--- title: "China's first fully-implantable wireless brain chip just met a paralyzed patient" date: 2026-09-30 category: Brain–Computer Interface site: NeuroAI canonical: https://neuroai.site/a/na-harbin-wireless-bci-implant language: en --- # China's first fully-implantable wireless brain chip just met a paralyzed patient > A 56-year-old man with cervical spinal-cord injury in Harbin received a neuron-level, high-throughput implantable wireless BCI built by CAS's Aerospace Information Research Institute — no external wires, no battery-replacement surgery. A 56-year-old man who lost the use of his limbs in a fall can now, in early training, imagine a grip and watch a robotic hand close. The chip inside his skull has no wires leaving the body, and no battery a surgeon must later dig out. This is not another demo clip. It is the first fully-implantable, wireless brain-computer interface (脑机接口) implant reported in Heilongjiang, performed at Harbin Medical University First Affiliated Hospital (哈尔滨医科大学附属第一医院) on a patient with cervical spinal-cord injury and high paraplegia (高位截瘫). ## The implant that lives entirely under the skin The device is described as a neuron-level, high-throughput implantable wireless BCI (植入式无线脑机接口). Its defining traits: - A **micrometer-thin flexible electrode** with good biocompatibility and long-term stability — thin enough that implantation trauma is minimal. - The ability to capture **single-neuron electrical signals**, which lifts decoding precision and the freedom of control well above surface-level recording. - A **fully buried wireless collector with a built-in rechargeable battery** — no external leads, and no second surgery just to swap a power cell. That last point is the quiet headline. Most invasive BCIs either leave a connector through the skin (an infection path) or need periodic battery surgery. A device meant for years of daily rehab has to remove both. ## Who built it, and why it is different from the others The surgery team was the hospital's neurosurgery department, reported by Xinhua's Heilongjiang channel as led by professor Shi Huaizhang (史怀璋). The system was developed by the **Chinese Academy of Sciences Aerospace Information Innovation Research Institute (中国科学院空天信息创新研究院)**. This matters for how China's BCI field is splitting into routes: - **NEO** (Neuracle / 博睿康) and **BeiNao-1 (北脑一号)** sit **extradurally (硬膜外)** — on top of the dura, not inside the tissue. Safer, but they trade some signal sharpness. - The Harbin system is **fully invasive (侵入式)** — it records single neurons, so it can, in principle, resolve intent far more finely. The cost is a higher surgical and long-term risk. The same system has a track record before this implant: in August 2025, Harbin and the CAS institute completed what was described as the world's first BCI-assisted brain-tumor boundary localization clinical trial. In May 2026, Beijing Tiantan Hospital (首都医科大学附属北京天坛医院) led the launch of a multicenter trial of this system for spinal-cord-injury patients, meant to assess safety and efficacy in those left quadriplegic by cord damage. ## What the patient is supposed to get back The patient, injured in a 2024 fall, could not stand, turn over, or perform fine hand movements. Under the trial protocol: - Training begins about **five weeks after surgery**, starting with motor imagery (运动想象), model calibration, and cursor control. - The core endpoint is the **six-month brain-controlled cursor success rate**, alongside device safety, transmission efficiency, and daily-living ability. - Later, a **smart rehab glove (智能康复手套)** is used so the patient can practice finger grasp and flexion — the step from "move a cursor" to "pick up a cup." The team's stated near-term goals are hand grasp and even dialing a phone — modest, concrete tasks that mean independence. ## Why "wireless + rechargeable" beats a flashy demo The field is full of impressive single-session videos. What actually blocks invasive BCI from leaving the ward is the boring stuff: infection risk from percutaneous wires, and the impossibility of asking a paralyzed patient to undergo repeated battery surgeries. A wireless, rechargeable, fully buried implant is the engineering choice that makes long-horizon rehab plausible. ## Honest limitations - This is a **single case at an early stage**. No peer-reviewed efficacy paper is available, and no functional restoration result has been published yet. - "Fully invasive" means **higher surgical risk** than the extradural devices now in larger trials; long-term biocompatibility in humans is unproven. - The developer is named as the CAS Aerospace Information Research Institute; a separate commercial entity was not clearly identified in the reports we found. - Battery life and recharge cycle were **not disclosed**. - Comparisons to other Chinese BCIs (NEO, BeiNao-1) are directional, not head-to-head, and the Harbin hospital was still recruiting 2–3 more cervical-spinal patients at the time of reporting. ## What readers can do now - **If you follow clinical BCI**, watch the Beijing Tiantan-led multicenter spinal-injury trial registrations — that is the dataset that will eventually show whether this system is safe and useful, not a single implant story. - **If you build implants**, treat "wireless + rechargeable + fully buried" as the differentiator worth engineering for; the wire and the battery are the two things that have historically kept invasive BCI out of daily rehab. - **If you report on the field**, keep "implanted and stable" separate from "restored function" — the first is what this case shows, the second is still a protocol away. --- Published by NeuroAI (https://neuroai.site/) — https://neuroai.site/a/na-harbin-wireless-bci-implant Free to quote with attribution and a link to the original.