P-wave Detection Research Articles

Abstract P504: Quality Validation Of Different ECG Leads, And An Automatic P-wave Detection Algorithm For Af, Based On Real-world Data

Background: Single-lead ECG has shown in research to be affected by artifacts leading to lower diagnostic yield of Atrial Fibrillation (AF). Use of multiple ECG leads and algorithms for detection of AF has shown to increase detection accuracy and reduce false positives. Methods: A novel RR- and P-wave based automatic algorithm implemented in the 2-lead Coala Heart Monitor (Coala) was evaluated for detection accuracy and quality by the comparison to blinded manual ECG interpretation. Evaluation was conducted on 100 consecutive anonymous printouts of chest- and thumb-ECG waveforms, where both an irregular RR-rhythm and strong P-waves in either chest or thumb recording were detected.The recordings, without exclusions, were generated from 5,512 real-world data recordings from actual Coala users in Sweden (both OTC and Rx users) during the period of March 5 to March 22, 2019, with no control or influence by the researchers or any other organization or individual. The prevalence of cardiac conditions in the user population was unknown. The blinded recordings were each manually interpreted and assessed for quality by a trained cardiologist. The manual interpretation was compared with the automatic analysis performed by the cloud-based detection algorithm to determine the detection quality of the respective ECG leads. Results: Strong P-waves were detected more often in the chest ECG as compared to the thumb ECG (90 vs 32 recordings). The assessed quality of the ECG tracings was higher in the chest ECGs as compared to the thumb ECGs (4.61 vs 3.88). Irregular RR-rhythms were detected in 569 recordings (10 % of a total of 5,512 recordings), the 100 non-AF recordings containing both irregular RR-rhythms and strong P-waves thus constituted 18% of all recordings with irregular RR-rhythms. Non-pathological rhythm (normal) was present in 84% of the recordings although all of these recordings contained irregular rhythm disturbances (respiratory sinus arrhythmia, PAC/PVC etc). Respiratory sinus arrhythmia was the single most prevalent condition and found in 47% of the recordings with irregular RR-rhythms with strong detected P-waves. Conclusion: The combination of chest and thumb ECG for detection of AF by an automatic P-wave based algorithm is shown to be more than 300% superior to thumb ECG alone with the majority of automatically detected P-waves and highest assessed ECG quality in the chest recordings.

Baseline Assessment of the Importance of Contributions from Regional Seismic Networks to the Pacific Tsunami Warning Center’s Operations

Abstract The Pacific Tsunami Warning Center’s (PTWC’s) core mission consists in saving lives and minimizing property damage through the issuance of timely, effective tsunami warnings and threat messages. For this purpose, the center relies primarily not only on the analysis of near-real-time seismic data streams provided by established networks such as the Global Seismic Network (GSN), but also by an increasing number of stations contributed by regional seismic networks (RSNs) from around the world. We used theoretical computations of the detection time of the first arriving P wave as a proxy to highlight the areas where RSN contributions have the greatest operational impact. To this goal, the P-wave detection times computed for the GSN provided a baseline to isolate the contribution of the RSN in the form of spatially distributed detection time gains. Inspection of the resulting global maps reveals detection time gains of more than 3 m for Hawaii, Alaska, the east and west coasts of the United States, South America, New Zealand, the southwest Pacific, Australia, and the Sunda arc. Despite higher density of GSN baseline stations across the U.S. west coast and the eastern Caribbean, contributions from RSNs result in detection time gains of more than 2 m. These gains in earthquake detection speed correlate well with the continuous increase in the number of stations ingested into the PTWC system and the gradual reduction of the operational median response time to under 6 m from origin during the last 5 yr. These results allow us to conclude that fulfillment of the PTWC’s international and domestic tsunami warning responsibilities critically depends on the support of the RSNs contributing data to PTWC’s daily operations.

