Brain‐computer interface in clinical application: How far is it from realization?

Brain-computer interface (BCI) is rapidly transitioning from concept verification to practical implementation,1 and clinical trials on BCIs are currently underway across the globe.2 The wireless minimally invasive implantable BCI, termed the Neural Electronic Opportunity (NEO), was developed by the team at Tsinghua University and successfully completed its inaugural clinical trial at Xuanwu Hospital in China in 2023. The next year, Beijing Tiantan Hospital in China reported a successful case where the NEO was used to assist a patient with high-level paraplegia in achieving mind control of cursor movement. Another successful case in China occurred in November 2024 in Huashan Hospital. The 38-year-old patient had been unable to grip and stand due to cervical spinal cord injury following a car accident. The patient recovered from the implantation procedure and was able to rise from bed and use a wheelchair on the third postoperative day. Clinical trials on BCIs are now in progress at many hospitals in China. Meanwhile, Elon Musk announced on X that Neuralink would implant its first chip into a human brain, marking the commencement of Neuralink's first human clinical trial on January 30, 2024. Neuralink received FDA approval for human trials in May 2023. As of September 2025, 12 individuals have received Neuralink's BCI device implants.3 Neuralink's first submission of research findings to the New England Journal of Medicine indicates that BCI technology is advancing to a new stage of development.4 Health Canada approved Neuralink's clinical trials to comprehensively evaluate the self-developed fully implantable wireless BCI system in 2024. The trials in China differ from the Neuralink trials of a fully implantable BCI in that the Chinese trials involve minimally invasive procedures which do not require opening the dura, resulting in a more stable electrode status, a lower risk of infection, and greater safety for patients. The advantage of BCI technology could be advantageous to patients living with disabilities due to spinal cord injury,5, 6 stroke,7-9 treatment-resistant depression,10 and other conditions,10 enhancing their ability to communicate,11-13 control and recover limb function,14 and improving quality of life. Although BCI technology has demonstrated clear clinical value for patients with disabilities, there remain bottlenecks in core algorithm breakthroughs.15-18 Additionally, the clinical application of BCIs is limited not only by technical challenges but also by medical ethics—which mandates minimizing potential harm—and by the significant cost of treatment. BCIs remain distant from widespread clinical implementation, primarily due to two prerequisites: first, demonstrable clinical efficacy must be established and substantiated to confirm sustained effectiveness; second, robust evidence-based validation is essential for its inclusion in clinical practice guidelines, which is a necessary condition for standardized adoption in routine medical care. Eventual large-scale clinical applications should use BCI medical products which are accessible, easy to use, effective and affordable. These are equally important considerations in the industrialization of BCIs. Shugeng Chen: Writing—review and editing; writing—original draft; funding acquisition; conceptualization; methodology; investigation; visualization; data curation; formal analysis. Lei Jiang: Writing—original draft; writing—review and editing; conceptualization. Jie Jia: Writing—review and editing; funding acquisition; conceptualization; methodology; investigation; supervision. This work was supported by the National Natural Science Foundation of China (82202798), the Project of Shanghai Science and Technology Commission (24YL1900202) and the Shanghai Sailing Program (22YF1404200). The authors declare no conflicts of interest. The National Natural Science Foundation of China, 82202798; Project of Shanghai Science and Technology Commission, 24YL1900202; Shanghai Sailing Program, 22YF1404200. The ethics approval is not needed in this study. Data sharing is not applicable to this article as no new data was created or analyzed in this study.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.