The appeal of liquid metals begins with a basic physical paradox. Conventional stretchable conductors rely on rigid fillers, such as silver flakes or carbon particles, embedded in an elastic matrix. Increasing the filler loading improves conductivity but stiffens the composite and accelerates fatigue under repeated strain; reducing the loading preserves softness but degrades electrical performance. This conductivity–softness trade-off has constrained generations of wearable and implantable devices. Liquid metals, most notably eutectic gallium–indium alloys that remain molten at room temperature, sidestep the problem entirely because the conductive phase is itself a fluid. Droplets of the alloy can be suspended in elastomers, where they deform, elongate and reconfigure with the surrounding polymer rather than resisting it, maintaining metallic properties across large strains.

The review emphasizes that liquid metal composites deliver far more than stretchable wiring. When droplets are dispersed through a soft polymer, they enhance not only electrical conductivity but also dielectric and thermal properties. Elongated liquid metal inclusions, for example, have been shown to raise the thermal conductivity of elastomers dramatically, enabling soft materials that can dissipate heat, and liquid metal inclusions have been used to create high-permittivity dielectric elastomers for capacitive and actuating applications. In one striking demonstration highlighted in the review, an electrically insulating liquid metal silicone compound combined electromagnetic interference shielding with direct electronic potting and thermal management, showing that a single soft composite can perform multiple functions once reserved for rigid metal housings.

Yet the same fluidity that gives liquid metals their conformability creates their greatest engineering headaches. A liquid conductor can leak, smear, delaminate or migrate within a device, destroying pattern fidelity and reliability. The authors argue that the decisive design variable is not the liquid metal itself but the interfaces it forms, both between liquid metal and the surrounding polymer and between liquid metal inks and the substrates and components they must connect. Interface engineering, they contend, is essential for mitigating leakage, delamination and functional degradation under deformation, and tailored interfacial design is what converts a promising ink into a robust, integrated device.

A rich toolkit of surface chemistry has emerged to control these interfaces. Gallium-based alloys spontaneously form a thin native oxide skin that dramatically alters their interfacial energy, lowering surface tension enough to allow the metal to be shaped, printed and stabilized as droplets. Researchers have exploited this oxide, and chemical modifications of it, to stabilize particles, tune wettability and promote adhesion. Polyphenol-induced adhesives have produced liquid metal inks that write directly on nearly any substrate, while thiol coordination chemistry has been used to soften liquid metal particles and switch their conductivity on demand. Hydrogen-doped viscoplastic microparticles have enabled stretchable printed metal lines, and finite-gel strategies have produced highly concentrated inks suitable for direct writing.

Patterning at high resolution remains one of the field’s defining challenges, and the review surveys how recent advances in droplet stabilization and ink formulation are addressing it. Ultrasonic sintering has been used to assemble liquid metal particles within polymers into elastic printed circuit boards, in which pressing or sonicating the particles ruptures their oxide shells and merges them into continuous, leakage-resistant traces. Laser sintering of liquid metal nanoparticles offers a scalable route to soft, flexible electronics, and projection lithography has enabled rapid printing of highly stretchable circuits. Meniscus-guided printing has produced stable semi-solid microgranular particles for soft electronics, while high internal phase emulsion inks have allowed direct-ink-writing of three-dimensional liquid metal structures. One recent metalgel formulation maintained a conductivity of three million siemens per metre through a million stretching cycles, a durability benchmark that would have seemed unattainable for a fluid conductor only a decade ago.

The applications catalogued in the review span the full breadth of soft electronics. In wearable healthcare, liquid metal electrodes and sensors have been fashioned into electronic tattoos, electromyography patches for tendon localization and muscle injury prevention, and pressure sensors for health monitoring. In implantable and bioelectronic systems, liquid metal microelectrodes have been wafer-patterned for chronic biocompatibility, printed directly on the cranium for soft neural probes, and formed into three-dimensional microelectrode arrays integrated with ultrathin retinal prostheses for vision restoration; liquid metal neuro-interfaces have even been used to stimulate and record from human hippocampal organoids. Phase-transition behavior adds another dimension: temperature-responsive intravenous needles that irreversibly soften on insertion, and biphasic microfibers for minimally invasive implantable electronics, exploit alloys that are rigid at room temperature and compliant at body temperature.

