A Deep Bidirectional GRU Network Model for Biometric Electrocardiogram Classification Based on Recurrent Neural Networks

Htet Myet Lynn,Sung Bum Pan,Pankoo Kim

doi:10.1109/access.2019.2939947

Abstract

In this paper, we propose a deep Recurrent Neural Networks (RNNs) based on Gated Recurrent Unit (GRU) in a bidirectional manner (BGRU) for human identification from electrocardiogram (ECG) based biometrics, a classification task which aims to identify a subject from a given time-series sequential data. Despite having a major issue in traditional RNN networks which they learn representations from previous time sequences, bidirectional is designed to learn the representations from future time steps which enables for better understanding of context, and eliminate ambiguity. Moreover, GRU cell in RNNs deploys an update gate and a reset gate in a hidden state layer which is computationally efficient than a usual LSTM network due to the reduction of gates. The experimental results suggest that our proposed BGRU model, the combination of RNN with GRU cell unit in bidirectional manner, achieved a high classification accuracy of 98.55%. Various neural network architectures with different parameters are also evaluated for different approaches, including one-dimensional Convolutional Neural Network (1D-CNN), and traditional RNNs with LSTM and GRU for non-fiducial approach. The proposed models were evaluated with two publicly available datasets: ECG-ID Database (ECGID) and MIT-BIH Arrhythmia Database (MITDB). This paper is expected to demonstrate the feasibility and effectiveness of applying various deep learning approaches to biometric identification and also evaluate the effect of network performance on classification accuracy according to the changes in percentage of training dataset.

Highlights

Due to the unique characteristics of electrocardiogram (ECG) signal, it has drawn increasing attention from biometric researchers in broad fields of information security in recent years
In order to speed up the training procedure, typically a bottleneck when running a deep network with multiple layers, our proposed network schemes are implemented using Tensorflow deep learning library written in Python, which can be executed on Graphics Processing Unit (GPU)
GPU generally brings at least 5 to 10 times speedups compared with Central Processing Unit (CPU) and can significantly accelerate the training procedure, and GeForce GTX 1080 GPU is used for our experiments

Summary

INTRODUCTION

Due to the unique characteristics of electrocardiogram (ECG) signal, it has drawn increasing attention from biometric researchers in broad fields of information security in recent years. Lynn et al.: Deep Bidirectional GRU Network Model for Biometric ECG Classification Based on RNNs sympathetic and parasympathetic nerves, it is unique and permanent as based on the size and shape of oneś heart and the orientation of valves. Several Recurrent Neural Networks (RNN) based models with different cell units, as well as, Onedimensional Convolution Neural Network (1D-CNN) model is proposed for classifying ECG signals. The main contributions of this work are: 1) We demonstrate the effectiveness of performing unidirectional and bidirectional RNNs based on both Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) cells, as well as 1D-CNN for biometrics identification.

RELATED WORKS

BACKGROUND

DATA ARGUMENTATION AND PREPROCESSING

MODELS OVERVIEW

CONCLUSION