【24h】

Using Behavioral Information to Contextualize BCI Performance

机译:使用行为信息将BCI绩效背景化

获取原文

摘要

Brain-computer interface (BCI) systems often require millisecond-level timing precision in order to function reliably. However, as BCI research expands to an ever-widening array of applications, including operation in real-world environments, such timing requirements will need to be relaxed. In addition, overall BCI system design must be improved in order to better dis-ambiguate the numerous, seemingly similar, neural responses that may arise in such environments. We argue that this new area of operational BCI will require the integration of neural data with non-neural contextual variables in order to function reliably. We propose a framework in which non-neural contextual information can be used to better scope the operational BCI problem by indicating windows of time for specific analyses as well as defining probability distributions over these windows. We demonstrate the utility of our framework on a sample data set and provide discussion on many of the factors influencing performance.
机译:脑机接口(BCI)系统通常需要毫秒级的计时精度才能可靠地运行。但是,随着BCI研究扩展到越来越多的应用程序,包括在现实环境中的操作,这种时序要求将需要放宽。另外,必须改善总体BCI系统设计,以更好地消除这种环境中可能出现的众多看似相似的神经反应的歧义。我们认为,业务BCI的这一新领域将要求将神经数据与非神经环境变量集成在一起,以便可靠地运行。我们提出了一个框架,在该框架中,可以通过指示特定分析的时间窗口以及在这些窗口上定义概率分布,将非神经环境信息用于更好地确定BCI问题的范围。我们在样本数据集上演示了我们框架的实用性,并提供了许多影响性能的因素的讨论。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号