Fan Zhao

Continuous monitoring of intra-abdominal pressure is important for detecting clinically silent pressure elevation and preventing life-threatening complications. However, existing sensing platforms rarely combine self-powered operation, high sensitivity, and tissue compatibility. Here, we establish a multiscale charge-dynamics engineering strategy for protein-based triboelectric materials using silk fibroin as a model system. At the molecular scale, native β-sheet nanofibrils are preserved during regeneration to promote crystallization and enhance charge retention. At the mesoscale, template-assisted electrospinning creates a checkerboard-patterned ridge-peak topography that increases interfacial contact and promotes charge generation. The resulting charge-dynamics-engineered silk fibroin (CDE-SF) sensor delivers an open-circuit voltage of 128.4 V, a short-circuit current of 4.7 μA and a sensitivity of 19.7 mV Pa-1, which, to our knowledge, is the highest reported among protein-based TENGs. In vivo rat tests showed respiration-associated abdominal signals in both wearable and implantable modes. In human-subject tests, the wearable sensor resolves 12 abdominal activities with an overall classification accuracy of 92% and assigns them to three predefined risk categories with class-wise accuracies of 85-92%. These findings establish charge-dynamics engineering as a generalizable materials-design framework for protein-based self-powered sensors and support its further development for wearable and implantable monitoring of abdominal pressure-related dynamics.

Triboelectric nanogenerators (TENGs) based on poly(vinylidene fluoride) (PVDF) have attracted considerable attention for wearable and self-powered biomedical electronics due to their flexibility, biocompatibility, and favorable triboelectric properties. However, their relatively limited electrical output still hinders practical applications. Herein, a synergistic strategy integrating hierarchically porous structure engineering and barium titanate (BaTiO3, BTO) nanoparticle (NP) incorporation was developed to enhance the triboelectric performance of PVDF nanofiber membranes. In particular, the hierarchical porous structure increases surface roughness and effective contact area, thereby promoting interfacial charge generation. BTO NPs compensate for the reduction in effective dielectric permittivity associated with the porous structure. The synergistic effect of these two factors achieved a short-circuit current of 0.863 ± 0.027 μA, a transferred charge density of 25.99 ± 0.62 μC m−2, and a power density of 650 mW m−2. The output voltage increased to 64.44 ± 0.83 V, corresponding to a 204.8% enhancement compared with the pristine device. Finally, the device was attached to the elbow joint, the contact area of the arm, and the sole of a shoe to verify its ability to monitor human motion signals and power electronic devices. The results demonstrated that the sensors can effectively distinguish between different motion states and can power small electronic devices for harvesting mechanical energy. Overall, this study developed a simple yet highly efficient strategy for enhancing the electrical output capacity of PVDF-TENG and holds promising potential for practical applications in human physiological signal monitoring and self-powering wearable devices.

Wireless intelligent tactile perception systems that can replicate human touch are urgently needed for sensory restoration in the dexterous neuroprosthetics and humanoid robots. However, a key challenge, is achieving simultaneous microstructural sensing and large‐area pattern recognition, primarily due to the difficulty in detecting both high‐frequency vibrations and static pressure. Herein, we report a bionic, textile‐based sliding tactile sensor (BTSTS) featuring a Janus 3D honeycomb‐structured sensing electrode paired with an interdigitated electrode. The BTSTS exhibits excellent hydrophobicity, abrasion resistance (withstanding record‐high abrasion 2000 cycles), and exceptional capabilities in perceiving both static pressure and vibrations (5–600 Hz). It represents the first piezoresistive tactile sensor with an ultra‐broad frequency‐detection range surpassing the human vibrotactile limit (98.8%) in identifying diverse objects, bridging material innovation with edge‐computing to advance haptic intelligence for human‐robot interaction and sensory augmentation.

Esophageal stent insertion is a key palliative therapy for malignant esophageal strictures, but the postoperative migration rate remains as high as 40%, significantly compromising clinical outcomes. Stent migration behavior is closely related to its structure and mechanical properties; however, the underlying mechanisms remain unclear, and there is a lack of effective in vitro evaluation methods to predict migration risk. Herein, we first developed a novel biomimetic swallowing peristalsis simulation device that highly replicates human physiological environments and swallowing waveforms—addressing the limitations of existing in vitro testing methods. Using this device, we demonstrated for the first time that stent migration is co-regulated by radial force and axial bending stiffness: higher radial force enhances anchoring via increased friction, while lower bending stiffness (superior flexibility) reduces migration risk by maintaining a larger stent–esophagus contact area and improving energy dissipation during swallowing. These conclusions are supported by our theoretical derivations and test results of stents with different densities. In addition, it was found that food viscosity and tumor block also influence stent migration risk. This study elucidates the synergistic mechanism of esophageal stent migration and provides a theoretical foundation and an in vitro validation platform for the design of a new generation of anti-migration esophageal stents.

Conventional ureteral stents often face a critical trade-off between radial support and lumen patency, compromising drainage performance under compression. To overcome this, we designed a novel interlocked bi-layer nitinol (NiTi) braided stent. A thread-inspired interlocking strategy between the inner and outer layers was achieved by introducing a thick NiTi wire into the inner layer, significantly enhancing interlayer bonding (64.10 ± 5.62 cN). The optimized stent featured an ultra-thin wall structure (0.31 ± 0.03 mm), which enabled a 29.03% larger lumen diameter while providing a 174.38% higher radial strength (2215.44 ± 42.87 cN) compared to commercial polyurethane (PU) ureteral stents. Moreover, it demonstrated superior flexibility, with a bending stiffness of 0.87 ± 0.07 N·mm2. Under simulated ureteral obstruction model (1.0 mm constant compression) with low renal pressure, the stent preserved 33.56% of its initial drainage rate, while the PU stent lost virtually all flow. Additionally, the retention force (24.08 ± 0.98 cN), optimized via J-end geometry, was equivalent to clinical standards. This study introduces a structurally reinforced, drainage-efficient ureteral stent that synergistically improves lumen patency, mechanical robustness, and anchoring performance. The proposed interlocked bi-layer braiding concept offers a scalable design strategy for next-generation stents and other minimally invasive tubular implants.

Implantable sensors are crucial for real‐time monitoring of internal soft tissue biomechanics, yet existing systems rarely combine high sensitivity, flexibility, and biodegradability. Here, we present an implantable sensor by engineering a flexo‐ferroelectric nanocomposite, where ferroelectric β‐glycine nanocrystals are in situ‐grown and stabilized along poly( L ‐lactic acid) (PLLA) nanofibers via an electric field‐assisted nanoconfinement strategy. This approach first uncovers flexoelectricity in PLLA nanofibers and template‐freely stabilizes pure β‐glycine nanocrystals without brittle polymer matrices. These components form a structurally coupled, self‐assembled interface, creating a co‐active charge generation network that integrates flexoelectric and ferroelectric effects new mechanism not previously demonstrated. Our sensor achieves unprecedented electromechanical performance, including piezoelectric coefficient of 246 pC/N, sensitivity of 205 mV/kPa, detection range of 3.5‐110 kPa, and tissue‐matched compliance. The integrated performance surpasses all reported bioresorbable sensors and commercial polyvinylidene fluoride‐based devices. Functionally, it demonstrates layer‐specific compatibility across soft tissues, capturing strain from surface to deep regions with minimal signal disruption. In vivo validation in rodent and rabbit models enables continuous cardiovascular monitoring and, notably, discriminates between metabolic and mechanical modes of acute heart failure. This work establishes a universal sensing platform for high‐fidelity soft tissue strain mapping, with broad implications for personalized mechano‐diagnostics across multiscale physiological systems.