[论文解读] Evidence-Based Filters for Signal Detection: Application to Evoked Brain Responses

Malik Ahmed Mubeen, Kevin H. Knuth|arXiv (Cornell University)|Jul 6, 2011

EEG and Brain-Computer Interfaces参考文献 20被引用 3

一句话总结

本文提出了一种基于证据的过滤方法，利用贝叶斯模型证据检测诱发脑响应，特别是脑电信号（EEG）中的P300事件相关电位。通过计算信号模型的边缘似然而非依赖相关性，该方法在单次试验检测的准确性和速度上显著优于传统的基于相关性的方法，如BMI数据的受试者工作特征曲线下面积（ROC）分析所示。

ABSTRACT

Template-based signal detection most often relies on computing a correlation, or a dot product, between an incoming data stream and a signal template. Such a correlation results in an ongoing estimate of the magnitude of the signal in the data stream. However, it does not directly indicate the presence or absence of the signal. The problem is really one of model-testing, and the relevant quantity is the Bayesian evidence (marginal likelihood) of the signal model. Given a signal template and an ongoing data stream, we have developed an evidence-based filter that computes the Bayesian evidence that a signal is present in the data. We demonstrate this algorithm by applying it to brain-machine interface (BMI) data obtained by recording human brain electrical activity, or electroencephalography (EEG). A very popular and effective paradigm in EEG-based BMI is based on the detection of the P300 evoked brain response which is generated in response to particular sensory stimuli. The goal is to detect the presence of a P300 signal in ongoing EEG activity as accurately and as fast as possible. Our algorithm uses a subject-specific P300 template to compute the Bayesian evidence that a applying window of EEG data contains the signal. The efficacy of this algorithm is demonstrated by comparing receiver operating characteristic (ROC) curves of the evidence-based filter to the usual correlation method. Our results show a significant improvement in single-trial P300 detection. The evidence-based filter promises to improve the accuracy and speed of the detection of evoked brain responses in BMI applications as well the detection of template signals in more general signal processing applications

研究动机与目标

• 为解决基于相关性的方法在检测诱发脑响应时无法以统计置信度量化信号存在的局限性。
• 开发一种基于贝叶斯模型比较的信号检测框架，特别利用边缘似然（证据）来评估信号存在性。
• 提升基于脑电信号的脑机接口（BMIs）中单次试验P300检测的准确性和速度。
• 提供一种通用的信号检测方法，可应用于神经科学以外的领域，尤其适用于基于模板的信号识别。

提出的方法

• 该方法计算给定数据窗口和受试者特异性P300模板下，信号模型的贝叶斯证据（边缘似然）。
• 将数据建模为噪声与信号分量的混合，其中信号由已知模板和未知幅度定义。
• 使用共轭先验进行解析计算，实现时间窗上的高效递归计算。
• 算法以滑动窗口方式处理输入的脑电信号，在每一步更新证据估计值。
• 通过直接评估模型的合理性，避免了阈值设定或幅度估计的需要。
• 通过受试者工作特征（ROC）曲线对比证据驱动滤波与标准相关性检测方法，评估性能。

实验结果

研究问题

• RQ1 贝叶斯模型证据是否能为诱发脑响应提供比相关性更可靠的信号检测度量？
• RQ2 在单次试验P300检测中，基于证据的滤波与相关性方法在检测准确性和速度方面有何比较？
• RQ3 所提出的方法能否在脑机接口应用中实现高效的实时脑电信号处理？
• RQ4 在使用贝叶斯证据时，使用受试者特异性模板是否能提升检测性能？

主要发现

• 基于证据的滤波器在检测单次试验P300响应方面显著优于相关性方法，表现为曲线下面积（AUC）更优。
• 该方法通过直接量化信号存在的概率，而非估计其幅度，实现了更快、更准确的检测。
• 贝叶斯方法在脑电信号中常见的低信噪比条件下，降低了假阳性率并提高了敏感性。
• 该算法计算效率高，适用于脑机接口系统中的实时实现。