Chinese Scientists Conduct First Invasive Brain-Computer Interface Clinical Trial

China has entered the clinical trial phase of invasive brain-computer interface (BCI) technology, marking a significant advancement. The Chinese Academy of Sciences' Institute of Brain Science and Smart Technology recently completed the country's first prospective clinical trial, establishing China as the second nation globally to reach this milestone after the United States. The research team from the Brain Science and Smart Technology Excellence Innovation Center developed and manufactured an ultra-flexible neural electrode. This electrode is currently the smallest and most flexible in the world, with a cross-sectional area only one-fifth to one-seventh that of similar foreign products and a flexibility surpassing them by over a hundred times. This innovation significantly reduces damage to brain tissue, addressing a critical bottleneck in implantable BCIs related to poor biocompatibility and narrow channel bandwidth. The high-density, large-scale, high-throughput, and long-term stable in-vivo neural signal acquisition capabilities of the ultra-flexible electrode have been validated in rodents, non-human primates, and human brains. These features make it particularly effective in capturing single-neuron spike signals, a key requirement for advanced BCI applications. The team's invasive BCI system is the first in China to receive a registration inspection report and is capable of consistently collecting such signals over extended periods. In terms of surgical friendliness, the implant is just 26 mm in diameter and less than 6 mm thick, making it the smallest in the world—about the size of a coin. Unlike foreign counterparts, this system can achieve comparable control levels with fewer implanted electrodes, thereby enhancing the patient's benefit-to-risk ratio. Real-time online decoding is a crucial aspect of BCI technology. The system must extract features, interpret motor intentions, and generate control commands within tens of milliseconds. The research team developed an innovative online learning framework that dynamically optimizes the neural decoder. This framework uses a parameter self-adjustment mechanism to align the decoder's optimization with neural plasticity, overcoming the limitations of traditional static models that struggle to adapt to the time-varying nature of neural signals. Combined with the stability of the flexible electrode and advanced high-precision neural activity estimation techniques, the system achieves low-latency, high-robustness, and day-to-day stable real-time motion decoding. In earlier animal experiments, the BCI system was implanted into the hand and arm function areas of the motor cortex in macaque monkeys. The system operated stably without any infections or electrode failures. After training, the monkeys were able to control a computer cursor using only their neural activity and could even type on a brain-controlled keyboard. Moreover, the feasibility of surgical upgrades was successfully demonstrated. The initial implant was safely removed, and a new one was inserted into the same cranial opening, with the system continuing to operate without issues. The monkeys quickly adapted to the new system and resumed smooth cursor control. The human participant in the trial is a male who lost all four limbs due to a high-voltage electrical accident. Before the surgery, the research team used functional magnetic resonance imaging (fMRI) and CT scans to reconstruct a detailed three-dimensional model of his brain, ensuring precise placement of the electrodes. The operation was performed with millimeter-level precision to maximize safety and effectiveness. Since the implantation of the BCI device in March 2025, the system has been stable, with no reports of infection or electrode failure over two months. Within just 2-3 weeks of training, the participant achieved skills such as playing chess and racing games on a computer, performing at levels comparable to those of ordinary users of computer touchpads. The next steps for the research team involve enabling the participant to use a robotic arm to perform physical tasks like grasping objects and picking up cups. They also plan to explore the control of more complex devices, including robotic dogs and embodied AI robots, further enhancing the participant's ability to interact with the physical world. Looking forward, this invasive BCI system holds the potential to significantly improve the quality of life for millions of patients suffering from complete spinal cord injuries, double upper-limb amputations, and amyotrophic lateral sclerosis (ALS). The ultra-flexible electrode, with its extremely small size—about one-hundredth the width of a human hair—ensures minimal invasiveness and enhances long-term stability. This breakthrough not only advances the field of neuroscience but also opens new possibilities for individuals with severe motor disabilities.