of traditional wired systems, including constraints imposed by tethered connections and the physiological effects of anesthetics6,7,8. These implantable systems enable the acquisition of ECG signals under more physiologically relevant conditions, reflecting the natural cardiac activity of freely moving mice. However, the use of implantable ECG modules introduces several limitations. First, the implantation procedure requires surgical expertise and must be performed by trained personnel with specialized skills. This surgical intervention not only increases the risk of post-operative complications but also necessitates a recovery period7, which can significantly limit throughput in experiments involving large cohorts of mice. Second, data acquisition from implantable devices often relies on specific toroidal equipment, restricting the flexibility of data collection and analysis workflows2,8,9,10. These constraints limit the scalability and adaptability of implantable ECG systems for high-throughput experimental designs. Therefore, achieving continuous, non-invasive, and high-fidelity ECG monitoring in large cohorts of freely moving, conscious, and undisturbed mice remains an outstanding challenge in the field of biomedical research. Overcoming these limitations will require the development of innovative non-invasive ECG monitoring technologies that combine precision, scalability, and operational flexibility to meet the demands of modern preclinical studies.
On the one hand, developing electrodes that conform closely to the skin is critical for acquiring high-quality ECG signals from active mice. Existing electrodes, including gel and dry types, face significant limitations. Gel electrodes, commonly used in clinical and large animal studies due to their low contact impedance and superior signal conduction, suffer from issues such as gel drying, signal degradation over time, and poor adhesion on the limited and irregular surface area of mouse skin11,12,13. Dry electrodes mitigate gel drying issues and offer some advantages in small animal experiments, but encounter challenges like high skin contact impedance, susceptibility to motion artifacts, and difficulties in accommodating the curved, wrinkled surfaces of mouse skin. These limitations lead to frequent detachment during prolonged use14. Additionally, mice may chew on the electrodes during activity, further compromising signal reliability. On the other hand, a miniaturized, soft-material wireless circuit system is essential for long-term physiological monitoring in freely moving mice. Traditional rigid circuits cause discomfort, avoidance behaviors, and disrupted data due to their structural and material limitations, interfering with natural physiology and experimental reliability. To address these challenges, wireless systems must prioritize miniaturization, lightweight design, and flexibility to reduce physical burden and maintain natural activity. Flexible materials like polyimide (PI) and silicone conform to curved surfaces, enhancing comfort and stability. Low-power communication technologies support real-time ECG monitoring with minimal battery maintenance. Cost efficiency and reusability are also critical for large-scale studies. This advancement significantly improves monitoring accuracy and feasibility in research.
This study proposes a non-invasive wearable ECG real-time monitoring platform for mice, integrating hollow ultra-flexible electrodes with a soft wireless circuit system. The platform enables real-time ECG monitoring in freely moving and awake mice. On one hand, the ultra-flexible electrodes, developed using microfabrication and transfer techniques, allow for large-scale manufacturing and customization based on experimental needs. Their hollow and stretchable structure ensures tight adhesion to the surfaces of the mouse’s heart, abdomen, and limbs, significantly reducing motion artifacts and lowering the risk of damage from chewing. Additionally, the electrodes improve breathability and wearing comfort, making them suitable for extended monitoring sessions. On the other hand, the miniaturized and lightweight soft wireless circuit system is encapsulated within an “electronic backpack” attached to the mouse’s back, featuring low-power wireless signal transmission capabilities. Combined with a mobile software platform, the system supports real-time data acquisition and differential analysis across multiple experimental groups, greatly enhancing data processing efficiency and experimental reproducibility. Furthermore, validation experiments on a mouse myocardial infarction model demonstrated the system’s ability to accurately capture key ECG features during the progression of myocardial infarction. The soft, porous, and breathable electrode material minimizes skin irritation and stress-induced interference in ECG recordings during long-term wear. Behavioral analysis and serum biomarkers (CORT, IL-1β, TNFα) confirm negligible chronic psychological stress in mice, ensuring high physiological relevance and reliability for sustained biomedical monitoring. This system shows potential to replace existing needle electrodes and implantable wireless monitoring systems, offering an innovative, efficient, and cost-effective tool for animal experiments and drug evaluation.
Results
Design and fabrication of epidermal electrodes
Stretchability serves as a critical parameter for assessing the mechanical performance of flexible epidermal electrodes, especially in scenarios requiring large deformation adaptability. The effectiveness of these electrodes heavily relies on the rational design of their geometric structures, which directly influences their mechanical resilience and functional stability. Various structural patterns, including serpentine, V-shaped, Z-shaped, wavy, and bio-inspired designs, have been widely explored to optimize stretchability15. As shown in Fig. 1a, finite element simulations were conducted using ABAQUS to evaluate the mechanical behavior of various electrode patterns, including Z-type, W-type, serpentine, serpentine interconnection, and serpentine island-bridge designs. Among them, the serpentine island-bridge configuration demonstrated the best stretchability. Subsequently, FEA was employed to systematically optimize the geometric parameters of the flexible epidermal electrodes under displacement loading, as illustrated in Fig. S1. Under 20% uniaxial tensile strain, the serpentine island-bridge structure exhibited outstanding mechanical compliance and deformation tolerance, indicating its strong potential for use in stretchable electronic applications. Additional details regarding the electrode structures and design rationale can be found in Supplementary Note S1. Comprehensive FEM simulation parameters, including material properties and boundary conditions, are summarized in Supplementary Tables S1–S4 to ensure clarity and reproducibility.
