A bio-inspired robotic hand developed by Hainan University（Illustration by Chen Haibing）

Hainan University's Wireless BCI Platform for Macaques（Photo: Hainan University）

At the opening ceremony of the 2014 FIFA World Cup in Brazil, a paraplegic youth named Juliano Pinto, wearing a “robotic exoskeleton”, successfully kicked off the first ball using “mind control” in front of hundreds of millions of viewers worldwide. This scene—reminiscent of a science fiction movie—was made possible by a brain-computer interface (BCI) system developed by a team at Duke University in the United States.

A brain-computer interface is a technology that enables information exchange between the brain and a machine by collecting and decoding neural signals.

Today, Chinese research teams have successfully made a significant leap in this cutting-edge field, advancing from trailing behind to keeping pace with global leaders. In April this year, Hainan University officially released its independently developed core technologies for invasive BCIs, along with a series of related products, which were debuted at the China International Consumer Products Expo 2025. Among them, three core chips have reached international world-class levels in key metrics such as precision of neural signal acquisition, freedom in parameter control and wireless transmission efficiency, breaking the long-standing reliance on imported chips in this field.

Notably, these three chips form a closed-loop system for signal collection, regulation, and transmission, covering the full technological chain required by BCIs. This marks China’s achievement of independent control over the BCI technologies, with Hainan continuously injecting “Chinese chip” power into brain science research and medical applications through scientific and technological innovation.

A

From Science Fiction to Reality: Hainan’s Exploration of BCI Technology

In a laboratory at Hainan University, a macaque is playing computer games using “mind control”. Fixing its gaze on the ball on the screen, it moves the ball using thought alone—without any physical movement. With the help of BCI technology, the whole process—from “signal acquisition” to “decoding” to “execution”—takes just few dozen of milliseconds. This is not a science fiction but a real world breakthrough enabled by BCI technology.

“The human brain is an extraordinarily complex structure, comprising 86 billion neurons and over 100 trillion connections, making it the most powerful information processor known,” said Yin Ming, head of BCI integrated Circuit & Neural Engineering Team at Hainan University. “BCI chips are the ‘core of the core’, serving as the bridge between the brain and the outside world. They can capture tiny electrical signals between neurons in real-time and convert them into commands to control external devices through AI algorithms.”

For many years, China relied on imported BCI chips.

In 2020, Hainan University established its BCI integrated circuit & neural engineering team, focusing on the independent development of high-density neural signal acquisition chips.

Over the past five years, the team has achieved a series of breakthroughs, successively developing three core chips: the SX-R128S4 high-throughput neural signal acquisition and stimulation chip, the SX-S32 high-degree-of-freedom neural modulation chip, and the SX-WD60 low-power wireless transmission chip. All have reached world-class performance standards. Notably, the 128-channel acquisition chip has twice the channel count of current leading international competitors’ products. While improving performance, it has reduced power consumption by more than 80% and size by 50%. Since its founding, the team has applied for nearly 20 patents, and several of its research outcomes have been published in leading international journals.

B

Five Years to Forge a Blade: The University’s “Groundbreaking Journey”

Looking back, this team has walked a path of innovation filled with challenges.

With strong backing from Luo Qingming, President of Hainan University and Member of CAS , the team was founded in 2020. Despite generous support from the university in many aspects, the cost of tackling critical “chokepoint” technologies remained prohibitively high.

To overcome funding hurdles, team members immersed themselves in intensive literature reviews and rigorous experimental design to boost their success rates, ultimately securing the approval for the China Brain Project.

In the early days, when BCI research was still an emerging field in China, the team faced formidable technical hurdles. BCI chips need to satisfy four nearly contradictory demands: ultra-high channel count and precision for real-time neural decoding, miniaturization for safe implantation, low noise to detect faint neural signals, and ultra-low power consumption to avoid heat buildup and tissue damage.

After repeated experiments and ongoing optimization, the team has adopted an innovative ultra-low-power design, ensuring post-implantation temperature rise stayed within a safe margin of 2°C without compromising performance.

The team has also made significant breakthroughs in size and energy efficiency. Over the past four years, the first 128-channel chip has undergone multiple iterations, reducing in size to 6.5mm x 5mm and in power consumption to just 15 milliwatts. In 2023, the team successfully developed a modular high-throughput acquisition chip with over 2560 channels and a compact of only 2.5cm x 2.6cm, achieving new technical milestones surpassing those of leading global competitors.

C

Breaking the Industrial Bottleneck: From Laboratory to Market

With key technological breakthroughs, the industrialization of Hainan University’s BCI chip technology has been accelerating. In January 2024, the team launched SensingX (Hainan) Co., Ltd., which has already released its commercial BCI chip products

“Our 32-channel stimulation chip and 128-channel acquisition chips have entered mass production,” said Guo Zheshan, a member of this team. “These chips can rival imported products in performance and offer significant cost advantages. Even more exciting, the next-generation chips with higher channel counts and enhanced functions are progressing well and will soon be released.”

Yang Quanwei, General Manager of Kedou (Suzhou) Brain-Computer Technology—an industry unicorn—said that his company has purchased a batch of 32-channel stimulation chips developed by the Hainan University team and completed performance testing of the 128-channel acquisition chip, which is now planned for procurement.

“Hainan University’s self-developed BCI chips outperform international competitors on multiple key metrics and are the first to achieve mass production in China, filling the gap in domestic high-performance BCI chips market,” Yang Quanwei said. He believes that as production scales up, prices will continue to fall, thereby lowering the cost of end products and broaden application coverage.

