Key result
PCA and FA improve automated sleep stage detection accuracy across most machine learning models.
Why the study?
Methods for automatic sleep stage detection rely on large polysomnographic data, and the impact of dimensionality reduction techniques on classifier performance is unclear.
Dimensionality reduction techniques like PCA and FA can improve the accuracy and reduce the computational load of automated sleep stage detection models.
Dimensionality reduction may streamline sleep stage classifiers; leaves open optimal technique for clinical PSG workflows.
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.
No takes yet. Share an insight, caveat, or question.
Tăuţan et al. (2020) studied Sleep stage detection. Dimensionality reduction techniques (PCA, FA, and autoencoders) vs. Standard classifiers without dimensionality reduction was evaluated on Accuracy of automated sleep stage detection. Dimensionality reduction techniques such as PCA and FA improved the accuracy of automated sleep stage detection models in most cases, while autoencoders maintained performance and reduced computational load.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: