A novel approach to extract the attractor feature of RR-Lorenz plot Based on SNN Density Clustering and Its Application in ECG Analysis

DOI

10.2991/aeecs-18.2018.60How to use a DOI?

Keywords

SNN, Lorenz-RR, feature extraction, ECG analysis.

Abstract

Computer-aided diagnosis has received intensive study in recent years, especially in an aging society which unforunately has limited medical resources but surging medical demands. It's a challenging work to determine health conditions with knowledge of dynamic successive long-term ECG.RR-Lorenz plot is an essential tool for analysing ECG, it's noise-immune, and time-domain HRVs (heart rate variability) are transformed to plots, which arrhythmia could be visually identified by expericed doctors. This paper proposes a novel approach for the problem based on DBSCAN(density-based spatial clustering of applications with noise) using SNN(shared nearest neighbor). It generates attractor features of RR-Lorenz with neither prior labels nor human interventions which would be later used to measure heart conditions. We employ this approach on datasets of PhysioNet CHF database and PhysioNet normal sinus rhythm database and comes with promising results(97.35% accuracy. And also, there is significant difference on the attractor features from NSR(normal sinus rhythm) and CHF(congestive heart failure) cases, Thus we believe the proposed approach are practical and clinically useful.

Copyright

© 2018, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yanling Liu
AU  - Xin'an Wang
AU  - Ran Li
PY  - 2018/03
DA  - 2018/03
TI  - A novel approach to extract the attractor feature of RR-Lorenz plot Based on SNN Density Clustering and Its Application in ECG Analysis
BT  - Proceedings of the 2018 2nd International Conference on Advances in Energy, Environment and Chemical Science (AEECS 2018)
PB  - Atlantis Press
SP  - 358
EP  - 365
SN  - 2352-5401
UR  - https://doi.org/10.2991/aeecs-18.2018.60
DO  - 10.2991/aeecs-18.2018.60
ID  - Liu2018/03
ER  -