P-Wave Detection Using a Fully Convolutional Neural Network in Electrocardiogram Images

Electrocardiogram (ECG) signal analysis is a critical task in diagnosing the presence of any cardiac disorder. There are limited studies on detecting P-waves in various atrial arrhythmias, such as atrial fibrillation (AFIB), atrial flutter, junctional rhythm, and other arrhythmias due to P-wave variability and absence in various cases. Thus, there is a growing need to develop an efficient automated algorithm that annotates a 2D printed version of P-waves in the well-known ECG signal databases for validation purposes. To our knowledge, no one has annotated P-waves in the MIT-BIH atrial fibrillation database. Therefore, it is a challenge to manually annotate P-waves in the MIT-BIH AF database and to develop an automated algorithm to detect the absence and presence of different shapes of P-waves. In this paper, we present the manual annotation of P-waves in the well-known MIT-BIH AF database with the aid of a cardiologist. In addition, we provide an automatic P-wave segmentation for the same database using a fully convolutional neural network model (U-Net). This algorithm works on 2D imagery of printed ECG signals, as this type of imagery is the most commonly used in developing countries. The proposed automatic P-wave detection method obtained an accuracy and sensitivity of 98.56% and 98.78%, respectively, over the first 5 min of the second lead of the MIT-BIH AF database (a total of 8280 beats). Moreover, the proposed method is validated using the well-known automatically and manually annotated QT database (a total of 11,201 and 3194 automatically and manually annotated beats, respectively). This results in accuracies of 98.98 and 98.9%, and sensitivities of 98.97 and 97.24% for the automatically and manually annotated QT databases, respectively. Thus, these results indicate that the proposed automatic method can be used for analyzing long-printed ECG signals on mobile battery-driven devices using only images of the ECG signals, without the need for a cardiologist.

충격 추진력 변화의 표준편차를 이용한 지진 P파 검출 알고리즘

In this paper, we propose a new seismic P-wave detection algorithm using the variation of impact movement and the variable threshold. In the proposed algorithm, the seismic P-wave is detected using a sudden change in the impact momentum calculated from the seismic jerk. The threshold to be the reference for sudden change is set from the mean and the standard deviation of the impact momentum changes. In addition, the mean and standard deviation estimated by the iterative method adjust the threshold value according to the SNR of the seismic wave. Using the observed real earthquakes and measured noise in the living area, computer simulations were performed to compare the conventional algorithms with the seismic P-wave detection performance of the proposed algorithm. From the experimental results, it is possible to confirm the superior detection performance of the proposed method and the effectiveness of the threshold adjustment.

P580Performance evaluation of dual vs. single lead automatic, real-world arrhythmic ECG recordings

Abstract Background Detection of treatment-demanding paroxysmal arrhythmias can be difficult. Use of single lead ECG recordings for detection of Atrial Fibrillation (AF) has shown to lead to high false positive discovery rates, and requirements of additional manual clinical interpretation or 12-lead ECG to reach satisfactory diagnostic precision. Use of sequential, dual-lead ECG recordings for detection of AF has indicated significant clinical improvement in detection yield of AF and reduction of false positive discovery rates as compared to single lead-ECG. Methods The Coala Heart Monitor (Coala Life AB, Stockholm Sweden) system was evaluated by manual interpretation of 1,000 consecutive anonymous printouts of chest- and thumb-ECG waveforms, without any exclusion. The anonymized printouts contained three 10 s. strips of ECG at 25 mm/s, including mean heart rate, RR median and any user-provided annotation but with personal identification and algorithm analysis results blinded. The recordings were derived from actual Coala users with no training, control or influence, under a defined time period. The prevalence of cardiac conditions in the user population was unknown. The blinded recordings were manually interpreted by a trained cardiologist. The interpretation was compared with the automatic analysis performed by an enhanced algorithm in the Coala Cloud to evaluate ECG signal performance and to calculate performance metrics of a combination of sequential dual chest- and thumb-ECG measurements as compared to the clinical performance metrics of singe lead ECG measurements only. Results Metric Result using dual chest- and thumb-ECG and P-wave detection algorithm Result based on single lead thumb-ECG only Prevalence of AF in the recordings 14.4% (143 of 990 recordings) 14.4% (143 of 990 recordings) Percentage of recordings with ECG quality good enough to permit manual interpretation. 99,2% (8 of 998 non-readable) 98,7% (13 of 998 non-readable) Negative predictive value (NPV) for detecting AF 0.992 0.995 Positive predictive value (PPV) for detecting AF 0.872 0.647 False positive discovery rate 12.8% 35,3% Sensitivity for detecting AF 0.951 0.972 Specificity for detecting AF 0.976 0.909 Accuracy 0.973 0.919 Conclusion Based on 1,000 real-world recordings the combination of dual chest- and thumb-ECG with P-wave detection yielded significantly higher Positive Predictive Values as compared to single lead ECG measurements. Dual chest- and thumb-ECG reduced false positive AF indications from 35,3% for single lead-ECG to 12,8% using combined chest- and thumb-ECG with P-wave detection.