Soft robotics and integrated systems feature prominently as well. Liquid metal–elastomer composites have demonstrated autonomous electrical self-healing, in which damaged circuits reconnect themselves as the liquid metal flows back together, and liquid metal inclusions have enabled shape-morphing elastomers that couple thermal transport with programmable actuation. Stretchable Arduinos embedded in soft robots, octopus-inspired sensorized arms, and stretchable core–shell cables that are patternable, recyclable and noise-resistant all illustrate how liquid metals can bridge soft and rigid electronic domains. Liquid crystal elastomer composites actuated by induction heating point toward untethered, programmable soft machines, while liquid metal antennas exploit fluidic deformability for reconfigurable radio-frequency devices.

The authors are candid about what stands between laboratory demonstrations and real-world deployment. Long-term interface durability and manufacturing precision remain the key translation challenges: devices must survive millions of deformation cycles in wet, biological environments without leaking, delaminating or drifting in performance, and patterning methods must reach the resolution and throughput that industrial electronics demand. Questions of biocompatibility are being actively studied, with gallium’s known antimicrobial activity offering intriguing therapeutic possibilities even as toxicity profiles continue to be mapped. The review’s central message is nonetheless one of momentum: by treating the liquid metal–material interface as the primary design object, rather than an afterthought, researchers now have a coherent strategy for building soft electronics that no longer force a choice between how well a device performs and how gracefully it bends. If that strategy matures, the circuits of the future may not merely sit on the body but move with it, heal with it and, in the case of implants, dissolve the boundary between machine and tissue altogether.

Subject of Research: Design strategies for liquid metal-based soft and stretchable electronics

Article Title: Design strategies for liquid metal-based soft electronics

Article References: Pan, C., Zhang, C., Wang, J., Han, M., CHIU, W. Y. P., Zhao, P., & Dickey, M. D. (2026). Design strategies for liquid metal-based soft electronics. Nature Reviews Electrical Engineering. https://doi.org/10.1038/s44287-026-00335-1

Image Credits: AI Generated

DOI: 10.1038/s44287-026-00335-1

Keywords: liquid metals, stretchable electronics, soft electronics, gallium alloys, interface engineering, wearable devices, implantable bioelectronics, soft robotics, elastomer composites, self-healing materials, printed electronics, dielectric elastomers

The core difficulty is familiar to anyone who has wrestled with wireless phone chargers. Conventional inductive coupling depends on two coils being tuned to the same resonant frequency, like two tuning forks sharing a pitch. When the coils are perfectly aligned and the load is steady, energy flows efficiently across the gap. But inside a body, nothing stays still. A patient moves, breathes, and shifts posture; the implant stretches with surrounding tissue; the electrical impedance at the implant’s output changes as it stimulates muscle. Each of these disturbances shifts the resonance of the system, and once transmitter and receiver drift out of tune, power transfer efficiency collapses. The result is an implant that works beautifully on the bench but fails unpredictably in the clinic, a problem engineers have documented since early analyses of radio-frequency coils in implantable devices in the 1980s.

The Korean team’s solution attacks the problem on two fronts simultaneously: the circuit architecture and the materials. On the circuit side, they built the power link around a nonlinear parity-time symmetric circuit, an approach first demonstrated for robust wireless power transfer by researchers at Stanford University in 2017. In a parity-time symmetric system, the transmitter and receiver form a symmetric resonator pair, and the transmitter continuously adjusts its own operating frequency to match whatever frequency the receiver settles at. Instead of demanding that the two coils stay perfectly tuned, the system lets the receiver’s resonance wander and follows it in real time. A feedback circuit in the wearable transmitter detects changes and provides automatic frequency adaptation, so that misalignment, mechanical deformation, or load impedance fluctuations no longer break the resonance that carries the power.

On the materials side, the implantable receiver is built from liquid metal. Rather than patterning the receiver’s coil and interconnects from rigid copper, the researchers used liquid metal conductors encapsulated in soft elastomers, fabricated through a multistep process involving photolithographically defined copper traces that guide liquid metal deposition, vertical interconnect vias etched through insulating elastomer layers, and off-the-shelf electronic components integrated onto liquid metal pads. Because the conductors are intrinsically stretchy, the receiver’s electrical resistance barely changes when the device deforms. That matters because resistance-induced losses are what erode efficiency in stretchable electronics: a conventional serpentine metal trace stretches by uncoiling and thinning, raising resistance and shifting the coil’s electrical properties. A liquid metal channel simply changes shape while keeping its cross-section and conductivity largely intact, minimizing resistance-induced power losses during deformation.

