Kim, Yewon Kang, Kyumin Koo, Ja Hoon Jeong, Yoonyi Lee, Sungjun Jung, Dongjun Seong, Duhwan Kim, Hyeok Han, Hyung-Seop Suh, Minah Kim, Dae-Hyeong Son, Donghee 2025-07-18T08:30:53Z 2025-07-18T08:30:53Z 2025-07-18 2025-09 0935-9648 https://pubs.kist.re.kr/handle/201004/152811 Soft bioelectronics mechanically comparable to living tissues have driven advances in closed-loop neuroprosthetic systems for the recovery of sensory-motor functions. Despite notable progress in this field, critical challenges persist in achieving long-term stable closed-loop neuroprostheses, particularly in preventing uncontrolled drift in the electrical sensitivity and/or charge injection performance owing to material fatigue or mechanical damage. Additionally, the absence of an intelligent feedback loop has limited the ability to fully compensate for sensory-motor function loss in nervous systems. Here, a novel class of soft, closed-loop neuroprosthetic systems is presented for long-term operation, enabled by spontaneous performance recovery and machine-learning-driven correction to address the material fatigue inherent in chronic wear or implantation environments. Central to this innovation is the development of a tough, self-healing, and stretchable bilayer material with high conductivity and exceptional cyclic durability employed for robot-interface touch sensors and peripheral-nerve-adaptive electrodes. Furthermore, two central processing units, integrated in a prosthetic robot and an artificial brain, support closed-loop artificial sensory-motor operations, ensuring accurate sensing, decision-making, and feedback stimulation processes. Through these characteristics and seamless integration, our performance-recoverable closed-loop neuroprosthesis addresses challenges associated with chronic-material-fatigue-induced malfunctions, as demonstrated by successful in vivo under 4 weeks of implantation and/or mechanical damage. English WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Performance-Recoverable Closed-Loop Neuroprosthetic System Article 10.1002/adma.202503413 1 Advanced Materials, v.37, no.37 Advanced Materials 37 37 Y scie scopus 001517901600001 Chemistry, Multidisciplinary Chemistry, Physical Nanoscience & Nanotechnology Materials Science, Multidisciplinary Physics, Applied Physics, Condensed Matter Chemistry Science & Technology - Other Topics Materials Science Physics Article CONDUCTORS INTERFACE closed-loop machine learning neuroprosthetic performance-recovery self-healing sensory-motor function stretchable