A temporally biosynchronized and physically transient peripheral nerve interface

Peripheral nerve injury often leads to severe and persistent motor and sensory deficits, thereby markedly reducing quality of life. Peripheral nerve interfaces offer opportunities to facilitate functional restoration by interrogating neural activity and delivering electrical therapy. However, existing devices are typically not designed for long-term monitoring of injured nerves and are fabricated from non-degradable materials that require surgical removal. Here we present a temporally biosynchronized, physically transient and flexible peripheral nerve interface that enables stable monitoring of neural activity throughout the post-injury recovery process. Integrated with machine-learning-based decoding, the interface achieves excellent motor intention recognition using peripheral nerve signals in a rodent sciatic nerve injury model, outperforming signals recorded from the primary motor cortex. Moreover, the device enables continuous tracking of nerve regeneration and supports bidirectional signal transmission, providing a foundation for closed-loop rehabilitation. This work establishes a transient and biomimetic peripheral nerve interface with the potential to advance neuroregeneration monitoring and adaptive neurorehabilitation strategies.