通向“人机融合”的下一步？ 超声波“读脑”技术初露头角

Core Insights - Brain-machine interface technology has made significant advancements, with companies like Neuralink demonstrating the ability to help severely disabled individuals regain communication capabilities through implanted devices [1] - Merge Labs, a newly established startup, aims to develop a non-invasive brain-machine interface using ultrasound technology, which could potentially treat mental and neurological disorders [1][2] Group 1: Company Overview - Merge Labs was founded in January this year and has secured $252 million in funding from investors, including OpenAI [1] - The company is exploring a new type of brain-machine interface technology that could not only decode information but also have therapeutic applications for conditions like depression and addiction [1] Group 2: Technology and Methodology - Traditional brain-machine interfaces rely on electrode arrays that require invasive surgeries, posing risks and uncertainties for long-term recovery [2] - Merge Labs utilizes functional ultrasound, which emits high-frequency sound waves to monitor blood flow changes in the brain, indirectly inferring neuronal activity [2] - This method is less invasive than traditional electrode implantation, as sensors can be placed beneath the skull or through a small opening, allowing for monitoring of multiple brain regions [2] Group 3: Advanced Techniques - Merge Labs is also investigating "sonogenetics," a technique that makes specific neurons more sensitive to ultrasound, enabling precise activation or inhibition of targeted neurons [3][4] - The combination of sonogenetics and ultrasound could lead to more personalized and accurate treatment options for neurological diseases, including potential applications in vision restoration and cancer treatment [4] Group 4: AI Involvement and Ethical Considerations - The significant investment in Merge Labs is closely tied to the involvement of artificial intelligence, which is expected to play a central role in analyzing complex brain signals obtained through ultrasound [5] - However, the integration of AI raises ethical and privacy concerns regarding the handling of sensitive brain data, questioning whether the technology is aimed solely at patient recovery or also at enhancing AI models [6] - There are inherent limitations to functional ultrasound, including a response time lag due to its reliance on blood flow changes, which may hinder real-time applications like speech decoding [6]