a Finite element analysis map of epidermal electrodes under different stretching shapes. b The conformal contact model for epidermal electrodes. c The preparation process of epidermal electrodes. d Section schematic of epidermal electrodes in the preparation process. e, f Epidermal electrode transfer process. g Photos of epidermal electrodes attached to the abdomen, legs, and heart surface of a mouse. h Photos of epidermal electrodes attached to the legs of the mouse
The epidermal electrode fabrication begins with polyvinyl alcohol (PVA) uniformly coated onto a 5-μm PI film, followed by thermal curing to ensure strong adhesion. Titanium (Ti, 70 nm) and gold (Au, 350 nm) layers are sequentially deposited via magnetron sputtering, forming a metal composite structure. Next, a positive photoresist is spin-coated and thermally cured, followed by photolithographic patterning and plasma etching to define the electrode structure (Fig. 1b–d). The water-soluble PVA layer, upon hydration, enables controlled transfer of the patterned electrode onto different substrates. Initially transferred onto a balloon surface as an intermediate substrate (Fig. 1e), the electrode can subsequently be transferred onto Ecoflex flexible substrates or polyurethane (PU) thin films (Fig. 1f). This streamlined microfabrication process was developed for Au electrodes on PI substrates via a single-step photolithography approach. This process effectively balances the trade-offs among electrode line width resolution, film thickness, and fabrication cost, addressing a bottleneck in scalable microelectronic fabrication for flexible bioelectronics. In addition, the results showed that when the electrodes were applied to the mouse’s abdomen, legs, and heart surface, they exhibited good conformability (Fig. 1g). Fig. 1h showed that electrode signals could be collected without causing tissue damage, while the mouse was able to move freely. Experimental results demonstrate that the epidermal electrode remains wrinkle-free and intact during transfer, ensuring structural integrity and functional stability. This process offers a reliable pathway for scalable fabrication and application of high-performance flexible epidermal electrodes.
Performance test of epidermal electrodes
To ensure accurate and stable ECG signal acquisition, epidermal electrodes must exhibit both high deformability and non-destructive characteristics. As illustrated in Fig. 2a, the electrodes maintain intimate contact with mouse skin surfaces even under compression, stretching, and twisting, demonstrating exceptional mechanical compliance. This characteristic is critical for ensuring consistent signal transmission at the skin-electrode interface, particularly in dynamic or deformation-prone conditions. Under unidirectional tensile strain exceeding 40% along the x-axis, no visible cracks or structural damage were observed on the electrode surface (Fig. 2b). To further elucidate the mechanical response under tensile deformation, finite element modeling (FEM) was performed to simulate the strain distribution across the electrode. Simulation results revealed that at 40% tensile strain, the maximum local strain was limited to 1.89%. The elongation of gold (Au) thin films varies significantly depending on film thickness, deposition method, cracks or defects, and whether the film is freestanding or supported by a flexible substrate. Typically, sputtered or evaporated Au films deposited on flexible PI substrates can tolerate 5–15% strain. With structural design optimizations of wrinkles, crack-bridging structures, serpentine layouts, or wavy configurations, the achievable strain range can be extended to 10–30%16,17. This agreement confirms the accuracy and predictive reliability of the finite element analysis. To evaluate the stretchability and electrical performance of the electrodes across various application scenarios, segmental tensile tests were conducted on Ecoflex-encapsulated epidermal electrodes, with electrical resistance monitored in real-time. Using a standard tensile testing platform, the electrodes were subjected to controlled stretching from 0% to 40%, with each stage held for 5 seconds while resistance data were recorded using a digital multimeter. Results indicated that at 50% tensile strain, the electrical resistance remained stable at 28.2 Ω, confirming robust and reliable electrical connectivity under significant mechanical deformation (Fig. 2d). Additionally, the excellent mechanical compliance of the electrodes enhances wearing comfort and reduces skin irritation. After being attached to mouse skin for over 24 hours, no visible damage or adverse skin reactions were detected (Fig. 2c), further validating their biocompatibility for prolonged use. In summary, epidermal electrodes demonstrate remarkable mechanical adaptability, electrical stability, and biocompatibility. Their ability to maintain structural integrity and functional performance under prolonged dynamic conditions establishes a robust foundation for their practical deployment in advanced biomedical sensing applications. To evaluate the interfacial adhesion between the epidermal electrode and mouse skin, a uniaxial tensile test was performed using a mechanical testing machine. After firmly attaching the electrode to the dorsal skin of the mouse, the bonded interface (with a contact area of 3.9 × 1.5 cm2, totaling 5.85 cm2) was subjected to peeling at a constant speed of 1.8 mm/min. The measured peeling force increased rapidly at the onset of detachment, reaching a peak value exceeding 1.2 N, and then plateaued, indicating a saturated adhesive interface. This strong and stable adhesion ensures reliable skin-electrode contact, which is critical for consistent signal acquisition in dynamic or long-term applications.