D

Medical New Era: Hyperlink to Mind-Controlled Reality

Overseas, the invasive BCI implant procedure by Elon Musk’s Neuralink (a BCI company) enabled a quadriplegic man to manipulate on-screen cursors through “thoughts” after surgery, allowing him to play online games.

Domestically, a 19-year-old epilepsy patient, following BCI surgery, not only mastered basic games such as “Pac-Man” through brain control, but also achieved precise operation of complex games like “Honor of Kings” and “Black Myth: Wukong”.

Aphasia patients speaking again, the visually impaired regaining sight, and people with paralysis controlling machines with their thoughts—once seemingly wild ideas, now appear within reach thanks to brain-computer interface (BCI) technology.

“Transforming self-developed BCI chips into clinical products requires going through processes like quality testing, ethical approvals, and clinical trials, which typically takes around 3-5 years,” stated Liang Fengyan, a member of Hainan University’s BCI chip team. He noted that while their team’s BCI chips haven’t yet entered clinical stages, they’ve established collaborations with Beijing Tiantan Hospital, Haikou People’s Hospital and several medical companies. “Many medical institutions are optimistic about BCI, recognizing its broad clinical adoption prospects,” he said.

Notably, the team’s “Brain-Spine-Machine” interface closed-loop system can directly collect brain motor signals, bypassing damaged spinal cords to activate muscles, rebuilding digital pathway between the brain and the spinal cord for paralysis patients, enabling true mind-controlled lower limb movement. Experiments conducted in May 2025 demonstrated human movement intentions successfully drove continuous left-right lower limb motions in anesthetized macaques, offering new hope for complete paralysis patients.

In the field of neurological disorder, the Deep Brain Stimulation (DBS) system shows tremendous potential. By monitoring abnormal brain activity in real time and automatically triggering precise regulation, it can effectively suppress tremors in Parkinson’s patients and block epileptic seizures. The neuromodulation chip developed by the team serves as the core component of DBS system. “By collaborating with medical enterprises, we can accelerate the clinical adoption of our self-developed chips, which significantly reduces costs for companies using imported chips,” said Liang Fengyan. He noted that their high-end BCI acquisition and neuromodulation chips will soon enter mass production, which is expected to lower the cost of current neuromodulation device and benefit patients with spinal cord injuries, Parkinson’s disease, epilepsy and other neurological disorders.

E

New Developmental Stage: “Future” Has Arrived, Yet Challenges Persist

BCI technology’s application and promotion still face multiple challenges. According to Yin Ming, from the technical perspective, human brain contains about 86 billion neurons, yet current devices can capture signals from only a few hundred to a few thousand. Large-scale recording and accurate decoding and encoding remain significant challenges. Another challenge is the long-term stability of implanted electrodes. “Typically, the signals often deteriorate significantly after 6-12 months,” he said.

In terms of clinical translation, Sui Yanfang, Deputy Director of the Rehabilitation Department of Haikou People’s Hospital, pointed out that invasive BCI devices face complex approval procedures, highlighting the urgent need for specialized review standards and regulatory frameworks. She also pointed out that the biocompatibility of current implant materials still requires improvement. In addition, high medical costs remain a major obstacle to widespread clinical adoption. Issues such as data security management, patient privacy protection, and social equity in technology application demand joint efforts from both academia and industry to develop effective solutions.

Encouragingly, the National Healthcare Security Administration recently issued the Guidelines for the Establishment of Pricing Items for Neurological Medical Services(Trial), paving the way for the clinical adoption of BCI technology. For the first time, this guideline established independent pricing items specifically for BCI technology, including fees for invasive BCI implantation and removal, as well as non-invasive BCI adaptation. This marks a significant step forward—once the technology matures, a clear pricing pathway for clinical adoption will already be in place. As local authorities align with and implement the guideline, BCI-related medical services will have a standardized basis for charging. As local authorities align with and implement the guideline, BCI-related medical services will establish standardized pricing frameworks.

Dr. Zhang Shaomin, a researcher at Zhejiang University’s Qiushi Institute for Advanced Studies and the Ministry of Education’s Frontier Science Center for Brain and Brain-Machine Integration, stated that as BCI technology continues to advance, it is expected to bring disruptive innovations to the daily lives of healthy individuals.

For example, in the entertainment sector, BCI technology could be deeply integrated with virtual reality (VR), enabling users to control game characters or adjust virtual environments through neural signals, significantly enhancing the sense of immersion. In the arts, it can capture real-time fluctuations in audience emotions and attention, dynamically adjusting digital artworks, and even allowing audiences to directly participate in creation through brain waves, ushering in a new paradigm of human-machine co-creation. In industrial settings, this technology enables workers to remotely operate heavy machinery through mental commands, reducing operational risks while improving precision.

Moreover, BCI technology holds great potential in fields such as education, social interaction and sports—for instance, enabling real-time assessment of learning focus, facilitating “thought-based” social communication, or enhancing athletes’ neural response speed.

Breakthroughs in full-chain R&D of BCI chips, electrodes by domestic research institutions such as Hainan University are opening up new frontiers for the development of BCI technology and unlocking doors of hope for millions of patients. In the near future, “Chinese chips” may lead the global BCI field into a new stage of development.

Translated by Wu Tong, Zhang Xiaohan

Proofread by Zhang Ying, Yang Jie