Modeling electroencephalography waveforms with semi-supervised deep belief nets: fast classification and anomaly measurement.

Modeling electroencephalography waveforms with semi-supervised deep belief nets: fast classification and anomaly measurement.
复制标题

DOI:
10.1088/1741-2560/8/3/036015
复制
发表时间:
2011-06
影响因子:
4
通讯作者:
Litt B
Litt B
中科院分区:
工程技术2区
文献类型:
--
作者:
Wulsin DF;Gupta JR;Mani R;Blanco JA;Litt B

文献摘要

参考文献

被引文献

相似文献

临床脑电图(EEG)记录了大量人类复杂数据,但主要仍由人工判读者进行审阅。深度置信网络(DBNs)是一种相对新型的多层神经网络,通常在二维图像数据上进行测试,但很少应用于像脑电图这样的时间序列数据。我们在半监督模式下应用深度置信网络对脑电图波形进行建模,以用于分类和异常检测。在我们的脑电图数据集上,深度置信网络的性能与标准分类器相当,并且发现其分类时间比其他高性能分类器快1.7到103.7倍。我们展示了深度置信网络学习的无监督步骤如何产生一种自动编码器,它可以自然地用于异常测量。我们比较了使用原始的、未处理的数据(在自动化生理波形分析中很少见)和人工选择的特征,发现原始数据产生了相当的分类效果以及更好的异常测量性能。这些结果表明,深度置信网络和原始数据输入对于在线自动化脑电图波形识别可能比其他常用技术更有效。
Clinical electroencephalography (EEG) records vast amounts of human complex data yet is still reviewed primarily by human readers. Deep Belief Nets (DBNs) are a relatively new type of multi-layer neural network commonly tested on two-dimensional image data, but are rarely applied to times-series data such as EEG. We apply DBNs in a semi-supervised paradigm to model EEG waveforms for classification and anomaly detection. DBN performance was comparable to standard classifiers on our EEG dataset, and classification time was found to be 1.7 to 103.7 times faster than the other high-performing classifiers. We demonstrate how the unsupervised step of DBN learning produces an autoencoder that can naturally be used in anomaly measurement. We compare the use of raw, unprocessed data—a rarity in automated physiological waveform analysis—to hand-chosen features and find that raw data produces comparable classification and better anomaly measurement performance. These results indicate that DBNs and raw data inputs may be more effective for online automated EEG waveform recognition than other common techniques.
DOI: 10.1023/b:aire.0000045502.10941.a9
发表时间: 2004-10-01
影响因子: 12
作者:
Hodge, VJ;Austin, J
通讯作者: Austin, J
DOI: 10.1016/j.clinph.2006.12.019
发表时间: 2007-05-01
影响因子: 4.7
作者:
Gardner, Andrew B.;Worrell, Greg A.;Litt, Brian
通讯作者: Litt, Brian
DOI: 10.1016/0013-4694(94)00286-t
发表时间: 1995-05-01
期刊: ELECTROENCEPHALOGRAPHY AND CLINICAL NEUROPHYSIOLOGY
影响因子: --
作者:
ALARCON, G;BINNIE, CD;POLKEY, CE
通讯作者: POLKEY, CE
DOI: 10.1111/j.1528-1157.1998.tb01430.x
发表时间: 1998-06-01
期刊: EPILEPSIA
影响因子: 5.6
作者:
Osorio, I;Frei, MG;Wilkinson, SB
通讯作者: Wilkinson, SB
DOI: 10.1016/s1388-2457(02)00296-1
发表时间: 2003-01-01
影响因子: 4.7
作者:
Flanagan, D;Agarwal, R;Gotman, J
通讯作者: Gotman, J