a Performance of epidermal electrodes under different mechanical deformations. b Tensile test results of the epidermal electrodes. c Effects of epidermal electrode attachment on the limbs of mice. d Resistance changes of epidermal electrodes under segmented stretching conditions. e Adhesion force test between the electrode and mouse skin
Design and implementation of a flexible mouse ECG monitoring system
A flexible printed circuit board (FPCB) measuring 25 mm × 32 mm was fabricated to integrate key components, including the BMD101 detection chip, low-power transmission chip CC2640R2F (MCU), DC voltage regulator, and power supply module, using a reflow soldering process (Fig. 3a, b). The mouse ECG data is wirelessly transmitted via a Bluetooth module to a computer, smartphone, or other user terminals. The system utilizes flexible conductive fabric to interconnect the epidermal electrodes and operates with a 3.7 V lithium battery. The system exhibits a standby current as low as 300 μA under a 3.3 V logic supply, enabling continuous operation for at least 5 days under typical usage conditions. The average power consumption of all circuit components is ~3 mW, which can be reliably supported by a standard micro lithium-polymer battery (100 mAh, 3.3 V). This integrated connection design ensures robust system stability and durability during operation. The assembled FPCB demonstrates remarkable flexibility and bendability, as shown in Fig. 3c, allowing it to be easily folded and stored in confined spaces. Compared to conventional rigid circuit boards, the FPCB is significantly lighter and thinner, effectively minimizing the overall weight of the device and making it particularly suitable for mouse ECG monitoring applications. To preserve signal fidelity, a well-established ECG acquisition scheme utilizing the BMD101 chip developed by NeuroSky is adopted, as illustrated in Fig. 3d, e, which is specifically designed for biomedical and neurophysiological applications. The BMD101 integrates an advanced analog front end with powerful digital signal processing capabilities. It features a built-in DSP that accurately captures and extracts cardiac electrical activity. The analog ECG signals are first processed through a differential amplifier, followed by a notch filter and a low-pass filter to effectively eliminate noise and interference while retaining the physiologically relevant components. Additionally, the chip performs real-time compression of the digitized ECG data, reducing the demand for data transmission and storage. The digitized data is further processed using a notch filter (50 Hz or 60 Hz) to eliminate power line interference and a low-pass filter with a cut-off frequency of 100 Hz to refine the signal quality. The processed data is communicated to the low-power Bluetooth module (CC2640R2F) via a UART interface (TX/RX). The CC2640R2F module, based on the ARM Cortex-M3 processor core, supports Bluetooth Low Energy (BLE) and Bluetooth 5.1 protocols and integrates peripherals such as ADC, universal asynchronous receiver-transmitter (UART), and universal serial interfaces (SPI/I2C). Within the Bluetooth module, the ECG signals are compressed, packaged, and modulated before being transmitted wirelessly to mobile devices. The software interface, developed using Android Studio, enables real-time visualization and storage of mouse ECG signals on mobile devices (Fig. 3f). To further enhance signal quality, a Butterworth bandpass filter with a passband frequency range of 0.8 Hz to 45 Hz was applied to the acquired ECG signals. The Butterworth filter is a widely used digital filter in signal processing, particularly in biomedical applications such as electrocardiography (ECG) and electroencephalography (EEG). Its key features include a maximally flat frequency response in the passband, which ensures minimal amplitude distortion and allows the preservation of critical signal components. In addition, the Butterworth filter exhibits a near-zero phase distortion characteristic, meaning it does not introduce significant phase shifts during filtering, thereby maintaining the original temporal structure of the signal.
a Physical diagram of the FPCB. b 3D diagram of the FPCB. c Bending diagram of the FPCB. d Functional block diagram. e Circuit module block diagram. f Algorithm flowchart
To provide a more comprehensive benchmarking, we have included an expanded comparison with recent flexible and dry electrode technologies for small animal ECG monitoring. A new comparative performance summary table has been added to the Supplementary Information (Tables S5 and S6). The mobile application collects raw ECG data via Bluetooth, processes it through pre-defined algorithms, and generates graphical representations for user interpretation. During operation, the mouse is fitted with the ECG monitoring system. Users can identify and pair the corresponding Bluetooth device via the mobile application, which then automatically establishes a data communication channel. Real-time ECG signals and heart rate data are continuously collected, processed, and securely stored, facilitating accurate medical diagnostics and preventive healthcare analysis. This lightweight, flexible, and fully integrated mouse-wearable ECG system was established for real-time, long-term physiological monitoring. Combining high-performance electrodes with signal acquisition, data display, and ergonomic design, the system enables reliable electrophysiological recording, offering strong potential for preclinical cardiovascular research and precision medicine.
