Arrhythmia Classification plays a major role while diagnosing heart diseases. Any change in the regular sequence of electric impulses is called as arrhythmia. Identifying arrhythmia as early as possible helps the patient in choosing appropriate treatment. Classification of ECG arrhythmia with high accuracy is a challenging problem. Arrhythmia classification requires preprocessing ECG Signal, extraction of features, and optimization of the features and classification of arrhythmia. This paper presents survey on issues concerned in ECG denoising, feature extraction, optimization and classification. Furthermore, methods used to analyze the performance are also discussed. Limitations and drawbacks involved in ECG denoising, Feature Extraction and Classification are discussed concluding remarks and future scope.

A Review on Arrhythmia Classification Using ECG Signals

Automatic External Defibrillator is a device that checks the heart rhythm and can send an electric shock to the heart muscle to restore the rhythm. Electric shock will be given in the case of abnormal heart which in termed as Arrhythmia. Cardiac Arrhythmias shows a condition of abnormal electrical activity in the heart which is a threat to humans. There are various algorithms proposed for the detection and classification of arrhythmia such wavelet transforms, Hilbert transform etc. Among these methods time domain analysis is simple with reduced complexity. This paper presents a feature extraction method in time domain for ECG feature extraction using adaptive threshold method for the detection of various peaks and classification of various heart conditions. This is achieved by extracting ECG features such as P, T wave, QRS complex, PR, QT, RR, ST intervals and ST segment deviations for the classification of heart beats according to different arrhythmias. Data were obtained from the records of MIT-BIH arrhythmia database (only one lead). Cardiac arrhythmias which are found are Tachycardia, Supraventricular Tachycardia and Ventricular Fibrillation.

Design and analysis of feature extraction algorithm for ECG signals using adaptive threshold method

In this work, a wearable, intelligent testing device based on nRF52832 microcontroller is designed to monitor the changes of life signs parameters of patients with depression and cardiovascular disease and determine disease seriousness and mood stability. The proposed device uses the ADS1292R chip and conductive silica gel for collection and preconditioning the electrocardiogram (ECG) and respiratory signals, which and then filtered by nRF52832. Since nRF52832 is a Bluetooth chip, it transmits the life signs signals to the mobile phone APP through Bluetooth to be displayed on a device screen. The wearable, intelligent detection devices have the advantages of small size, low power consumption, low cost, and convenience to be wearable.

Design of Wearable Device for Measurement of Multi-signature Vital Parameters

Waveform detection is the basis for automatic analysis of ECG signals, and the detection accuracy has an important effect on the assay performance In a cardiac cycle, R wave characteristics are very obvious and easy to identify, so most researchers take ECG R wave detection as the foundation of feature extraction This paper studies the principle of ECG R wave detection and carries out an in-depth discussion on some key technologies As a result, an intelligent algorithm of R wave detection is proposed The algorithm uses filter, multi-order difference and Shannon energy transformation to process ECG signals, and then it adopts an improved adaptive threshold method for detecting R waves The author used ECG data files from the MIT-BIH arrhythmia database (mitdb) to measure the performance of the algorithm Overall, the experiment obtained 9984% sensitivity and 9992% precision of R wave detection, while the error rate is only 024% Experiments show that, this algorithm could be also applied for R wave real-time detection under the requirements of high accuracy and low error rate

Zhao Wang

Papers

Research on intelligent algorithm for detecting ECG R waves