Detection of cardiac arrhythmia in electrocardiograms using adaptive feature extraction and modified support vector machines

Shen, Chia-Ping; Kao, Wen-Chung; Yang, Yueh-Yiing; Hsu, Ming-Chai; Wu, Ya; Lai, Feipei

doi:10.1016/j.eswa.2012.01.093

Cited by 53 publications

(36 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SVM is widely used for ECG classification due to its simplicity, robustness and efficiency [2,3], which was confirmed in our previous study, too [4,5].…”

Section: Introductionsupporting

confidence: 49%

SVM Based ECG Classification Using Rhythm and Morphology Features, Cluster Analysis and Multilevel Noise Estimation

Smíšek

Hejc

Ronzhina

et al. 2017

Computing in Cardiology Conference (CinC)

View full text Add to dashboard Cite

show abstract

“…SVM is widely used for ECG classification due to its simplicity, robustness and efficiency [2,3], which was confirmed in our previous study, too [4,5].…”

Section: Introductionsupporting

confidence: 49%

SVM Based ECG Classification Using Rhythm and Morphology Features, Cluster Analysis and Multilevel Noise Estimation

Smíšek

Hejc

Ronzhina

et al. 2017

Computing in Cardiology Conference (CinC)

View full text Add to dashboard Cite

show abstract

“…The selected algorithms are fast and the results easy to interpret, as shown in both [36,40]. Once the P and T waves and the complex QRS are detected, an automatic diagnosis can be performed by using databases with different pathologies, the most common being MIT-BIH Arrhythmia, on which classifiers as SVM [41], KNN [42], decision rules [43] are applied. Given that any classifier can be applied to the database to generate a predictive model, this field was not considered very important.…”

Section: State Of the Artmentioning

confidence: 99%

“…In any case, a basic functionality for detecting problems is included, although it is not necessarily a diagnostic system because the hardware is not considered safe for application in serious diseases. Previous studies [41] generated classifiers, such as SVM, from a trained database [46], and were able to reach an accuracy rate of near to 90%.…”

Section: Proposed Systemmentioning

confidence: 99%

Monitoring and Detection Platform to Prevent Anomalous Situations in Home Care

Villarrubia

Bajo

Paz

et al. 2014

Sensors

View full text Add to dashboard Cite

Monitoring and tracking people at home usually requires high cost hardware installations, which implies they are not affordable in many situations. This study/paper proposes a monitoring and tracking system for people with medical problems. A virtual organization of agents based on the PANGEA platform, which allows the easy integration of different devices, was created for this study. In this case, a virtual organization was implemented to track and monitor patients carrying a Holter monitor. The system includes the hardware and software required to perform: ECG measurements, monitoring through accelerometers and WiFi networks. Furthermore, the use of interactive television can moderate interactivity with the user. The system makes it possible to merge the information and facilitates patient tracking efficiently with low cost.

show abstract

“…The approach we proposed attains an F 1 -Score of 0.992 that is the weighted average score of 10-fold cross-validation with 10% testing data and 90% training data. Compared with available studies [8][9][10][11][12][13][14][15][16][17] , the approach we proposed achieved the highest accuracy score by using all the data files in MIT-BIH database.…”

mentioning

confidence: 96%

Optimal Multi-Stage Arrhythmia Classification Approach

Zheng

Chu

Struppa

et al. 2020

Sci Rep

View full text Add to dashboard Cite

Arrhythmia constitutes a problem with the rate or rhythm of the heartbeat, and an early diagnosis is essential for the timely inception of successful treatment. We have jointly optimized the entire multi-stage arrhythmia classification scheme based on 12-lead surface ECGs that attains the accuracy performance level of professional cardiologists. The new approach is comprised of a three-step noise reduction stage, a novel feature extraction method and an optimal classification model with finely tuned hyperparameters. We carried out an exhaustive study comparing thousands of competing classification algorithms that were trained on our proprietary, large and expertly labeled dataset consisting of 12-lead ECGs from 40,258 patients with four arrhythmia classes: atrial fibrillation, general supraventricular tachycardia, sinus bradycardia and sinus rhythm including sinus irregularity rhythm. Our results show that the optimal approach consisted of Low Band Pass filter, Robust LOESS, Non Local Means smoothing, a proprietary feature extraction method based on percentiles of the empirical distribution of ratios of interval lengths and magnitudes of peaks and valleys, and Extreme Gradient Boosting Tree classifier, achieved an F 1 -Score of 0.988 on patients without additional cardiac conditions. The same noise reduction and feature extraction methods combined with Gradient Boosting Tree classifier achieved an F 1 -Score of 0.97 on patients with additional cardiac conditions. Our method achieved the highest classification accuracy (average 10-fold cross-validation F 1 -Score of 0.992) using an external validation data, MIT-BIH arrhythmia database. The proposed optimal multi-stage arrhythmia classification approach can dramatically benefit automatic ECG data analysis by providing cardiologist level accuracy and robust compatibility with various ECG data sources.ECGs represent the filtered electrical activity generated by the heart. An ECG from lead II presents a normal heartbeat under sinus rhythm that has a characteristic shape with three features, a P-wave presenting the atrial depolarization process, a QRS complex denoting the ventricular depolarization process, and a T-wave representing the ventricular repolarization. The normal feature sequence of the cardiac cycle is P-wave, QRS complex, and T-wave with sections between them called segments. Three such major segments are the PR, ST, and TP segments. Important periods within and between ECG waves are the PR, QT, and RR intervals.Damage to the heart muscle or nerves can change the electrical activity of the heart and induce a corresponding change in the shape of the ECGs. Thus, ECG is a major clinical diagnostic tool for various heart abnormalities. Arrhythmias are a family of conditions characterized by aberrations from the normal rate or rhythm of the heartbeats. There are several dozen classes of arrhythmia with various distinct manifestations, excessively slow or fast heartbeats such as sinus bradycardia and atrial tachycardia, irregular rhythm with missing or distorte...

show abstract

Detection of cardiac arrhythmia in electrocardiograms using adaptive feature extraction and modified support vector machines

Cited by 53 publications

References 35 publications

SVM Based ECG Classification Using Rhythm and Morphology Features, Cluster Analysis and Multilevel Noise Estimation

SVM Based ECG Classification Using Rhythm and Morphology Features, Cluster Analysis and Multilevel Noise Estimation

Monitoring and Detection Platform to Prevent Anomalous Situations in Home Care

Optimal Multi-Stage Arrhythmia Classification Approach

Contact Info

Product

Resources

About