The performance numbers reported in the paper are striking. The system maintains power transfer efficiency above 50 percent even when the receiver is stretched by 30 percent or displaced laterally by 30 millimeters from the transmitter. For context, conventional inductively coupled systems can lose the bulk of their efficiency at a fraction of that misalignment. The team also tested the system under bending, tilting, and rotational misalignment, tracking both operating frequency and transfer efficiency through each disturbance, and compared the parity-time symmetric architecture against a conventional negative-impedance-converted system in simulations and in live animals. In those comparisons, the conventional system’s efficiency maps showed sharp drop-offs as separation and strain increased, while the parity-time symmetric system held a broad, stable operating range. A radar chart comparison against previously reported stretchable wireless bioelectronic devices, scored on efficiency, alignment tolerance, strain insensitivity, stretchability, and conductivity, placed the new system at the most balanced and superior overall performance among state-of-the-art devices.

The demonstration that turns this from an elegant circuit exercise into a potential medical technology is cardiac pacing. The researchers packaged the liquid metal receiver as a wireless pacemaker with electrodes made from a silver-gold nanowire composite embedded in an elastomer, designed to be sutured onto the surface of the heart. The wearable transmitter, powered by an 11-volt lithium-ion battery and managed by a microcontroller-controlled power unit, was mounted on the outside of the body. The received power is rectified on the implant and used to deliver controlled electrical stimulation to the cardiac tissue. Because the entire power link tolerates the constant motion of a large animal, the pacing could continue under highly dynamic in vivo conditions that would destabilize a conventional link.

In experiments, the team first validated the system in rabbit models, implanting the liquid metal receiver subcutaneously and using an LED indicator on the receiver to visualize successful power transfer as the axial and lateral separation between transmitter and receiver coils increased. The parity-time symmetric system maintained consistent activation across the tested positions, whereas the conventional system failed to power the implant at increased distances and lateral misalignments. The team then moved to a porcine model, whose heart size and physiology are much closer to humans. Fully untethered pigs with wearable transmitters attached to their backs received wireless epicardial pacing, with electrodes sutured onto the heart surface. The system not only paced the heart reliably but also terminated tachyarrhythmias, dangerously fast heart rhythms, under those same dynamic conditions, demonstrating that the power link could support not just routine pacing but active intervention during cardiac events.

The control side of the system reflects a deliberate design for real-world use. The wearable power management unit regulates power from the battery and provides regulated outputs to the transmitter circuits through a microcontroller-controlled switch, and the microcontroller was programmed using the Arduino IDE, with source code available from the corresponding authors. Smartphone control was implemented using a commercially available Bluetooth Low Energy terminal application, meaning a clinician or patient could in principle adjust the system with ordinary consumer hardware rather than bespoke equipment. That kind of practical engineering detail, mundane as it sounds, often separates laboratory demonstrations from technologies that can actually be deployed and maintained outside a research environment.

The implications extend well beyond pacemakers. Fully implantable bioelectronic systems, from neuromodulation devices and brain-computer interfaces to bioresorbable stimulators and injectable sensors, all face the same power bottleneck, and many of the most exciting recent devices in the field, including millimetre-scale bioresorbable optoelectronic systems and programmable ultrasonic implants, depend on some form of wireless energy delivery. A power link that tolerates strain, misalignment, and load variation could serve as a general-purpose energy backbone for soft, body-conformal devices that move with organs rather than fighting them. The authors’ comparison framework, scoring devices on efficiency, alignment tolerance, strain insensitivity, stretchability, and conductivity, also gives the field a clearer yardstick for evaluating future designs.

Cautious optimism is warranted. The work was demonstrated in animal models, and the path to human devices will require the usual gauntlet of biocompatibility validation, long-term reliability testing, and regulatory review; the liquid metal, elastomers, and nanocomposite electrodes involved have strong precedents in the soft bioelectronics literature but must prove themselves in chronic implants. Still, the combination of a self-tuning nonlinear circuit and an intrinsically stretchable liquid metal receiver addresses the two failure modes, resonance drift and deformation losses, that have most reliably broken wireless implants. If the variation tolerance demonstrated in freely moving pigs carries through to clinical devices, the era of truly untethered, maintenance-free implantable medicine moves considerably closer, powered by a transmitter you wear and a receiver that bends with every beat of your heart.