P1456Evaluation of an enhanced, cloud-based AF-detection algorithm based on real-world arrhythmic recordings

Abstract Background Cloud-based solutions offer the ability to centrally and continuously enhance detection algorithms for arrhythmias such as Atrial Fibrillation (AF) based on generated data. Methods The Coala Heart Monitor (Coala) system was evaluated by manual interpretation of 1,000 consecutive anonymous printouts of chest- and thumb-ECG waveforms, without any exclusion. The anonymized printouts were blinded from algorithm analysis, apart from gender and age within a 10-year span. The recordings were derived from actual Coala users in Sweden with no training, control or influence, under a defined time period. The prevalence of cardiac conditions in the user population was unknown. The blinded recordings were manually interpreted by a trained cardiologist. The interpretation was compared with the automatic analysis performed by an algorithm in the Coala Cloud to evaluate ECG signal performance and calculate performance metrics. An enhanced algorithm utilizing P-wave detection was then evaluated on the data set and compared with the performance metrics of the existing algorithm. Results Metric Results with current algorithm Results with enhanced algorithm Prevalence of AF in the recordings 14.4% (143 of 990 recordings) 14.4% (143 of 990 recordings) Sensitivity for detecting AF 0.972 (95% CI = 0.930–0.992) 0.951 Specificity for detecting AF 0.946 (95% CI = 0.928–0.960) 0.976 Negative predictive value for detecting AF 0.995 (95% CI = 0.987–0.999) 0.992 Positive predictive value for detecting AF 0.751 (95% CI = 0.683–0.812) 0.872 Accuracy 0.950 0.973 Conclusion The enhanced algorithm was found to improve the Positive Predictive Value for detecting AF as compared to the existing algorithm (0.872 vs 0.751). The reduced sensitivity for the enhanced algorithm was due to 3 consecutive recordings from a single individual who had misplaced the Coala with corresponding altered morphology of the ECG signal. The recordings were still reported as having an irregular rhythm by the algorithm. The evolution demonstrates that cloud-based systems offer an ability to enhance detection accuracy by using reference data to train algorithms.

An STFT-LSTM System for P-Wave Identification

The identification of P-waves in data with substantial noises is essential in seismic phase recognition, which is the foundation of seismic localization and an essential part of automatic seismic monitoring and processing. We treat the identification of P-waves as a sequence classification problem and build a long short-term memory (LSTM) network to model the difference between P-waves and noises. LSTM is designed to process time series data and has the ability to model the connections among short and long time steps; such temporal changes are characteristic of P-waves. In addition to temporal changes, spectral characteristics also distinguish between P-waves and noises. To leverage this difference, we introduce the short-time Fourier transform (STFT) at the front-end of the network to retain the implicit information, as well as to reduce the burden of the network. The results show that a deep neural network with STFT fails to function when the P-waves are not strong, but in the proposed STFT–LSTM system, the average recall for the P-wave detection is 90%, and the precision is approximately 75%. These results suggest that the proposed STFT–LSTM system is capable of distinguishing P-waves from background noises.