Design of a wearable mouse ECG monitoring system
Figure 4a illustrates the wearable mouse ECG monitoring system, which comprises a vest and a backpack. The backpack is securely attached to the vest using Velcro straps, ensuring a stable connection at the mid-scapular level of the mouse. The FPCB is housed within the backpack, providing a compact and lightweight design optimized for minimal interference with the mouse’s natural movements. Flexible epidermal electrodes are strategically placed on the left forelimb, right forelimb, and right hindlimb of the mouse, serving as measurement electrodes and a ground electrode. These electrodes are interconnected with the FPCB using flexible conductive fabric, ensuring robust and reliable electrical signal transmission. As shown in Fig. 4b, the ECG backpack, when not worn, highlights its key role in the system by preventing accidental pulling or electrode displacement, which could otherwise result in signal distortion or positional misalignment. The vest design includes two openings for the forelimbs, allowing a secure and comfortable fit. The central Velcro strap ensures the backpack remains properly positioned during data acquisition. This lightweight and ergonomic design minimizes physical stress on the mouse and enhances overall system stability. To maintain structural integrity during fabrication processes such as photolithography, metal deposition, and water-transfer printing, the electrode edges were initially designed with wider structures; however, these non-functional peripheral regions were trimmed before use to enhance stretchability without compromising signal quality, as confirmed by comparative performance tests.
a Design diagram of the ECG backpack. b Photo of the mouse without the backpack. c Photos and details of the mouse wearing the backpack. d Overall testing scenario
Prior to equipping the mouse with the system, the animal undergoes brief anesthesia to reduce stress and facilitate proper placement. The mouse is then positioned supine on a heated experimental platform, allowing for secure attachment of the epidermal electrodes to the limb surfaces, ensuring optimal skin contact and signal quality. A small amount of glycerol–water solution was applied during electrode attachment, enabling temporary adhesion through capillary force, which becomes stable after solvent evaporation. Following electrode placement, the mouse is laid flat, the vest is carefully fitted, and the FPCB-containing backpack is securely attached to the vest using the Velcro fasteners. Bite resistance is especially ensured via ultrathin conformal design, robust PI-based encapsulation, and dorsal-mounted configuration, minimizing exposure and mechanical damage during long-term use. The final wearable configuration is depicted in Fig. 4c. The experimental environment (Fig. 4d) consists of a clean, softly illuminated, and noise-free laboratory setting to minimize external interference during data acquisition. Both anesthetized mice and freely moving awake mice are placed in a transparent enclosure, enabling unobstructed monitoring. Upon successful Bluetooth pairing between the ECG monitoring system and a mobile terminal, the system establishes a real-time data transmission channel. ECG signals and heart rate data are continuously acquired, processed, and displayed on the mobile application interface, with all data securely stored for subsequent analysis and diagnostic evaluation. After that, the ECG signals during resting and running are collected to confirm signal stability and assess motion tolerance in real scenarios, as shown in supporting Fig. S2B.
The functional testing of epidermal electrodes
In this study, the Lead II configuration was selected (right forelimb: green, left hindlimb: red, right hindlimb: black) for ECG signal acquisition in mice. For the needle electrode setup, three silver needle electrodes were inserted subcutaneously into the ankle regions, ensuring secure electrode-skin contact while avoiding muscle penetration to minimize potential electromyographic signal interference18. In contrast, epidermal electrodes were adhered directly to the skin surface at the left forelimb, right forelimb, and right hindlimb, ensuring tight conformal contact (Fig. 5a). ECG signals were continuously recorded using the RM2640E multichannel physiological signal acquisition system (Shanghai Yuyan, China). As illustrated in Fig. 5c–e, both needle electrodes and epidermal electrodes successfully captured clear ECG waveforms, with distinct QRS segments. These results demonstrate that epidermal electrodes are equally effective as traditional needle electrodes for ECG monitoring, validating their signal acquisition reliability and functional feasibility. In addition, compared to the needle electrodes, the epidermal electrodes demonstrated a 7.7 dB higher signal-to-noise ratio (SNR) and a 12.45% increase in amplitude. To further evaluate the anti-interference capability of the electrodes under dynamic conditions, a stress response was induced in mice by spraying water droplets on their faces, simulating sudden external stimulation scenarios (Fig. 5f, g). During this stress test, epidermal electrodes demonstrated superior signal stability and resistance to motion artifacts compared to needle electrodes. The ECG waveforms remained clear and analyzable, highlighting the robust anti-interference performance of epidermal electrodes under motion-induced noise. Overall, these findings confirm that epidermal electrodes offer reliable and stable ECG signal acquisition, even in dynamic environments. Their mechanical compliance, signal fidelity, and anti-interference performance make them particularly suitable for long-term cardiac monitoring applications, providing a solid foundation for future biomedical sensor developments.
a Electrode placement method for ECG signal acquisition in mice. b Pictures of the mouse’s leg after using two types of electrodes. c Mouse ECG signals tested using two electrodes. d Test of interference immunity of an ECG acquisition system using epidermal electrodes. e Test of interference immunity of an ECG acquisition system using needle electrodes. f Test of interference immunity of an ECG acquisition system using needle electrodes. g Test of interference immunity of an ECG acquisition system using epidermal electrodes
As shown in Fig. 5f, g, the epidermal electrodes demonstrated a significantly shorter signal recovery time (0.28 s) compared to needle electrodes (0.35 s) following external interference. Moreover, while the baseline of the needle electrode signals displayed noticeable instability post-recovery, the epidermal electrodes rapidly restored their original waveform characteristics, indicating higher signal stability and accuracy. These results highlight the superior anti-interference performance of epidermal electrodes, attributed to their flexibility and conformal contact, which enable them to adapt effectively to skin deformations and minimize motion artifacts during signal acquisition. In long-term monitoring scenarios, the invasive nature of needle electrodes, which require skin and muscle penetration, can lead to tissue damage, inflammation, or irritation, potentially compromising the reliability of experimental results (Fig. 5b). Additionally, the performance of needle electrodes is highly dependent on operator skill, and improper handling may result in unstable or inconsistent data quality. In contrast, epidermal electrodes, as a non-invasive alternative, eliminate the need for skin penetration, effectively reducing tissue damage, minimizing animal discomfort and stress responses (Fig. 5b).
