脑机接口分类算法近十年进展：深度学习与 EEG 研究

需积分: 50 78 浏览量更新于2024-07-16 2 收藏 1.49MB PDF 举报

"这篇文章是关于近十年来脑机接口（BCI）分类算法的综合回顾，主要关注基于EEG（脑电图）的系统。作者分析了过去十年中在BCI领域的重要进展，特别是涉及分类算法的发展，这些算法用于解析EEG信号并转化为可操作的指令。文章涵盖了各种技术，包括深度学习、遗传特征选择、图形傅里叶变换以及适用于混合BCI的工具箱等。" 在过去的十年里，脑机接口技术经历了显著的进步，尤其是基于EEG的BCI，这使得直接通过读取大脑活动来控制外部设备成为可能。这篇综述文章深入探讨了这一领域的关键分类算法，这些算法是理解并解码EEG信号的关键。首先，文章可能讨论了传统机器学习方法在BCI中的应用，如支持向量机（SVM）、人工神经网络（ANN）等，这些方法在早期的BCI研究中占据了主导地位。随着深度学习的兴起，EEG数据的处理方式也发生了变化。文章可能会介绍EEGNet，这是一个专门为BCI设计的紧凑型卷积神经网络，它能够有效地捕捉EEG信号中的时空模式。此外，深度学习的运用，如深度信念网络（DBN）和循环神经网络（RNN），也可能被提及，这些模型在处理复杂非线性问题时表现出了强大的能力。在特征选择方面，遗传算法可能被用来优化BCI系统的效率，通过选取对任务最相关的特征来减少计算负担。文章可能还提到了连接导向的图傅里叶变换在运动想象BCI解码中的应用，这是一种利用大脑网络结构信息的方法。同时，针对EEG分析的系统性回顾也涵盖了深度学习的应用，这表明研究人员正在探索更高级的模型来提升BCI的性能。此外，Gumpy，一个Python工具箱，可能被介绍为用于混合BCI开发的实用资源，它支持多种输入类型和算法集成，促进了BCI系统的灵活性和可扩展性。这篇综述为读者提供了近十年来BCI分类算法的全面视图，从传统的统计方法到最新的深度学习技术，强调了这些技术如何改变我们理解和利用大脑活动的方式，以及它们对未来BCI系统的影响。对于想要深入了解该领域的研究者和工程师来说，这是一份宝贵的参考资料。

Topical Review

performance for each class [199]. If these conditions are not

met, the Kappa metric or the confusion matrix are more infor-

mative performance measures [66]. The sensitivity-specicity

pair, or precision, can be computed from the confusion matrix.

When the classication depends on a continuous parameter

(e.g. a threshold), the receiver operating characteristic (ROC)

curve, and the area under the curve (AUC) are often used.

Classier performance is generally computed ofine on

pre-recorded data, using a hold-out strategy: some datasets

are set aside to be used for the evaluation, and are not part of

the training dataset. However, some authors also report cross-

validation measures estimated on training data, which may

over-rate the performance.

The contribution of classier performance to overall BCI

performance strongly depends on the orchestration of the BCI

subcomponents. This orchestration is highly variable given the

variety of BCI systems (co-adaptive, hybrid, passive, self- or

system- paced). The reader is referred to [212] for a compre-

hensive review of evaluation strategies in such BCI contexts.

3. Past methods and current challenges

3.1. A brief overview of methods used ten years ago

In our original review of classication algorithms for EEG-

based BCIs published ten years ago, we identied ve main

families of classiers that had been explored: linear classi-

ers, neural networks, non-linear Bayesian classiers, nearest

neighbour classiers and classier combinations [141].

Linear classiers gather discriminant classiers that use

linear decision boundaries between the feature vectors of each

class. They include linear discriminant analysis (LDA), regu-

larized LDA and support vector machines (SVMs). Both LDA

and SVM were, and still are, the most popular types of clas-

siers for EEG based-BCIs, particularly for online and real-

time BCIs. The previous review highlighted that in terms of

performances, SVM often outperformed other classiers.

Neural networks (NN) are assemblies of articial neurons,

arranged in layers, which can be used to approximate any non-

linear decision boundary. The most common type of NN used

for BCI at that time was the multi-layer perceptron (MLP),

typically employing only one or two hidden layers. Other NN

types were explored more marginally, such as the Gaussian

classier NN or learning vector quantization (LVQ) NN.

