Abstract 12792: Enhancing the Accuracy of Cardiac Rhythm Analysis in Automated External Defibrillators During Ongoing Cardiopulmonary Resuscitation by Applying a Deep Encoder-Decoder Filtering Model

Circulation ◽  
2021 ◽  
Vol 144 (Suppl_2) ◽  
Author(s):  
Shirin Hajeb Mohammadalipour ◽  
Alicia Cascella ◽  
Matt Valentine ◽  
Ki Chon
Keyword(s):  
Cardiopulmonary Resuscitation ◽  
Signal To Noise Ratio ◽  
Normal Sinus Rhythm ◽  
Motion Artifact ◽  
Automated External Defibrillators ◽  
Normal Sinus ◽  
Before And After ◽  
Rhythm Analysis ◽  
Distinct Features ◽  
Ecg Data

Survival from out-of-hospital cardiac arrests depends on an accurate defibrillatory shock decision during cardiopulmonary resuscitation (CPR). Since chest compressions induce severe motion artifact in the electrocardiogram (ECG), current automatic external defibrillators (AEDs) do not perform CPR during the rhythm analysis period. However, performing continuous CPR is vital and dramatically increases the chance of survival. Hence, we demonstrate a novel application of a deep convolutional neural network encoder-decoder (CNNED) method in suppressing CPR artifact in near real-time using only ECG data. The encoder portion of the CNNED uses the frequency and phase contents derived via time-varying spectral analysis to learn distinct features that are representative of both the ECG signal and CPR artifact. The decoder portion takes the results from the encoder and reconstructs what is perceived as the motion artifact removed ECG data. These procedures are done via multitude of training of CNNED using many different arrhythmia contaminated with CPR. In this study, CPR-contaminated ECGs were generated by combining clean ECG with CPR artifacts from 52 different performers. ECG data from CUDB, VFDB, and SDDB datasets which belong to the Physionet’s Physiobank archive were used to create the training set containing 89,984 14-second ECG segments. The performance of the proposed CNNED was evaluated on a separate test set comprising of 23,816 CPR-contaminated 14-second ECG segments from 458 subjects. The results were evaluated by two metrics: signal-to-noise ratio (SNR), and Defibtech’s AED rhythm analysis algorithm. CNNED resulted in the increase of the mean SNR value from -3 dB to 5.63 dB and 6.3 dB for shockable and non-shockable rhythms, respectively. Comparing Defibtech’s AED rhythm classifier before and after applying CNNED on the CPR-contaminated ECG, the specificity improved from 96.57% to 99.31% for normal sinus rhythm, and from 91.5% to 96.57% for other non-shockable rhythms. The sensitivity of shockable detection also increased from 67.68% to 87.76% for ventricular fibrillation, and from 62.71% to 81.27% for ventricular tachycardia. These results indicate continuous and accurate AED rhythm analysis without stoppage of CPR using only ECG data.

scholarly journals Automated Condition-Based Suppression of the CPR Artifact in ECG Data to Make a Reliable Shock Decision for AEDs during CPR

Sensors ◽  
2021 ◽  
Vol 21 (24) ◽  
pp. 8210
Author(s):  
Shirin Hajeb-Mohammadalipour ◽  
Alicia Cascella ◽  
Matt Valentine ◽  
Ki H. Chon
Keyword(s):  
Performance Metrics ◽  
Signal To Noise Ratio ◽  
Normal Sinus Rhythm ◽  
Spectral Characteristics ◽  
Stop Band ◽  
Decision Algorithm ◽  
Shockable Rhythm ◽  
Normal Sinus ◽  
Rhythm Analysis ◽  
Ecg Data

Cardiopulmonary resuscitation (CPR) corrupts the morphology of the electrocardiogram (ECG) signal, resulting in an inaccurate automated external defibrillator (AED) rhythm analysis. Consequently, most current AEDs prohibit CPR during the rhythm analysis period, thereby decreasing the survival rate. To overcome this limitation, we designed a condition-based filtering algorithm that consists of three stop-band filters which are turned either ‘on’ or ‘off’ depending on the ECG’s spectral characteristics. Typically, removing the artifact’s higher frequency peaks in addition to the highest frequency peak eliminates most of the ECG’s morphological disturbance on the non-shockable rhythms. However, the shockable rhythms usually have dynamics in the frequency range of (3–6) Hz, which in certain cases coincide with CPR compression’s harmonic frequencies, hence, removing them may lead to destruction of the shockable signal’s dynamics. The proposed algorithm achieves CPR artifact removal without compromising the integrity of the shockable rhythm by considering three different spectral factors. The dataset from the PhysioNet archive was used to develop this condition-based approach. To quantify the performance of the approach on a separate dataset, three performance metrics were computed: the correlation coefficient, signal-to-noise ratio (SNR), and accuracy of Defibtech’s shock decision algorithm. This dataset, containing 14 s ECG segments of different types of rhythms from 458 subjects, belongs to Defibtech commercial AED’s validation set. The CPR artifact data from 52 different resuscitators were added to artifact-free ECG data to create 23,816 CPR-contaminated data segments. From this, 82% of the filtered shockable and 70% of the filtered non-shockable ECG data were highly correlated (>0.7) with the artifact-free ECG; this value was only 13 and 12% for CPR-contaminated shockable and non-shockable, respectively, without our filtering approach. The SNR improvement was 4.5 ± 2.5 dB, averaging over the entire dataset. Defibtech’s rhythm analysis algorithm was applied to the filtered data. We found a sensitivity improvement from 67.7 to 91.3% and 62.7 to 78% for VF and rapid VT, respectively, and specificity improved from 96.2 to 96.5% and 91.5 to 92.7% for normal sinus rhythm (NSR) and other non-shockables, respectively.