Subject of Research: Variation-tolerant wearable-to-implant wireless power transfer for implantable bioelectronics such as wireless cardiac pacemakers

Article Title: A wearable-to-implant wireless power transfer technology with variation tolerance

Article References: Nam, S., Seo, T., Yoo, S., Park, C., Kim, T., Yu, M., Kim, Y., Kang, H., Yeom, D., Cho, Y., Cho, H., Lee, D., Kim, J. H., Sunwoo, S.-H., Lee, S., Moon, J., Lee, S.-P., Kim, S., & Kim, D.-H. (2026). A wearable-to-implant wireless power transfer technology with variation tolerance. Nature Electronics. https://doi.org/10.1038/s41928-026-01714-0

Image Credits: AI Generated

DOI: 10.1038/s41928-026-01714-0

Keywords: wireless power transfer, parity-time symmetry, liquid metal electronics, implantable bioelectronics, cardiac pacing, stretchable electronics, wearable devices, resonant inductive coupling, soft materials, biomedical engineering, Nature Electronics, tachyarrhythmia

Implantable electronics are usually hidden once surgeons place them beneath tissue. That makes it difficult to determine whether an implant has shifted, fractured, degraded prematurely or stopped working. In conventional systems, clinicians may need to rely on imaging procedures, indirect physiological measurements or exploratory surgery. Batteries add another layer of difficulty: they have finite lifetimes, can leak, occupy space and often contain components that remain in the body after the intended treatment has ended. A device designed to disappear could avoid a second operation, but only if doctors can determine when it is still functioning and how rapidly it is breaking down. The new implant addresses this monitoring problem by linking the device’s physical appearance to its electrical performance, creating an external visual readout of events normally concealed beneath the skin.

The researchers built the platform around two complementary technologies: triboelectric energy generation and optical phosphorescence. Triboelectric devices produce electrical charge through contact electrification and electrostatic induction. When two materials with different tendencies to gain or lose electrons interact, charge separation can occur. Mechanical motion then changes the arrangement of those charges, producing a usable alternating electrical signal. In this implant, ultrasound supplies the mechanical stimulus. Acoustic waves passing through tissue drive movement within the device, allowing it to generate electricity without an implanted battery or a direct wired connection. The electrical output is therefore activated from outside the body, on demand, rather than being continuously supplied by a stored power source.

The optical component relies on nanoscale iridium(III) complexes, described in the study as iridophosphors. These molecules emit phosphorescent light after absorbing energy from an external source. Unlike ordinary fluorescence, which generally stops almost immediately when excitation ends, phosphorescence can persist for longer periods because the excited state involves a transition that is less likely to release energy instantly. That persistence makes the material useful as a visual reporter: under handheld illumination, the implant produces a recognizable glowing outline that can be observed through the surrounding tissue. Crucially, the optical signal is not merely an identification tag. Because the phosphorescent material is integrated into the device’s structure, changes in the outline can reveal deformation, breakage, fragmentation or resorption.

Selecting a material capable of both emitting light and supporting triboelectric charge generation was a central challenge. The researchers used a data-driven materials-to-device workflow that combined machine-learning-assisted photophysical screening with a selection strategy based on molecular dipole moments. A molecule’s dipole moment describes how electric charge is distributed within it; larger or appropriately oriented dipoles can influence polarization and charge-transfer behavior, properties that are important for triboelectric performance. At the same time, the molecular structure must produce suitable optical behavior and remain compatible with a resorbable implant. By screening candidate iridium(III) compounds computationally and then selecting materials according to their photophysical characteristics and dipole moments, the team sought to optimize two functions at once rather than treating light emission and energy harvesting as separate engineering problems.

The resulting devices were designed to be read through the skin while also converting ultrasound into electrical energy. When illuminated externally, they produced outlines visible to the unaided eye under handheld light. Those outlines provided information about where the implant was located and what its overall morphology looked like. When ultrasound was applied, the triboelectric structure generated up to 3.6 volts peak-to-peak at 0.5 watts per square centimetre. Peak-to-peak voltage represents the difference between the highest positive and lowest negative portions of an alternating electrical waveform, rather than a steady direct-current voltage. The value indicates that the implant can generate a substantial transient electrical signal, although the practical usefulness of that output will depend on the electrical load, operating frequency, coupling through tissue and the requirements of any future therapeutic or sensing circuit.