Non-linear Bayesian classiers are classiers modeling the

probability distributions of each class and use Bayes’ rule to

select the class to assign to the current feature vector. Such

classiers notably include Bayes quadratic classiers and hid-

den Markov models (HMMs).

Nearest neighbour classiers assign a class to the cur-

rent feature vector according to its nearest neighbours. Such

neighbours could be training feature vectors or class proto-

types. Such classiers include the k-nearest neighbour (kNN)

algorithm or Mahalanobis distance classiers.

Finally, classier combinations are algorithms combin-

ing multiple classiers, either by combining their outputs

and/or by training them in ways that maximize their comple-

mentarity. Classier combinations used for BCI at the time

included boosting, voting or stacking combination algorithms.

Classier combination appeared to be amongst the best per-

forming classiers for EEG based BCIs, at least in ofine

evaluations.

3.2. Challenges faced by current EEG signal classication

methods

Ten years ago, most classiers explored for BCI were rather

standard classiers used in multiple machine learning

problems. Since then, research efforts have focused on iden-

tifying and designing classication methods dedicated to the

specicities of EEG-based BCIs. In particular, the main chal-

lenges faced by classication methods for BCI are the low

signal-to-noise ratio of EEG signals [172, 228], their non-

stationarity over time, within or between users, where same-

user EEG signals varying between or even within runs [56, 80,

109, 145, 164, 202], the limited amount of training data that is

generally available to calibrate the classiers [108, 137], and

the overall low reliability and performance of current BCIs

[109, 138, 139, 229].

Therefore, most of the algorithms studied these past ten

years aimed at addressing one or more of these challenges.

More precisely, adaptive classiers whose parameters are

incrementally updated online were developed to deal with

EEG non-stationarity in order to track changes in EEG prop-

erties over time. Adaptive classiers can also be used to deal

with limited training data by learning online, thus requir-

ing fewer ofine training data. Transfer learning techniques

aim at transferring features or classiers from one domain,

e.g. BCI subjects or sessions, to another domain, e.g. other

subjects or other sessions from the same subject. As such

they also aim at addressing within or between-subjects non-

stationarity and limited training data by complementing the

few training data available with data transferred from other

domains. Finally in order to compensate for the low EEG

signal-to-noise ratio and the poor reliability of current BCIs,

new methods were explored to process and classify signals in

a single step by merging feature extraction, feature selection

and classication. This was achieved by using matrix (notably

Riemannian methods) and tensor classiers as well as deep

learning. Additional methods explored were targeted speci-

cally at learning from limited amount of data and at dealing

with multiple class problems. We describe these new families

of methods in the following.

4. New EEG classication methods since 2007

4.1. Adaptive classiers

4.1.1. Principles.

Adaptive classiers are classiers whose

parameters, e.g. the weights attributed to each feature in a lin-

ear discriminant hyperplane, are incrementally re-estimated

and updated over time as new EEG data become available

[200, 202]. This enables the classier to track possibly chang-

ing feature distribution, and thus to remain effective even with

non-stationary signals such as an EEG. Adaptive classiers

for BCI were rst proposed in the mid-2000s, e.g. in [30, 72,

J. Neural Eng. 15 (2018) 031005

剩余28页未读，继续阅读

qmzqmz

粉丝: 0
资源: 2

脑机接口分类算法近十年进展：深度学习与 EEG 研究

脑机接口综述，关键词：脑机接口、非侵入式、运动想象、解码算法、系统应用

脑机接口（Brain Computer Interaction)资料文献整理

基于AdaBoost的脑机接口分类算法研究 (2013年)

脑机接口的信号处理算法

机器学习分类算法综述

在脑机接口技术中，如何应用AdaBoosting算法提高情绪识别的准确性？请结合《基于脑机接口的脑电波情绪识别实验：准确率提升与应用潜力》一文进行分析。

脑电波义肢脑机接口技术可以用到哪些算法，要用到哪些知识

脑机接口的未来应用1000字

介绍一下脑机接口，不得低于八千字

脑机接口学习详细路线及相关书籍

最新资源