To verify the stability of the signal during an extended measurement process, we conducted a 30-minute ECG monitoring (Fig. S2A). Throughout the monitoring period, no significant degradation in signal quality was observed. In addition, we recorded the ECGs of mice under different physiological states, including resting and walking (Fig. S2B). Compared to the resting state, the signal amplitude decreased by 40%, and the SNR dropped by 8.45 dB during movement. Nevertheless, the SNR remained at an acceptable level of 20.4 dB, demonstrating that the ECG monitoring system effectively preserves the quality of physiological signals even under dynamic conditions.
To facilitate fabrication, the electrodes were designed with relatively wide edges and without serpentine structures. To determine whether the edge region affects the overall signal integrity, ECG signals were collected both before and after trimming the electrode edges (Fig. S3A, B). The results showed that clear ECG waveforms were obtained in both cases, with no significant differences in SNR or amplitude, indicating that the edge structure does not compromise signal quality (Fig. S3C). Additionally, as a demonstration, we continuously monitored the ECG of the mouse during surgery and collected its heart for subsequent experiments (Fig. S3D–F). The red arrow indicates the start of the surgery, during which the mouse’s ECG underwent significant changes (Fig. S3D–F). As the surgery progressed, the ECG signal fluctuated continuously until the mouse’s heart was removed, at which point the signal became a straight line (Fig. S3D–F). In summary, epidermal electrodes demonstrate significant advantages in terms of anti-interference performance, biocompatibility, and signal stability, proving their feasibility and reliability for long-term ECG monitoring in animals.
Long-term stability and biocompatibility evaluation of the ECG monitoring system
For long-term ECG monitoring, maintaining signal quality poses a significant challenge. In this experiment, the feasibility of wearing the proposed ECG system was verified for prolonged wear time of up to 5 days, which has been the suggested period for increasing the detection rate of subclinical heart disease (e.g., arrhythmias and atrial fibrillation). The system was implemented in mice, allowing them to engage freely in activities like eating, moving about, and sleeping. The ECG waveforms measured at Day 1 and Day 5 showed slight degradations in the amplitude (22.5%) and the SNR (8.04 dB), which was caused by accumulated dead cells on the epidermis layer (Fig. 6a). Owing to its ultrathin, hollow-carved architecture that permits direct exposure of skin pores, the design facilitates efficient vapor diffusion across the electrode interface, thereby mitigating the risks of pruritus, allergic responses, and erythema.
a ECG waveforms recorded on Day 0, Day 1, Day 3, and Day 5. b Serum biochemical indices (ALT, AST, BUN, and CRE). c, d Hematoxylin and Eosin (H&E) stain present in ×400 magnification. ns not significant
To further evaluate the biocompatibility of the epidermal electrodes, the epidermal electrodes applied to the legs of mice, and the key blood biochemical markers, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatinine (CRE), and blood urea nitrogen (BUN), were measured after 5 days (Fig. 6b). As shown in Fig. 6b, no significant alterations in the evaluated biochemical indicators were observed in the group with epidermal electrodes compared to the control. Furthermore, hematoxylin and eosin (H&E) staining of major organs–including the heart, liver, spleen, lung, kidney, and hind limb muscle–revealed no discernible histopathological abnormalities (Fig. 6c). These results indicate that the epidermal electrodes possess excellent biosafety. The good biocompatibility of the epidermal electrodes contributes to improved skin comfort in mice during long-term physiological signal monitoring.
Psychological stress indicators in mice following ECG monitoring system wear
To evaluate the potential psychological stress induced by the use of an ECG monitoring system in mice, an open field test was performed to assess key behavioral parameters, including total locomotor distance, movement velocity, number of center zone entries, and time spent in the center (Fig. 7a–f). These indicators were used to compare spontaneous locomotor activity and exploratory behavior between the test group and the control group. The results showed that mice in the test group exhibited a slight reduction in overall activity compared to the control group; however, the difference was not statistically significant. This suggests that, following an acclimation period, the mice were able to tolerate the presence of the ECG monitoring system without marked behavioral alterations.
a, b Trajectory maps of mice during the open field test. c Total locomotor distance. d Average movement velocity. e Number of entries into the center zone. f Duration spent in the center zone. g–i Serum levels of corticosterone (CORT), interleukin-1β (IL-1β), and tumor necrosis factor-α (TNF-α). ns not significant
Corticosterone (CORT) is a well-established biomarker of chronic psychological stress in rodents, often associated with conditions such as anxiety and learned helplessness. Chronic stress has also been implicated in the activation of neuroinflammatory pathways, leading to elevated levels of pro-inflammatory cytokines, such as interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α). To further investigate potential physiological stress responses, serum samples were collected after five days of continuous ECG system wear, and the concentrations of CORT, IL-1β, and TNF-α were measured (Fig. 7g–i). Results showed a slight increase in these markers in the test group relative to controls, but the differences were not statistically significant (Fig. 7g–i). These findings collectively suggest that prolonged use of the ECG monitoring system does not elicit a significant chronic stress response at the behavioral or molecular level in mice.