The most important feature of the system was the relationship between what researchers could see and what the implant could do electrically. In mouse experiments, optical readouts tracked implant position and morphology, while visible structural defects were associated with a loss of electrical output. That correlation turns the implant’s shape into a functional diagnostic signal. A broken or distorted outline is not simply evidence of physical damage; it can indicate that the energy-harvesting mechanism has also been compromised. Such a relationship could help distinguish a device that has naturally resorbed as intended from one that has failed unexpectedly. It could also support timely intervention in applications where an implant is meant to deliver stimulation, power a temporary sensor or assist tissue healing for a defined period.

The researchers followed the implants in living mice for 38 weeks, using the optical signal to monitor integrity and bioresorption over time. Bioresorption means that the implanted material gradually breaks down and is cleared or assimilated by the body, eliminating the need for surgical retrieval. A transient device must balance two opposing requirements: it must remain stable and functional long enough to perform its medical role, yet eventually disappear without leaving a persistent foreign object. Monitoring that transition is difficult when the device is invisible and battery-free. The new approach provides a way to observe the changing implant directly through its emitted pattern, potentially revealing whether degradation is proceeding uniformly or whether structural damage has occurred earlier than expected.

The concept could be especially valuable for temporary medical technologies, including implants intended for short-term electrical stimulation, localized sensing or controlled therapeutic support. A battery-free design reduces the risks associated with leakage and removes one of the major limits on implant lifetime. Ultrasound provides a non-invasive route for delivering mechanical energy through tissue, while optical readout offers a comparatively simple way to inspect the device with external illumination. Yet the work remains an early-stage demonstration in mice. Human tissue is thicker and optically more complex, and the visibility of a phosphorescent signal will depend on implant depth, tissue composition and the intensity and wavelength of the excitation light. Future studies will also need to examine long-term biocompatibility, the fate of the iridium-containing compounds during resorption, manufacturing consistency and whether the generated electrical power is sufficient for specific clinical functions.

Even with those questions unresolved, the platform introduces a significant shift in how transient implants could be designed. Most bioresorbable electronics are engineered primarily to operate and then vanish; their failure or degradation may remain hidden until symptoms appear or the device is examined after removal. This implant is designed to make its own condition observable. Its morphology acts as a physical status display, while its triboelectric response provides a measurable electrical output. By combining molecular materials discovery, ultrasound-driven power generation and visible phosphorescence, the researchers have created an implant that can communicate its state from inside the body without a battery and without a permanent retrieval procedure. If the approach can be adapted for human use, future temporary bioelectronics may not simply disappear on schedule—they may show clinicians exactly how they are performing until the moment they do.

Subject of Research: Battery-free, bioresorbable triboelectric implants with optical monitoring of structural integrity, energy harvesting and resorption

Article Title: Bioresorbable and optically readable triboelectric implants enabled by nanoscale iridophosphors

Article References: Meng, X., Kwon, Y.H., Wang, B. et al. “Bioresorbable and optically readable triboelectric implants enabled by nanoscale iridophosphors.” Nature Nanotechnology (2026). https://doi.org/10.1038/s41565-026-02256-4

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41565-026-02256-4

Keywords: bioresorbable implants, triboelectric energy harvesting, iridophosphors, implantable bioelectronics, ultrasound power, phosphorescent imaging, transient electronics, molecular dipole moments

The advancement centers on a transformative fabrication process known as “spiral transformation,” where two-dimensional thin films harboring microfabricated devices are geometrically reconfigured into one-dimensional, cylindrical fibers. This inventive method not only allows for precise spatial control over the longitudinal, angular, and radial distribution of active components but also enables the creation of highly compact, soft fibers dense with multimodal bioelectronic functionality. Such a capability circumvents the limitations imposed by traditional microfabrication methods like photolithography, which struggle to conform devices onto thin, curved, and elongated fiber substrates.

Existing bioelectronic fibers have historically been plagued by rigidity, low density of active components, and constraints on layout precision. These factors have restricted their usability in dynamic biological environments, particularly where soft tissue interfaces demand highly flexible and biocompatible devices. The newly developed Spiral-NeuroString (S-NeuroString) fibers directly confront these challenges, offering a pliant, minimally invasive platform that supports a diverse array of sensing and stimulation capabilities within a single fiber structure.