scholarly journals Ventricular fibrillation detection algorithm for automated external defibrillators

2021 ◽  
Author(s):  
Marjan M. Kusha
Keyword(s):  
Ventricular Fibrillation ◽  
Distance Measure ◽  
Normal Sinus Rhythm ◽  
Heart Association ◽  
Detection Algorithm ◽  
Detection Accuracy ◽  
Automated External Defibrillators ◽  
Normal Sinus ◽  
Dominant Pole ◽  
Ar Modeling

The automatic external defibrillator (AED) is a lifesaving device, which processes and analyzes the electrocardiogram (ECG) and prompts a defibrillation shock if ventricular fibrillation is determined. This project investigates the possibility of developing a Ventricular Fibrillation (VF) detection algorithm based on Autoregressive Modeling (AR Modeling) and dominant poles for the use in AEDs. In particular, the ECG segment is modeled using AR modeling and the dominant poles are extracted from the model transfer function. The dominant pole frequencies were then used in classification based on the distance measure. The potential use of this method to distinguish between VF and Normal sinus rhythm (NSR) is discussed. The method was tested with ECG records from the widely recognized databases of American Heart Association (AHA) and the Creighton University (CU). Sensitivity and specificity for the new VF detection method were calculated to be 66% and 94% respectively. The proposed method has some advantages over other existing VF detection algorithms; it has a high detection accuracy, it is computationally inexpensive and can be easily implemented in hardware.

Left ventricular resynchronization therapy in a canine model of left bundle branch block

2002 ◽  
Vol 282 (6) ◽  
pp. H2238-H2244 ◽  
Author(s):  
Lili Liu ◽  
Bruce Tockman ◽  
Steven Girouard ◽  
Joseph Pastore ◽  
Greg Walcott ◽  
...  
Keyword(s):  
Left Bundle Branch Block ◽  
Wall Motion ◽  
Normal Sinus Rhythm ◽  
Posterior Wall ◽  
Canine Model ◽  
Left Ventricular ◽  
Pressure Derivative ◽  
Bundle Branch Block ◽  
Normal Sinus ◽  
Before And After

Positive responses to left (LV) and biventricular (BV) stimulation observed in heart failure patients with left bundle branch block (LBBB) suggest a possible mechanism of LV resynchronization. An anesthetized canine LBBB model was developed using radio frequency ablation. Before and after ablation, LV pressure derivative over time (dP/d t) and aortic pulse pressure (PP) were assessed during normal sinus rhythm with right ventricle (RV), LV, or BV stimulation combined with four atrioventricular delays in six dogs. In three more dogs, M-mode echocardiograms of septal and LV posterior wall motion were obtained before and after LBBB and during LV stimulation. LBBB caused QRS widening and hemodynamics deterioration. Before ablation, stimulation alone worsened LV dP/d t and PP. After ablation, LV and BV stimulation maximally increased LV dP/d t by 16% and PP by 7% ( P < 0.001), whereas little improvement was observed during RV stimulation. M-mode echocardiogram showed that LBBB resulted in a paradoxical septal wall motion that was corrected by LV stimulation. In conclusion, LV and BV stimulation improved cardiac function in a canine LBBB model via resynchronization of LV excitation and contraction.

scholarly journals Robust Discrimination of Normal Sinus Rhythm and Atrial Fibrillation on ECG Using a Multiscale Frequency Technique

2017 ◽  
Author(s):  
Shivaram Poigai Arunachalam ◽  
Elizabeth M. Annoni ◽  
Suraj Kapa ◽  
Siva K. Mulpuru ◽  
Paul A. Friedman ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Sinus Rhythm ◽  
Time Series Data ◽  
Normal Sinus Rhythm ◽  
The Body ◽  
Series Data ◽  
Robust Detection ◽  
Normal Sinus ◽  
Surface Ecg ◽  
Ecg Data