Results of ECG signal testing in mice model of myocardial infarction
Myocardial infarction, caused by ischemic arterial occlusion, is characterized by an acute onset, complex complications, and high mortality rates. Early real-time ECG monitoring and feedback therapy are essential for effective disease management19,20,21,22. In this study, an ISO-induced myocardial infarction mouse model was employed to evaluate the accuracy and reliability of epidermal electrodes in ECG signal acquisition. As shown in Fig. 8a, b, ECG signals were synchronously recorded using epidermal electrodes and needle electrodes, and the waveforms from both methods exhibited high similarity. Further comparison between normal and MI groups (Fig. 8c, d) revealed significant ST-segment elevation and baseline irregularities in the ISO group, consistent with clinical diagnostic criteria for myocardial infarction. To verify the pathological characteristics of the ISO model, H&E staining and Masson staining were performed on cardiac tissue sections. H&E staining (Fig. 8e) showed that in the normal group, myocardial cells were orderly arranged, with small intercellular spaces and intact cell morphology. In contrast, the ISO group exhibited fiber rupture, disorganized alignment, nuclear fragmentation, and uneven inflammatory cell infiltration. The results of Masson staining (Fig. 8f) further confirmed these pathological changes. In the normal group, myocardial cells displayed orderly alignment with minimal collagen deposition. However, in the ISO group, there was disorganized myocardial alignment, significant collagen accumulation, and uneven distribution in the extracellular matrix. These combined findings from H&E and Masson staining validate the successful establishment of the ISO mouse model and demonstrate the accuracy and reliability of epidermal electrodes in ECG signal acquisition. This study provides strong experimental evidence supporting the potential application of epidermal electrodes in real-time cardiac disease monitoring.
a, b Testing electrocardiographic signals in mice with myocardial infarction using two types of electrodes. c, d Monitoring electrocardiographic signals in normal and myocardial infarction mice using epidermal electrodes. e Hematoxylin and Eosin (H&E) stain present in ×400 magnification. f Masson Staining present in ×400 magnification
Materials and methods
Preparation of metal layers
PVA (USOLF, CHN, 1788 L) powder and PI (MXJYCL, CHN) film were obtained from commercial suppliers. The PVA solution was prepared by dissolving PVA powder in deionized water at a 1:10 mass ratio. The PVA solution was first spin-coated onto a cleaned glass substrate, followed by 5 μm-thick PI film on top. The substrate was then heated to remove residual solvent. Using magnetron sputtering, a 70 nm titanium (Ti) layer was initially deposited on the PI film, followed by a 350 nm gold (Au) layer. Subsequently, a positive photoresist was spin-coated onto the resulting composite and baked. A photomask was employed to define the metal pattern through exposure and development, and the sample was immersed in an etching solution consisting of 4 g potassium iodide, 1 g iodine, and 40 mL water to remove the unpatterned regions of the metal layer. Afterwards, plasma etching was performed to remove the remaining photoresist and the PI film. Finally, by simply wetting the epidermal electrode, the device was transferred onto another substrate without damage. A serpentine island-bridge architecture was selected in this work.
Seamless transfer of the epidermal electrode
Seamless transfer of the epidermal electrode is achieved by exploiting the water solubility of the cured PVA layer. Upon wetting the electrode, the PVA dissolves in water, allowing the device to be transferred to a three-dimensional substrate by initially placing the wetted electrode onto the surface of a balloon (serving as an intermediate substrate), and then moving it onto an Ecoflex substrate. Alternatively, it may be transferred to a two-dimensional substrate by simply wetting and removing it from the glass and attaching it to a PU film. Throughout the transfer process, the epidermal electrode remains wrinkle-free and undamaged, with the PI layer oriented toward the receiving substrate.
Animals and experimental design
About 10-week-old male mice were purchased from Pengyue Experimental Animal Breeding Co., Ltd (Jinan, Shandong, China) [SCXK (Lu) 20220006]. All animals were housed under standard barrier conditions with controlled temperature (21 °C–24 °C) and humidity (50%–65%) on a 12-h light/dark cycle and had free access to a standard rodent diet and water. All animal care and experimental procedures were conducted according to the Guide for the Care and Use of Laboratory Animals and approved by the Animal Care and Use Committee of the affiliated hospital of Qingdao University. The mice in the ISO group were subcutaneously injected with ISO (100 mg/kg, I5627, purchased from Sigma-Aldrich, United States) for 2 consecutive days to induce experimental myocardial infarction, as reported previously23,24. In addition, the mice in the control group were subcutaneously injected with normal saline for 2 days. After the last ISO injection, the ECGs of mice were collected, and the animals were then euthanized for the collection of the hearts.
