瑞士联邦理工学院研究团队开发了一种名为LogSync的可植入神经芯片，能够实时追踪脑区之间的通信，并识别认知和情绪状态1。该系统通过机器学习算法分析网络级生物标志物，在功耗低于200微瓦的约束下实现跨区域神经信号处理1，这一功耗水平远低于现有植入式神经接口系统的典型功耗1。
该研究由EPFL集成神经技术实验室主任Mahsa Shoaran领导1。系统采用对数计算创新设计，简化了脑网络分析所需的复杂运算1。研究人员在人类脑记录和大鼠实验中验证了该芯片区分认知和情绪状态的能力1。这项技术有望用于精神疾病的自适应深部脑刺激治疗1。
研究成果获得了美国国家心理健康研究所（NIMH）660万瑞士法郎的资助1，并已发表于《IEEE固态电路杂志》2026年9月刊1。
Researchers at Switzerland's Federal Institute of Technology (EPFL) have developed an implantable neural chip called LogSync that can track communication between brain regions in real time and identify cognitive and emotional states 1. The system processes cross-regional neural signals through machine learning algorithms while operating at extremely low power consumption of less than 200 microwatts, significantly below the typical power requirements of existing implanted neural interface systems 1. This advance has potential applications in adaptive deep brain stimulation therapy for psychiatric disorders 1.
The LogSync chip employs an innovative logarithmic computation design that simplifies the complex mathematical operations necessary for analyzing brain networks 1. The research has been validated in both human brain recordings and rat experiments, demonstrating the system's capability to distinguish between different cognitive and emotional states 1. The work, led by Mahsa Shoaran, director of EPFL's Integrated Neuromorphic Technology Laboratory, received funding of 6.6 million Swiss francs from the U.S. National Institute of Mental Health (NIMH) 1. The findings were published in the IEEE Journal of Solid-State Circuits in September 2026 1.
评论
还没有评论，欢迎留下第一条。