Classification of ECG signals for detection of arrhythmia and congestive heart failure based on continuous wavelet transform and deep neural networks

Olanrewaju, Rashidah Funke; Ibrahim, Siti Noorjannah; Asnawi, Ani Liza; Altaf, Hunain

doi:10.11591/ijeecs.v22.i3.pp1520-1528

Cited by 12 publications

(10 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it is tested on known classical datasets such as CIFAR-10, like other CNN architectures. Additionally, it is utilized on Physikalisch-Technische Bundesanstalt (PTB) Diagnostic ECG Database (Özaltın and Yeniay 2021 ; Goldberger, et al 2000 ). This proposed CNN is performed for not only signals but also brain computed tomography, detailed in Ozaltin et al ( 2022 ).…”

Section: Methodsmentioning

confidence: 99%

A novel proposed CNN–SVM architecture for ECG scalograms classification

Özaltın

Yeniay

2022

Soft Comput

View full text Add to dashboard Cite

Nowadays, the number of sudden deaths due to heart disease is increasing with the coronavirus pandemic. Therefore, automatic classification of electrocardiogram (ECG) signals is crucial for diagnosis and treatment. Thanks to deep learning algorithms, classification can be performed without manual feature extraction. In this study, we propose a novel convolutional neural networks (CNN) architecture to detect ECG types. In addition, the proposed CNN can automatically extract features from images. Here, we classify a real ECG dataset using our proposed CNN which includes 34 layers. While this dataset is one-dimensional signals, these are transformed into images (scalograms) using continuous wavelet transform (CWT). In addition, the proposed CNN is compared to known architectures: AlexNet and SqueezeNet for classifying ECG images, and we find it more effective than others. This study, which not only performed CWT but also implemented short-time Fourier transform, examines the success in recognizing ECG types for the proposed CNN. Besides, different split methods: training and testing, and cross-validation are applied in this study. Eventually, CWT and cross-validation are the best pre-processing and split methods for the proposed CNN, respectively. Although the results are quite good, we benefit from support vector machines (SVM) to obtain the best algorithm and for detecting ECG types. Essentially, the main aim of the study increases classification results. In this way, the proposed CNN is utilized as deep feature extractor and combined with SVM. As a conclusion of this study, we achieve the highest accuracy of 99.21% from the proposed CNN–SVM when using CWT. Therefore, we can express that this framework can be used as an aid to clinicians for ECG-type identification.

show abstract

Section: Methodsmentioning

confidence: 99%

A novel proposed CNN–SVM architecture for ECG scalograms classification

Özaltın

Yeniay

2022

Soft Comput

View full text Add to dashboard Cite

show abstract

“…It should be noted that FPGA is very beneficial when compared to digital signal processing (DSP) processors as they are of low cost and high speeded reprogrammable devices. The real time processing of ECG signals is done using FPGA, and Hermite functions are used to process the acquired ECG signals and heart beat classification [11], [12]. A system that could be beneficial and economical for patients to measure ECG using an Arduino Nano board, which acts as a sampler and analog to digital converter (ADC) [13].…”

Section: Literature Reviewmentioning

confidence: 99%

Design and implementation of portable electrocardiogram recorder with field programmable gate arrays and IoT interface

Raj

Priya²,

Fernández³

2022

IJECE

View full text Add to dashboard Cite

The electrical activities of the heart are used to monitor cardiovascular diseases. It can be measured using electrocardiogram (ECG), a simple, painless test that can be recorded graphically. The physician, to predict the patient’s heart conditions and recommend suitable treatments, uses electrodes placed on the patient’s skin surface, to record these signals. The P, Q, R, S, T waves in the ECG signal can be used to determine the normality and abnormality of the heart's condition. The time interval differs for each cardiovascular condition of the heart. In this work, the ECG signal is acquired real-time using an intelligent sensor module, and the recorded value is processed to find the peak values. The data is sent to the web serverusing internet of things technology at a minimal time, where the physician can view it and proper decision can be taken. The real-time ECG data acquisition is also made using the field programmable gate array kit as it is a low cost, high-speed device and the output is viewed in the computer. The developed model is validated through MATLAB software and implemented for real-time applications.

show abstract

“…For the GA-BPNN approach, the authors obtained an accuracy of 97.87%. Olanrewaju et al [25] proposed a method to accurately predict heart disease using continuous wavelet transform (CWT) and deep neural deep neural networks.…”

Section: Introductionmentioning

confidence: 99%

A deep learning-based cardio-vascular disease diagnosis system

Hamdan

Aksa

Omari

et al. 2022

IJEECS

View full text Add to dashboard Cite

Recently ehealth technologies are becoming an overwhelming aspect of public health services that provides seamless access to healthcare information. Machine learning tools associated with IoT technology play an important role in developing such health technologies. This paper proposes a decision support system-based system (DSS) to make diagnosis of cardiovascular diseases. It uses deep learning approaches that classify electrocardiogram (ECG) signals. Thus, a two-stage long-short term memory (LSTM) based neural network architecture, along with an adequate preprocessing of the ECG signals is designed as a diagnosis-aided system for cardiac arrhythmia detection based on an ECG signal analysis. This deep learning based cardio-vascular disease diagnosis system (namely ‘DLCVD’) is built to meet higher performance requirements in terms of accuracy, specificity, and sensitivity. This must also be capable of an online real-time classification. Experimental results using the Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) arrhythmia database show that DLCVD led to outstanding performance

show abstract

Classification of ECG signals for detection of arrhythmia and congestive heart failure based on continuous wavelet transform and deep neural networks

Cited by 12 publications

References 24 publications

A novel proposed CNN–SVM architecture for ECG scalograms classification

A novel proposed CNN–SVM architecture for ECG scalograms classification

Design and implementation of portable electrocardiogram recorder with field programmable gate arrays and IoT interface

A deep learning-based cardio-vascular disease diagnosis system

Contact Info

Product

Resources

About