Mouse ECG signal
The mice were anesthetized by inhaling 1% to 3% isoflurane vapor at an oxygen flow rate of 2 ml/min before and during the experiment. The mice were transferred to a clean, softly lit, and noise-free laboratory after confirmation of loss of consciousness. The needle electrodes group of mice were placed in a supine position on a 38 °C insulation experimental platform, and three silver needle electrodes were inserted subcutaneously into the mice’s ankles18. Paws were affixed to the insulation experimental platform with epidermal electrodes and tape to collect ECGs in the epidermal electrodes group. Continuous 24-hour ECG data were recorded using the multi-channel physiological signal acquisition and processing system (YAN6000, Yuyan, China). The data were analyzed with origin.
Histopathological examination
After sacrifice, the hearts were collected, rinsed with ice-cold saline, and preserved in a 10% buffered formalin solution. Subsequently, the hearts underwent processing using incremental alcohol concentrations before being embedded in paraffin. Sections of 10 µm thickness were then cut, followed by deparaffinization, hydration, and staining with H&E or Masson trichrome. The stained sections were observed using the light microscope (Olympus, Japan) and examined in a blind fashion.
Masson trichrome collagen staining
The prepared tissue sections were baked in a 60 °C oven for 1 hour, followed by deparaffinization in xylene (initial immersion for 10 minutes, then replaced with fresh xylene and immersed for an additional 5 minutes). Subsequently, the sections were sequentially immersed in 100%, 95%, 90%, and 85% ethanol for 5 minutes each, followed by rinsing with tap water for 5 minutes. Masson staining solution, phosphomolybdic acid, and aniline blue were applied for 5 minutes each. Differentiation was performed by adding the differentiation solution for 30–60 seconds, followed by rinsing with distilled water. The sections were then dehydrated in 95% and 100% ethanol, cleared with xylene, and mounted with neutral balsam. Finally, images were acquired under a microscope (Olympus, Japan).
Biochemical assays
Serum alanine ALT and AST levels were measured using an Autodry Chemistry Analyzer.
ELISA
The concentrations of serum CRE, BUN, CORT, IL-1β, and TNF-α were measured using the ELISA kits (R&D Systems, Minneapolis, MN, USA and BD OptEIA™, New Jersey, USA) according to the manufacturers’ instructions.
Discussion and conclusion
This study presents an ultrathin, highly flexible epidermal electrode fabrication process that is scalable, cost-effective, and enables one-step mass production, demonstrating successful application in ECG signal acquisition from mice. To address existing challenges in cost, substrate transfer, and conformal contact, an efficient MEMS-based batch manufacturing method was introduced, enabling rapid large-scale production of stretchable, multifunctional epidermal electrodes suitable for diverse applications. Using FEM and mechanical analysis, the structural parameters were optimized. Results showed that stretchability is positively correlated with periodic length and film thickness but negatively correlated with array unit line width. The optimized serpentine island-bridge structure was defined with a line width of 0.1 mm, a periodic length of 1.3 mm, and a 5-μm-thick PI film as the substrate. Experimental results demonstrated that the epidermal electrodes maintained tight conformal contact with dynamic and curved mouse skin, exhibited excellent stretchability, low contact impedance, and no skin irritation. Even under 40% tensile strain, the electrodes retained stable electrical performance. The mouse ECG monitoring system features a compact and wearable design, allowing the mouse to carry a small backpack with encapsulated soft circuits. This setup enables accurate and long-term ECG signal acquisition during unrestricted movement. A non-invasive ECG monitoring method using epidermal electrodes was validated against traditional needle electrodes. Both electrodes captured clear QRS waveforms with comparable signal amplitudes (Fig. 9). In a myocardial infarction mouse model, the epidermal electrodes detected ST-segment elevation and baseline irregularities, aligning with clinical diagnostic features, confirming their accuracy and reliability in animal ECG monitoring. This technology provides robust technical support and theoretical foundations for ECG signal acquisition in mice and cardiac disease research.
The proposed system is mounted on the back of the mouse to collect ECG signals
Although the epidermal electrode technology proposed in this study has demonstrated significant success in ECG signal acquisition from mice, it also presents several avenues for future improvement. For instance, in our current experiments, the system successfully captures distinct heart rate changes in mice under different behavioral states, such as rest and locomotion. Although the system has not yet been applied to models of arrhythmia or HRV-related disorders, these results demonstrate its technical capability and suggest strong potential for future expansion into more nuanced cardiac monitoring applications. After that, conducting large-scale, long-term animal studies to evaluate the performance of the epidermal electrodes across diverse physiological states and disease models will be crucial in assessing their broader applicability. The integration of artificial intelligence and machine learning techniques into the electrode monitoring system could further improve its capabilities by enabling automated data analysis, early disease detection, and personalized health monitoring. These advancements would pave the way for the development of more versatile epidermal electrodes that seamlessly combine both monitoring and diagnostic functions, broadening their potential for clinical and research applications.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
Ha, T. W., Oh, B. & Kang, J. O. Electrocardiogram recordings in anesthetized mice using lead II. J. Vis. Exp. https://doi.org/10.3791/61583 (2020).
