Classification algorithms of error-related potentials in brain-computer interface

被引：4

作者：

Sun J. ^{[1
]}

Jung T.-P. ^{[1
,2
,3
]}

Xiao X. ^{[1
]}

Meng J. ^{[1
]}

Xu M. ^{[1
,2
]}

Ming D. ^{[1
,2
]}

机构：

[1] Academy of Medical Engineering and Translational Medicine. Tianjin University, 300072, Tianjin

[2] School of Precision Instrument and Opto-electronics Engineering, Tianjin University, 300072, Tianjin

[3] University of California and Swartz Center for Computational Neuroscience, 92093, CA

来源：

Shengwu Yixue Gongchengxue Zazhi/Journal of Biomedical Engineering | 2021年 / 38卷 / 03期

基金：

中国国家自然科学基金;

关键词：

brain-computer interface; error-related potentials; pattern recognition; single trial recognition;

D O I：

10.7507/1001-5515.202012013

中图分类号：

学科分类号：

摘要：

基于错误相关电位（ErrP）的错误自检测有助于改善脑-机接口系统的实用性。但单试次 ErrP 信号的准确识别仍是阻碍这一技术发展的难题。为了衡量不同算法解码 ErrP 信号的能力，本文使用两个不同的公开数据集，对研究与相关应用中常见的 4 种线性判别分析算法、2种支持向量机、逻辑回归以及判别典型模式匹配（DCPM）共 8 个算法展开对比。文中主要分析了算法的分类正确率和算法性能随训练集样本数量的变化情况。实验结果表明 DCPM 具有最佳的综合性能。本研究揭示了各算法性能与训练样本数目和 ErrP 试验范式间的相互影响，为研究与实际应用中 ErrP 解码算法的选择提供参考。.; Error self-detection based on error-related potentials (ErrP) is promising to improve the practicability of brain-computer interface systems. But the single trial recognition of ErrP is still a challenge that hinters the development of this technology. To assess the performance of different algorithms on decoding ErrP, this paper test four kinds of linear discriminant analysis algorithms, two kinds of support vector machines, logistic regression, and discriminative canonical pattern matching (DCPM) on two open accessed datasets. All algorithms were evaluated by their classification accuracies and their generalization ability on different sizes of training sets. The study results show that DCPM has the best performance. This study shows a comprehensive comparison of different algorithms on ErrP classification, which could give guidance for the selection of ErrP algorithm.

引用

页码：463 / 472

页数：9