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia affecting approximately 3 million Americans, and is a prognostic marker for stroke, heart failure and even death [1]. 12-lead electrocardiogram (ECG) is used to monitor normal sinus rhythm (NSR) and also detect AF. Although the persistent form of AF can be detected relatively easy, detecting paroxysmal AF is often a challenge since requiring continuous monitoring, which becomes expensive and cumbersome to collect lot of ECG data [1]. Several researchers have attempted to develop new methods to discriminate NSR and AF which are based on R-R interval analysis, linear methods, filtering, spectral analysis, statistical approaches such as entropy etc. which faces limitation of successfully detecting AF of all types with high sensitivity and specificity using short time ECG data [1–3]. The major issues with these approaches is that they often distort the ECG by several pre-processing steps with filters, do not provide reliable discrimination using short ECG time series data and many of them lack real-time capability that makes it difficult to trust the data for diagnosis and treatment. Both clinical and scientific communities recognize these difficulties and the necessity to develop novel methods that can enable accurate monitoring and detection of AF [2]. In addition, robust detection and classification algorithms are essential for delivering appropriate therapy for implantable cardioverter defibrillators (ICD) to provide lifesaving timely action. In this work, the authors propose and demonstrate the application of a multiscale frequency (MSF) approach [4] for accurate detection and discrimination between AF and NSR ECG traces taken from publically available Physionet database. The MSF approach takes into account the contribution from various frequencies in ECG and thus yield valuable information regarding the chaotic nature of AF. Therefore, we demonstrate that MSF can capture the complexity of AF which is associated with higher MSF value compared with NSR thus enabling robust discrimination e AF manifests itself with numerous chaotic frequencies within the body surface ECG,. We validate the feasibility of this technique to discriminate NSR from AF.

scholarly journals Ventricular fibrillation detection algorithm for automated external defibrillators

2021 ◽  
Author(s):  
Marjan M. Kusha
Keyword(s):  
Ventricular Fibrillation ◽  
Distance Measure ◽  
Normal Sinus Rhythm ◽  
Heart Association ◽  
Detection Algorithm ◽  
Detection Accuracy ◽  
Automated External Defibrillators ◽  
Normal Sinus ◽  
Dominant Pole ◽  
Ar Modeling

The automatic external defibrillator (AED) is a lifesaving device, which processes and analyzes the electrocardiogram (ECG) and prompts a defibrillation shock if ventricular fibrillation is determined. This project investigates the possibility of developing a Ventricular Fibrillation (VF) detection algorithm based on Autoregressive Modeling (AR Modeling) and dominant poles for the use in AEDs. In particular, the ECG segment is modeled using AR modeling and the dominant poles are extracted from the model transfer function. The dominant pole frequencies were then used in classification based on the distance measure. The potential use of this method to distinguish between VF and Normal sinus rhythm (NSR) is discussed. The method was tested with ECG records from the widely recognized databases of American Heart Association (AHA) and the Creighton University (CU). Sensitivity and specificity for the new VF detection method were calculated to be 66% and 94% respectively. The proposed method has some advantages over other existing VF detection algorithms; it has a high detection accuracy, it is computationally inexpensive and can be easily implemented in hardware.

Hemodynamics and Echocardiograms Before and After Cardioversion of Atrial Fibrillation to Normal Sinus Rhythm

CHEST Journal ◽  
1979 ◽  
Vol 76 (5) ◽  
pp. 521-526 ◽  
Author(s):  
Joan R. Orlando ◽  
Robert van Herick ◽  
Wilbert S. Aronow ◽  
Harold G. Olson
Keyword(s):  
Atrial Fibrillation ◽  
Sinus Rhythm ◽  
Normal Sinus Rhythm ◽  
Normal Sinus ◽  
Before And After

scholarly journals Enhanced Local Patterns Using Deep Learning Techniques for ECG Based Identity Recognition System

2021 ◽  
Author(s):  
Lotfi Mostefai ◽  
Benouis Mohamed ◽  
Denai Mouloud ◽  
Bouhamdi Merzoug
Keyword(s):  
Deep Learning ◽  
Normal Sinus Rhythm ◽  
Recognition System ◽  
Learning Approaches ◽  
Identity Recognition ◽  
Ecg Signals ◽  
Normal Sinus ◽  
Learning Techniques ◽  
Local Patterns ◽  
Distinct Features

Abstract Electrocardiogram (ECG) signals have distinct features of the electrical activity of the heart which are unique among individuals and have recently emerged as a potential biometric tool for human identification. The paper attempts to address the problem of ECG identification based on non-fiducial approach using unsupervised classifier and a Deep Learning approaches. This work investigates the ability of local binary pattern to extract the significant pattern/feature that describes the heartbeat activity for each person’s ECG and the use of staked autoencoders and deep belief network to further enhance the extracted features and classify them based on their heartbeat activity. The proposed approach is validated using experimental datasets from two publicly available databases MIT-BIH Normal Sinus Rhythm and ECG-ID and the results demonstrate the effectiveness of this approach for ECG-based human authentication.

Sign in / Sign up

Export Citation Format

Share Document