当前位置: X-MOL 学术Biomed. Eng. Biomed. Tech. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Dimensionality reduction for EEG-based sleep stage detection: comparison of autoencoders, principal component analysis and factor analysis
Biomedical Engineering / Biomedizinische Technik ( IF 1.3 ) Pub Date : 2021-04-01 , DOI: 10.1515/bmt-2020-0139
Alexandra-Maria Tăuţan 1 , Alessandro C Rossi 2 , Ruben de Francisco 2 , Bogdan Ionescu 1
Affiliation  

Methods developed for automatic sleep stage detection make use of large amounts of data in the form of polysomnographic (PSG) recordings to build predictive models. In this study, we investigate the effect of several dimensionality reduction techniques, i.e., principal component analysis (PCA), factor analysis (FA), and autoencoders (AE) on common classifiers, e.g., random forests (RF), multilayer perceptron (MLP), long-short term memory (LSTM) networks, for automated sleep stage detection. Experimental testing is carried out on the MGH Dataset provided in the “ You Snooze, You Win: The PhysioNet/Computing in Cardiology Challenge 2018 ”. The signals used as input are the six available (EEG) electoencephalographic channels and combinations with the other PSG signals provided: ECG – electrocardiogram, EMG – electromyogram, respiration based signals – respiratory efforts and airflow. We observe that a similar or improved accuracy is obtained in most cases when using all dimensionality reduction techniques, which is a promising result as it allows to reduce the computational load while maintaining performance and in some cases also improves the accuracy of automated sleep stage detection. In our study, using autoencoders for dimensionality reduction maintains the performance of the model, while using PCA and FA the accuracy of the models is in most cases improved.

中文翻译:

基于 EEG 的睡眠阶段检测的降维:自编码器的比较、主成分分析和因子分析

为自动睡眠阶段检测开发的方法利用多导睡眠图 (PSG) 记录形式的大量数据来构建预测模型。在这项研究中,我们研究了几种降维技术,即主成分分析 (PCA)、因子分析 (FA) 和自动编码器 (AE) 对常见分类器的影响,例如随机森林 (RF)、多层感知器 (MLP) ),长短期记忆 (LSTM) 网络,用于自动睡眠阶段检测。实验测试在“You Snooze, You Win: The PhysioNet/Computing in Cardiology Challenge 2018”中提供的 MGH 数据集上进行。用作输入的信号是六个可用的 (EEG) 脑电图通道以及与提供的其他 PSG 信号的组合:ECG – 心电图,EMG – 肌电图,基于呼吸的信号——呼吸努力和气流。我们观察到,在使用所有降维技术时,在大多数情况下获得了相似或改进的准确度,这是一个很有希望的结果,因为它允许在保持性能的同时减少计算负载,并且在某些情况下还提高了自动睡眠阶段检测的准确度。在我们的研究中,使用自动编码器进行降维可以保持模型的性能,而使用 PCA 和 FA 在大多数情况下提高了模型的准确性。这是一个很有希望的结果,因为它允许在保持性能的同时减少计算负载,并且在某些情况下还提高了自动睡眠阶段检测的准确性。在我们的研究中,使用自动编码器进行降维可以保持模型的性能,而使用 PCA 和 FA 在大多数情况下提高了模型的准确性。这是一个很有希望的结果,因为它允许在保持性能的同时减少计算负载,并且在某些情况下还提高了自动睡眠阶段检测的准确性。在我们的研究中,使用自动编码器进行降维可以保持模型的性能,而使用 PCA 和 FA 在大多数情况下提高了模型的准确性。
更新日期:2021-03-26
down
wechat
bug