Algorithm
Allam Jaya prakash,Samit Ari
Department of Electronics and Communication Engineering, National Institute of Technology, Rourkela,India
E-mail: allamjayaprakash@gmail.com
ECG plays a most important role in finding cardiac disorders of the heart. Cardiac arrhythmias occur in a short duration of time which can’t
be distinguishable by a human eye. The finding of arrhythmias is a tedious task since slight changes in electro cardiogram signal (ECG)
may lead to life-threatening. Diagnosis and medication at an early stage of cardiac arrhythmia may facilitate to reduce the mortality rate of
the heart patients. This paper presents an accurate system for the classification of 5 types of ECG arrhythmias namely Paced (P), Premature
ventricular contraction (V), Normal (N), Right bundle branch block (R) and Left bundle branch block (L). In the proposed technique, random
forest classifier (RF) is used as classifier to classify ECG arrhythmias. The projected model of cardiac arrhythmia recognition system covers
3 stages they’re pre-processing, feature extraction and classification. In the first stage, filtered the ECG signal raw data and finding the R-peak
locations of the ECG signal. Dual tree complex wavelet transform (DTCWT) is employed to extract the feature vector from the ECG signal
within the second stage. The final feature vector consists of an extracted feature set using DTCWT and four other temporal features of the
ECG signal. The final feature set is applied as an input to the random forest (RF) classifier. Classifier performance is evaluated using MIT-BIH
physionet database. Finally RF classifier achieved overall accuracy of 98.78% on an individual basis once tested over five kinds of physionet
MIT-BIH arrhythmia database. The proposed arrangement of framework effectively classified five kinds of ECG arrhythmias. The proposed
RF classifier performs better than other reported techniques for the ECG arrhythmia classification.
1. Introduction: Electrocardiogram (ECG) is the recording quantization (LVQ) and self-organizing maps (SOM) are two
of the electrical phenomenon at the heart that represents the classification models based on clustering technique, LVQ is a
variability of heart rate. Any deviation from the normal heart rate, supervised learning technique which classifies the feature vector
disturbance in rhythm, site of origin or conduction of cardiac corresponding to the label of the cluster pattern (code word) into
electric impulse is considered as an arrhythmia. Most of the which is clustered. In SOM, each centring of the cluster (prototype
arrhythmias are seldom occurring and can’t be distinguished by a or code word) is represented by some weights of a neuron which is
person’s eye. Long duration of ECG record called as ambulatory assigned to correlate in the feature map. The classifier is modelled
electrocardiogram are required to identify the abnormalities present using mixture of SOM and LVQ approaches. The network is
in a patient. Moreover, huge variations in temporal and some other designed in such a way that LVQ gives superior classification
morphological characteristics from one patient to the other patient performance for classes 1 and 3 however for class 2 and 4 SOM
make detection of abnormalities is a challenging task. Hence it is gives superior performance. Overall classification accuracy of 94%
very difficult to analyse and recognize these arrhythmias manually is reported using this mixture of expert’s approach.
by an expert cardiologist also. Hence we require an automatic A unique technique is proposed in [7] for a patient-
computer-aided diagnostic (CAD) system that can quickly detect adapted ECG heartbeat classification that consists of four stages
abnormalities within the patient. namely, pre-processing, feature extraction, feature selection and
In literature, there are many algorithms proposed on automatic classification. Features are extracted in temporal and frequency
classification of ECG arrhythmias. Most of them have done ECG domain. Frequency domain features include coefficients obtained
classification in the subsequent stages i) pre-processing ii) feature by applying stock well transform. 184 samples around each R
extraction iii) classification. In this feature extraction techniques peak are taken by combining 4 temporal features (Pre R-R, Post
include both time and frequency domain features. Some of the R-R, average R-R, local R-R) and 180 samples of frequency
temporal features include R-R intervals, Q-R intervals, QRS domain features and obtained better classification results but poor
complex duration, R-S intervals, ST time segments. These temporal sensitivity as reported in [7].
features are not enough for classification since there is a huge To address the above limitations, a novel approach of random
variation in the other morphological patterns within the same forest (RF) classifier is proposed for the classification of ECG
patient. Hence some of the researchers have reported the mixture arrhythmias. Random forest is a highly powerful machine learning
of both frequency and time features as reported in [1]. Many algorithm which is based on supervised learning.[8] Random forest
transformation techniques like S transform, Fourier transform (FT) algorithm can use individually for both classification and the
and discrete wavelet transform (DWT) [2] are used to extract the regression kind of difficulties. As name suggests that this algorithm
features from the pre-processed data. Extracted features are given creates the forest with many number of random trees .
to the input of a classifier for classification into respective classes. The remaining paper is organized as follows: Section 2 presents
Some of the classifiers are artificial neural networks (ANN) [3], MIT-BIH database that is used to evaluate the performance of
support vector machine (SVM) [4] etc. the proposed classification algorithm. Section 3 contains proposed
Detection of cardiac arrhythmias by applying Hidden Markov framework. Section 4 shows the results and discussion of the
models is reported in [5].This algorithm has demonstrated very proposed algorithm. The conclusion of the work is explained in
good promising results in classifying ventricular arrhythmias and section 5.
detection of low amplitude P wave detection. In [6], the author
presented a customized electrocardiogram (ECG) beat classifier 2. ECG data Processing: MIT-BIH ECG Arrhythmia database
using a mixture of experts (MOE) methodology. Linear vector is utilised for assessing the performance of the proposed classifier
(i) Pre-processing
Normalization
from 24 hours recordings of 48 specific patients. The first 23
recordings correspond to the routine clinical recordings while the
remaining recordings contain the complex ventricular, junctional, Filtering
and supraventricular arrhythmias [9]. These ECG recordings are
sampled at 360 Hz and band-pass filtered at 0.1-100 Hz. Fig.1 R-peak detection
represents the plot of record 100 from MIT-BIH ECG signal
database. Annotation file contains labels for each rhythm that are
detected by using a simple slope sensitive detector. These labels are
used in the training and testing phase of the implemented algorithm.
Morphological features
Temporal features
AC power
In this work, five different types of cardiac arrhythmias namely
Timing
DTCWT features
information
P V N RBBB LBBB