Machine Learning Approach to Predict Ventricular Fibrillation Based on QRS Complex Shape

Taye, Getu Tadele; Shim, Eun Bo; Hwang, Han-Jeong; Lim, Ki Moo

doi:10.3389/fphys.2019.01193

Cited by 37 publications

(25 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…They extracted the features using the aforementioned three traditional analysis methods. In our previous study, we investigated the feasibility of QRS complex shape features on VF onset prediction, where we used ANN and demonstrated the superiority of QRS features as compared to traditional HRV features in terms of VF prediction performance 9 .…”

mentioning

confidence: 99%

Application of a convolutional neural network for predicting the occurrence of ventricular tachyarrhythmia using heart rate variability features

Taye

Hwang

Lim

2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

predicting the occurrence of ventricular tachyarrhythmia (VtA) in advance is a matter of utmost importance for saving the lives of cardiac arrhythmia patients. Machine learning algorithms have been used to predict the occurrence of imminent VtA. in this study, we used a one-dimensional convolutional neural network (1-D CNN) to extract features from heart rate variability (HRV), thereby to predict the onset of VtA. We also compared the prediction performance of our cnn with other machine leaning (ML) algorithms such as an artificial neural network (ANN), a support vector machine (SVM), and a k-nearest neighbor (KNN), which used 11 HRV features extracted using traditional methods. The proposed CNN achieved relatively higher prediction accuracy of 84.6%, while the ANN, SVM, and KNN algorithms obtained prediction accuracies of 73.5%, 67.9%, and 65.9% using 11 HRV features, respectively. Our result showed that the proposed 1-D CNN could improve VTA prediction accuracy by integrating the data cleaning, preprocessing, feature extraction, and prediction. Heartbeat is regulated by electrical signals conducted across the four chambers of the heart: two atria and two ventricles. When electrical activity is normal, the heart beats approximately 60 to 100 times per minute. However, abnormal electrical signals in the heart lead to disorganized electrical activities such as ventricular tachyarrhythmia (VTA), which causes fast heart rate 1. Thus, early VTA prediction helps physicians to take immediate medical procedure to reduce the risk. Ventricular tachycardia (VT) and ventricular fibrillation (VF) are the most common VTAs. VT arises from improper electrical activity in the ventricles, and can cause sudden cardiac arrest. VF is caused by chaotic electrical activity in the ventricles, which is similar to VT, but is a fatal condition that requires immediate medical attention. In VF, the heart shivers instead of pumping blood. Developing earlier preventive interventions would reduce the risk of experiencing an imminent VT and VF events. Researchers used noninvasive tests by measuring and analyzing electrocardiograms (ECGs), where heart rate variability (HRV) is extracted to train machine learning (ML) algorithms for predicting VT or VF in advance 2. HRV is the most commonly employed biomarker for isolating VT or VF subject from the normal subject 3. It is a time variation of heartbeats among two successive QRS complexes (Q, R, and S waves in ECG). In recent years, HRV indices have been used as a noninvasive biomarkers to forecast life-threatening arrhythmias 4. Previous studies mainly used the three traditional analysis methods: time domain, frequency domain, and Poincare nonlinear analyses, to extract features from HRV. Furthermore, they used these features as input to machine learning algorithms to predict the occurrence of VT, VF, or both. The machine learning techniques are used to classify the complex feature patterns and enable early prediction of VT or VF events with high accuracy. Acharya et al. used features extracted fr...

show abstract

mentioning

confidence: 99%

Application of a convolutional neural network for predicting the occurrence of ventricular tachyarrhythmia using heart rate variability features

Taye

Hwang

Lim

2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…Tong et al [22] and Joo et al [24] did not show the precision values of their results. More recently, Joo et al [24], Lee et al [25] and Taye et al [26] used ANN models with ECG signal to construct VT/VF predicting machines. Specially, Getu et al constructed an ANN model with morphological features of ECG signal to predict VT and claimed the accuracy of their approach reached 98.6% [26], however this study only detected VT.…”

Section: Discussionmentioning

confidence: 99%

“…More recently, Joo et al [24], Lee et al [25] and Taye et al [26] used ANN models with ECG signal to construct VT/VF predicting machines. Specially, Getu et al constructed an ANN model with morphological features of ECG signal to predict VT and claimed the accuracy of their approach reached 98.6% [26], however this study only detected VT. Compared with the studies above, our approach predicts two types of MAs (VT/VF) and achieves the highest overall accuracy, sensitivity, and specicity.…”