Ho, D. et al. Heart rate and electrocardiography monitoring in mice. Curr. Protoc. Mouse Biol. 1, 123–139 (2011).
Korte, T. et al. In-vivo electrophysiological study in mice with chronic anterior myocardial infarction. J. Inter. Card. Electrophysiol. 6, 121–132 (2002).
Calvet, C. & Seebeck, P. What to consider for ECG in mice-with special emphasis on telemetry. Mamm. Genome 34, 166–179 (2023).
Lux, R. L. Uncertainty of the electrocardiogram: old and new ideas for assessment and interpretation. J. Electrocardiol. 33, 203–208 (2000). Suppl.
Cesarovic, N., Jirkof, P., Rettich, A. & Arras, M. Implantation of radiotelemetry transmitters yielding data on ECG, heart rate, core body temperature and activity in free-moving laboratory mice. J. Vis. Exp. 21, 3260 (2011).
Kim, C. et al. A wearable ECG monitoring system using microneedle electrodes for small animals. IEEE Sens. J. 23, 21873–21881 (2023).
Wright, J. P. et al. A fully implantable wireless bidirectional neuromodulation system for mice. Biosens. Bioelectron. 200, 113886 (2022).
Mishra, V., Gautier, N. M. & Glasscock, E. Simultaneous video-EEG-ECG monitoring to identify neurocardiac dysfunction in mouse models of epilepsy. J. Vis. Exp. 29, 57300 (2018).
Sato, M. et al. Simultaneous monitoring of mouse respiratory and cardiac rates through a single precordial electrode. J. Pharm. Sci. 137, 177–186 (2018).
Kabiri Ameri, S. et al. Graphene electronic tattoo sensors. ACS Nano 11, 7634–7641 (2017).
Yokus, M. A. & Jur, J. S. Fabric-based wearable dry electrodes for body surface biopotential recording. IEEE Trans. Biomed. Eng. 63, 423–430 (2015).
Zhou, A. et al. Water-in-salt electrolyte promotes high-capacity FeFe (CN) 6 cathode for aqueous Al-ion battery. ACS Appl. Mater. Interfaces 11, 41356–41362 (2019).
Lin, X. et al. Self‐adhesive and biocompatible dry electrodes with conformal contact to skin for epidermal electrophysiology. Interdiscip. Mater. 3, 775–790 (2024).
Hocheng, H. & Chen, C. M. Design, fabrication and failure analysis of stretchable electrical routings. Sensors 14, 11855–11877 (2014).
Kim, D. H. et al. Epidermal electronics. Science 333, 838–843 (2011).
Kim, D. H. et al. Stretchable and foldable silicon integrated circuits. Science 320, 507–511 (2008).
Boukens, B. J., Rivaud, M. R., Rentschler, S. & Coronel, R. Misinterpretation of the mouse ECG: ‘musing the waves of Mus musculus. J. Physiol. 592, 4613–4626 (2014).
Yu, C. et al. Chronological adhesive cardiac patch for synchronous mechanophysiological monitoring and electrocoupling therapy. Nat. Commun. 14, 6226 (2023).
Schiavone, G. & Lacour, S. P. Conformable bioelectronic interfaces: mapping the road ahead. Sci. Transl. Med. 11, eaaw5858 (2019).
Choi, Y. S. et al. Fully implantable and bioresorbable cardiac pacemakers without leads or batteries. Nat. Biotechnol. 39, 1228–1238 (2021).
Li, Y., Li, N., De Oliveira, N. & Wang, S. Implantable bioelectronics toward long-term stability and sustainability. Matter 4, 1125–1141 (2021).
Yang, Y. et al. Ciprofol attenuates the isoproterenol-induced oxidative damage, inflammatory response and cardiomyocyte apoptosis. Front. Pharm. 13, 1037151 (2022).
Wang, H. et al. Curdione inhibits ferroptosis in isoprenaline-induced myocardial infarction via regulating Keap1/Trx1/GPX4 signaling pathway. Phytother. Res. 37, 5328–5340 (2023).
Acknowledgements
This work was supported by the National Key R&D Program of China (no. 2024YFB3214000), National Natural Science Foundation of China (no. 62371275, 62201328 and 62401345), Natural Science Foundation of Shandong Province (no. ZR2023MF009, ZR2021QH221 and ZR2021QF121), Natural Science Foundation of Qingdao (no. 23-2-1-155-zyyd-jch, 24-4-4-zrjj-95-jch and No.24-4-4-zrjj-98-jch), Industry Cluster Leading Program of Qingdao (no. 25-1-1-gjgg-18-gx).
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Ethics declarations
This study was approved by the Animal Care and Use Committee of the affiliated hospital of Qingdao University (no. 20250103YYL). All animal care and experimental procedures were conducted according to the Guide for the Care and Use of Laboratory Animals.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Yao, Y., Pan, C., Li, J. et al. A non-invasive, flexible, lightweight, and wearable system for real-time ECG monitoring in freely moving mice. Microsyst Nanoeng 11, 218 (2025). https://doi.org/10.1038/s41378-025-01013-2
Received:
Revised:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41378-025-01013-2
