Assessment of Outliers and Detection of Artifactual Network Segments Using Univariate and Multivariate Dispersion Entropy on Physiological Signals.

Evangelos Kafantaris,Tsz-Yan Milly Lo,Javier Escudero,Ian Piper

doi:10.3390/e23020244

Abstract

Network physiology has emerged as a promising paradigm for the extraction of clinically relevant information from physiological signals by moving from univariate to multivariate analysis, allowing for the inspection of interdependencies between organ systems. However, for its successful implementation, the disruptive effects of artifactual outliers, which are a common occurrence in physiological recordings, have to be studied, quantified, and addressed. Within the scope of this study, we utilize Dispersion Entropy (DisEn) to initially quantify the capacity of outlier samples to disrupt the values of univariate and multivariate features extracted with DisEn from physiological network segments consisting of synchronised, electroencephalogram, nasal respiratory, blood pressure, and electrocardiogram signals. The DisEn algorithm is selected due to its efficient computation and good performance in the detection of changes in signals for both univariate and multivariate time-series. The extracted features are then utilised for the training and testing of a logistic regression classifier in univariate and multivariate configurations in an effort to partially automate the detection of artifactual network segments. Our results indicate that outlier samples cause significant disruption in the values of extracted features with multivariate features displaying a certain level of robustness based on the number of signals formulating the network segments from which they are extracted. Furthermore, the deployed classifiers achieve noteworthy performance, where the percentage of correct network segment classification surpasses 95% in a number of experimental setups, with the effectiveness of each configuration being affected by the signal in which outliers are located. Finally, due to the increase in the number of features extracted within the framework of network physiology and the observed impact of artifactual samples in the accuracy of their values, the implementation of algorithmic steps capable of effective feature selection is highlighted as an important area for future research.

Highlights

With 16 experimental setups and 15 features extracted from the network segments of each setup, a total of 240 feature distributions are produced corresponding to artifactual network segments, alongside 15 feature distributions corresponding to normal network segments
Since in each experimental setup, only one of the physiological signals contains outliers, out of the 15 total features extracted per network segment, only eight of these features were extracted from signal combinations that include the “artifactual” signal
Each network segment consists of four synchronised physiological signal segments: EEG, RESP, blood pressure (BP), and ECG, resulting in a total of 16 experimental setups, with each setup being defined by the signal containing the artifactual outliers and the percentage of samples set as outliers as specified by the corresponding P factor, with possible values being: 0.1%, 0.5%, 1%, and 5%

Summary

Introduction

As a paradigm, aims to describe the interaction across diverse organ systems in the form of physiological networks. Within this framework, each system is represented as a node of the network, and interactions across systems are projected as edges between the nodes [1,2]. Each system is represented as a node of the network, and interactions across systems are projected as edges between the nodes [1,2] This approach allows the monitoring of complex physiological interactions in the body, through the detection of topological transitions which occur within the networks when their physiological state changes.

Objectives

Methods

Results

Discussion

Conclusion