Section: Discussionmentioning

confidence: 99%

“…constructed an ANN binary model using 14 parameters from HRV and respiratory rate variability to predict VT occurrence and claimed that their model can predict VT 30 seconds prior to the event with a sensitivity of 0.88, a specificity of 0.82, and an area under the receiver operating characteristic curve (AUC ROC) of 0.93 [25]. Taye et al constructed an ANN model with morphological features of ECG signal to predict VT 30 seconds prior to its onset and claimed that the accuracy of their approach reached 98.6% [26].…”

mentioning

confidence: 99%

See 1 more Smart Citation

The Feasibility of Predicting Impending Malignant Ventricular Arrhythmias on the basis of Signal Complexity of Heartbeat Intervals

Chen¹,

Kido²,

Ono³

et al. 2019

Preprint

View full text Add to dashboard Cite

Malignant ventricular arrhythmias (MAs), such as ventricular tachycardia (VT) that presages cardiac arrest, present the highest hurdle for the healthcare community to overcome. Given that MAs occur unpredictably and lead to emergencies, convenient tracking devices, e.g. photoplethysmogram (PPG), that could predict MAs would be irreplaceably valuable. Since the use of heartbeat intervals (HbI) to predict the occurrence of arrhythmias is becoming more feasible, a further attempt to establish a new convenient approach for predicting impending MAs with HbI is worth trying. Assuming that intrinsic characteristics of MAs (VT and ventricular fibrillation: VF) can be revealed by a suitable approach on the basis of signal complexity, we propose an approach that first expresses the physiological status of the heart by HbI; then delineates the patterns of HbI by a new complexity metric (refined composite multi-scale entropy: RCMsEn); and finally trains a nonlinear machine learning model (random forest: RF) to learn the specific patterns of MAs so as to differentiate them from the normal sinus heart rhythm (N) and other prevalent arrhythmias (atrial fibrillation: AF, and premature ventricular contraction: V).For calculating entropy values and predicting MAs as early as possible (which is the aim of this study), two specifications are of interest: the minimal length of HbI needed to delineate the MAs patterns sufficiently (lenmin), and the maximum time length at which our model can predict impending MAs (timemax). We compared the RF model with support vector machine (SVM) models based on linear and Gaussian kernels. Results show that the RF model performs the best, reaching a 99.24% recall and a 99.87% precision for a HbI of 500 heartbeats (the lenmin) 374 seconds (the timemax) preceding the occurrence of MAs. The HbI samples in this study were extracted from an electrocardiograph (ECG). However, given the subtle difference (0.1 ms typically) between the R-R interval of ECG and the P-P interval of PPG, this approach could be extended to HbI acquired by the PPG sensor and thus should be of substantial theoretical and practical significance in cardiac arrest prevention.

show abstract

“…The method proposed in this study involves data preprocessing, feature extraction, and predictions using one dimensional deep convolution neural networks. Even though other algorithms exist that can be used in performing feature extraction and classification of sequence or time-series data, convolution neural networks have been found to be superior [13][14][15][16] in extracting unique features and patterns in sequences [17][18][19] , images and video data. In fact, deep convolution neural networks algorithms are behind some of the state-of-the-art detection and tracking algorithms found in everyday images and videos.…”

Section: Introductionmentioning

confidence: 99%

Tracking and Classifying Global COVID-19 Cases by using 1D Deep Convolution Neural Networks

Amo-Boateng

2020

Preprint

View full text Add to dashboard Cite

The novel coronavirus disease (COVID-19) and pandemic has taken the world by surprise and simultaneously challenged the health infrastructure of every country. Governments have resorted to draconian measures to contain the spread of the disease despite its devastating effect on their economies and education. Tracking the novel coronavirus 2019 disease remains vital as it influences the executive decisions needed to tighten or ease restrictions meant to curb the pandemic. One-Dimensional(1D) Convolution Neural Networks (CNN) have been used classify and predict several time-series and sequence data. Here 1D-CNN is applied to the time-series data of confirmed COVID-19 cases for all reporting countries and territories. The model performance was 90.5% accurate. The model was used to develop an automated AI tracker web app (AI Country Monitor) and is hosted on https://aicountrymonitor.org. This article also presents a novel concept of pandemic response curves based on cumulative confirmed cases that can be use to classify the stage of a country or reporting territory. It is our firm believe that this Artificial Intelligence COVID-19 tracker can be extended to other domains such as the monitoring/tracking of Sustainable Development Goals (SDGs) in addition to monitoring and tracking pandemics.

show abstract

Machine Learning Approach to Predict Ventricular Fibrillation Based on QRS Complex Shape

Cited by 37 publications

References 22 publications

Application of a convolutional neural network for predicting the occurrence of ventricular tachyarrhythmia using heart rate variability features

Application of a convolutional neural network for predicting the occurrence of ventricular tachyarrhythmia using heart rate variability features

The Feasibility of Predicting Impending Malignant Ventricular Arrhythmias on the basis of Signal Complexity of Heartbeat Intervals

Tracking and Classifying Global COVID-19 Cases by using 1D Deep Convolution Neural Networks

Contact Info

Product

Resources

About