One of the most compelling demonstrations of the S-NeuroString’s potential lies in its application to the gastrointestinal system of awake, behaving pigs. The soft fibers exhibit excellent biocompatibility and mechanical compliance with the complex and constantly moving environment of the gut. This compatibility facilitates continuous multimodal monitoring of gut motility, a critical physiological parameter, alongside the capability for targeted electrical stimulation. Such functionality signals a significant step forward for post-operative monitoring and real-time therapeutic interventions in clinical settings.

Beyond gastrointestinal applications, the researchers have showcased the power of their fibers for neuroscience research. In vivo experiments involved chronic implantation of multi-channel arrays for electrical recording within mouse brains. Remarkably, these fibers maintained stable single-unit activity recordings for up to four months, pointing to their durability and minimal tissue response over extended periods. This achievement addresses a longstanding need for long-term neural interface devices that maintain signal fidelity without causing chronic tissue damage.

The extraordinarily high channel count achievable within these fibers is another distinguishing feature. The research team fabricated an S-NeuroString fiber featuring an unprecedented 1,280 independent channels embedded within a soft fiber only 230 micrometers in diameter. This level of integration opens exciting avenues for capturing spatially rich neural and physiological data, which could revolutionize brain-machine interfaces, prosthetics, and biofeedback-controlled therapeutic devices.

Technically, the spiral transformation fabrication approach allows for unique arrangements of functional components within the fiber cross-section, including electrodes, sensors, and microstimulators. By carefully designing the initial planar devices and transforming them into spiraled fibers, the devices achieve an optimized spatial organization that enhances sensing resolution and functional multiplexing without compromising flexibility or mechanical robustness.

The use of soft materials further enhances the fibers’ suitability for chronic implantation. The fibers’ elastic properties closely match those of surrounding biological tissues, dramatically reducing mechanical mismatch, which is a primary driver of inflammation and scarring in implantable devices. This compliance, paired with precise control over device architecture, represents a new paradigm in designing bioelectronic interfaces that harmonize with the body’s natural movements.

Moreover, the multimodal sensing capabilities embedded within these fibers encompass a wide range of physiological signals, from electrophysiological recordings to chemical sensing modalities. This rich data spectrum facilitates a more comprehensive understanding of complex biological systems, enabling researchers and clinicians to develop nuanced treatment strategies tailored to individual physiological states.

From a clinical perspective, the S-NeuroString fibers hold particular promise for minimally invasive implantation procedures. Their slender profile and mechanical softness allow navigation through constrained biological pathways with minimal tissue disruption. This advantage is critical for translating advanced bioelectronic devices into practical medical tools for diagnostics, therapeutics, and long-term health monitoring.

In sum, the introduction of the Spiral-NeuroString technology marks a pivotal advancement in bioelectronic device engineering. It marries state-of-the-art microfabrication techniques with revolutionary geometric design principles to realize ultra-dense, soft, and biocompatible fibers capable of complex multimodal functionality. This breakthrough opens new frontiers for implantable electronics, offering transformative potential for neuroscience, gastroenterology, and beyond.

The multidisciplinary nature of this work, intersecting materials science, microengineering, biology, and medicine, underscores the growing importance of integrated approaches to tackling challenges in bioelectronics. As these advanced fibers progress toward clinical adoption, they are poised to enable a new generation of diagnostics and therapies that seamlessly interface with the human body in both health and disease.

Ultimately, the high-density soft bioelectronic fibers developed through spiral transformation set a compelling benchmark for future research and innovation. By overcoming key obstacles related to component density, flexibility, and spatial control, they illuminate a path toward next-generation bioelectronic systems that are minimally invasive, highly functional, and long-lasting.

Subject of Research: High-density soft bioelectronic fibers for multimodal sensing and stimulation.

Article Title: High-density soft bioelectronic fibres for multimodal sensing and stimulation.

Article References:
Khatib, M., Zhao, E.T., Wei, S. et al. High-density soft bioelectronic fibres for multimodal sensing and stimulation. Nature 645, 656–664 (2025). https://doi.org/10.1038/s41586-025-09481-2

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41586-025-09481-2