Uday Maji

Estimation of arrhythmia episode using variational mode decomposition technique

Detection of life threatening arrhythmia like ventricular flutter (VFL) or ventricular tachycardia (VT) and atrial flutter (AFL) at the earlier stage may save life by defibrillation therapy. Different type of mechanism has been proposed earlier in time domain or by spectral analysis of ECG signal. In this work a frequency domain approach is proposed by spectral decomposition of ECG signal. Spectral decomposition of signal is done with the help of variational mode decomposition (VMD) technique. VMD model is used to obtain the required number of spectral mode of the test signal and their central mode of oscillation. This central frequency of decomposed signal and maximum phase change within a specified window are used to characterize ventricular tachycardia and atrial flutter and compared to normal rhythms by K-near neighbour (KNN) classification method. Accuracy of 98.6% is obtained for VT classification. The proposed method eliminates the requirement of detecting fiducial points of ECG signal as necessary in conventional classification methods.

Study of atrial activities for abnormality detection by phase rectified signal averaging technique.

Abstract Non-invasive detection of Atrial Fibrillation (AF) and Atrial Flutter (AFL) from ECG at the time of their onset can prevent forthcoming dangers for patients. In most of the previous detection algorithms, one of the steps includes filtering of the signal to remove noise and artefacts present in the signal. In this paper, a method of AF and AFL detection is proposed from ECG without the conventional filtering stage. Here Phase Rectified Signal Average (PRSA) technique is used with a novel optimized windowing method to achieve an averaged signal without quasi-periodicities. Both time domain and statistical features are extracted from a novel SQ concatenated section of the signal for non-linear Support Vector Machine (SVM) based classification. The performance of the proposed algorithm is tested with the MIT-BIH Arrhythmia database and good performance parameters are obtained, as indicated in the result section.

Empirical mode decomposition vs. variational mode decomposition on ECG signal processing: A comparative study

Most of the non-stationary signals need adaptive processing technique for denoising, signal processing for feature extraction and analysis. In this regard, signal decomposition methods plays a vital role as selective reconstruction extracts the enhanced version of the signal buried in the noise. Decomposition mode based analysis also becomes popular especially in case of biosignals due to their highly non-stationary nature. Biosignals are better decomposed by a technique where basis function is derived from the signal itself. This data adaptive decomposition of biosignals into different frequency modes is very effective irrespective of multiple periodicities present in the signal or unknown sampling rate. This paper aims to study the performance of Empirical Mode Decomposition (EMD) and the Variational Mode Decomposition (VMD) technique over the popular ECG signal in terms of different periodicities during various cardiac abnormalities. The results highlight the main differences between the methods in range of signal decomposition levels as well as ability of extracting both low and high frequency from the signal.

Differentiating normal sinus rhythm and atrial fibrillation in ECG signal: A phase rectified signal averaging based approach

Abnormal electrical activities in the heart cause various types of arrhythmia or cardiac dysrhythmia. Atrial fibrillation and atrial flutter are most important among them. Automatic detection of different cardiac abnormalities is an emerging field of study in assistive diagnosis technology for cardiac diseases. Atrial fibrillation (AF) is a kind of arrhythmia which increases risk of heart attack especially to the older people. Detection of AF at the early stage may cause prevention of serious stoke. In this paper a new technique based on phase rectified signal averaging (PRSA) is proposed to characterize and classify the AF rhythm from normal. The performance of this method is tested with the MIT-BIH arrhythmia data base and sensitivity and specificity of more than 96% is obtained.

Detection of Atrial Flutter using PRSA

Automatic detection of different cardiac abnormalities is an emerging field of study in assistive diagnosis technology for cardiac diseases. A study on the feasibility of automatic detection of Atrial Flutter (AFL) based on time and frequency domain features has been presented in this paper to prevent the serious heart failure by detecting it at early stage. The proposed algorithm is developed based on feature subsets of a set of statistical time-frequency-domain parameters by using phase rectified signal average (PRSA) method. Classification of the abnormality using the derived features has been performed with the help of two class clustering method by Support Vector Machine (SVM). This classifier is tested on 382 and 587 numbers of AFL and normal cardiac cycles respectively taken from MIT-BIH Arrhythmia database. Satisfactory result is obtained as the 96% sensitivity and 98% specificity is observed.

Imposed target based modification of Taguchi method for feature optimisation with application in arrhythmia beat detection

This paper shows a system for feature optimization using modified Taguchi method.This method can reduce the number of features, and classification hazards.This study enhances the rules of Taguchi method for the system has no exact output.This study classifies multiple clusters with less number of parameters.This method possesses minimal pre-processing with anchor point feature selection. Development of an expert system for clinical application includes automation in diagnosis of abnormality and patient monitoring based on features derived from continuous data set. This paper presents a novel method for feature optimization and classification of electrocardiogram (ECG) for arrhythmia analysis. A feature set optimization technique can reduce the classification hazard by selecting few comprehensive features to cater all kind of abnormalities under consideration. Proposed work deals with ranking and selection of an optimized pair of features using Taguchi method from eleven possible features normally used for characterizing arrhythmic beats like left bundle branch (LBBB), right bundle branch (RBBB) and premature ventricular contraction (PVC) are compared to normal beats. An imposed target based modification of Taguchi method is also suggested for the systems where the output is not pre-defined as in the case of biomedical applications. The proposed method is advantageous for the expert systems in which individual identity of the features are to be stored while reducing the dimensionality of the feature set. Multiclass Navis Bayes classifier is used to classify the beats in a single run and good performance parameters are obtained as reported in the result section.

Characterizing cardiac arrhythmia by optimized window length based PRSA technique

Phase Rectified Signal Averaging (PRSA) technique is a promising method for analysis of quasi-periodic signals which helps to identify characteristic frequencies contained in the data by disregarding the artifacts and noises. But the performance of the PRSA technique largely depends on the choice of window length (WL). It is required to optimize WL for better detection of precise but important periods present in signal. In this paper a method to optimize the window length is proposed based on the spectral analysis of original and PRSA signal. The proposed method is applied on ECG signal to characterize and classify the atrial fibrillation (AF), atrial flatter (AFL) and ventricular flutter (VFL) rhythms with statistical features. Classification of the fibrillatory episode is done by K-nearest-neighbor (KNN) and support vector machine (SVM) clustering method with derived features. Extracted features are clustered with a new approach of Root Mean Square (RMS) Technique. This algorithm is applied on to the MIT-BIH arrhythmia database and checks the performance. Both quantitative and qualitative analysis is made and sensitivity and specificity 98.24% and 96.08% respectively is achieved.