Effectiveness of the Chest Strap Electrocardiogram to Detect Atrial Fibrillation

Atrial fibrillation (AF) is a significant cause of cardioembolic strokes. AF is often symptomless and intermittent, making its detection challenging. The aim of this study was to assess the possibility to use a chest strap (Suunto Movesense) to detect AF both by cardiologists and automated algorithms. A single channel electrocardiogram (ECG) from a chest strap of 220 patients (107 AF and 111 sinus rhythm SR with 2 inconclusive rhythms) were analyzed by 2 cardiologists (Doc1 and Doc2) and 2 different algorithms (COSEn and AFEvidence). A 3-lead Holter served as the gold standard ECG for rhythm analysis. Both cardiologists evaluated the quality of the chest strap ECG to be superior to the quality of the Holter ECG; p <0.05/p <0.001 (Doc1/Doc 2). Accurate automated algorithm-based AF detection was achieved with sensitivity of 95.3%/96.3% and specificity of 95.5/98.2% with 2 AF detection algorithms from chest strap and 93.5%/97.2% and 98.2%/95.5% from Holter, respectively. P waves were detectable in 93.7% (Doc1) and 94.6% (Doc2) of the cases from the chest strap ECG with sinus rhythm and 98.2% (Doc1) and 95.5% (Doc2) from the Holter (p = n.s). In conclusion, the ECGs from both methods enabled AF detection by a cardiologist and by automated algorithms. Both methods studied enabled P-wave detection in sinus rhythm.

Real-Time Classification of Earthquake using Deep Learning

Existing Earthquake Early Warning Systems (EEWSs) calculates the location and magnitude of an earthquake using real-time waveforms from seismic stations within a few seconds. Typically, three to six stations are necessary to estimate earthquake parameters. Waiting for primary (P-) wave information from closest stations results in a blind-zone area where the arrival of secondary (S-) wave cannot be provided around the epicenter of an earthquake. If an earthquake occurred under a city center, EEWSs would not work even though each building has a seismic sensor in a smart city in future. Here, we present a methodology to classify earthquake vibrations into near-source or far-source within one second after P-wave detection. This will allow warnings to citizens who are the residence of earthquake epicenter in case of an earthquake very close by. We trained a deep learning Long Short-Term Memory (LSTM) network for sequence-to-label classification. 305 three component accelerations recorded between 2000 and 2018 in Japan are used to train the artificial network by extracting thirteen features of one second of P-wave. The accuracy of the methodology is 98.2%. 54 out of 55 near-source waveforms classified correctly and only 2 of 80 waveforms were misclassified. We tested the LSTM network with 2018 Northern Osaka (M 6.1.) earthquakes in Japan where closest stations are correctly identified with 83.3% accuracy. Therefore, smart cities donated with smart automated shut-on/off machines and sensors will be more resilient against earthquake disaster even EEWSs are not available in the blind zone area in future.

Multi-window detection for P-wave in electrocardiograms based on bilateral accumulative area

Background and objectiveP-wave detection is one of the most challenging aspects in electrocardiograms (ECGs) due to its low amplitude, low frequency, and variable waveforms. This work introduces a novel multi-window detection method for P-wave delineation based on the bilateral accumulative area. MethodThe bilateral accumulative area is calculated by summing the areas covered by the P-wave curve with left and right sliding windows. The onset and offset of a positive P-wave correspond to the local maxima of the area detector. The position drift and difference in area variation of local extreme points with different windows are used to systematically combine multi-window and 12-lead synchronous detection methods, which are used to screen the optimization boundary points from all extreme points of different window widths and adaptively match the P-wave location. ResultsThe proposed method was validated with ECG signals from various databases, including the Standard CSE Database, T-Wave Alternans Challenge Database, PTB Diagnostic ECG Database, and the St. Petersburg Institute of Cardiological Technics 12-Lead Arrhythmia Database. The average sensitivity Se was 99.44% with a positive predictivity P+ of 99.37% for P-wave detection. Standard deviations of 3.7 and 4.3ms were achieved for the onset and offset of P-waves, respectively, which is in agreement with the accepted tolerances required by the CSE committee. ConclusionCompared with well-known delineation methods, this method can achieve high sensitivity and positive predictability using a simple calculation process. The experiment results suggest that the bilateral accumulative area could be an effective detection tool for ECG signal analysis.

단계별 기저선 정렬을 이용한 ECG 신호에서 P파와 T파 검출 알고리즘

The detection of P-waves and T-wave in the electrocardiogram signal analysis is an important issue. But the accuracy of the boundary detection algorithm is an insufficient level in the change of slow transition in the signal compared to the QRS complex. This study proposes an algorithm to detect P-wave and T-wave sequentially after determining local baseline using QRS complex. First, we detected the peak points based on local baseline and determined the onset and offset through the calculation of the area of the section. After modifying the baseline using detected waveform, we detected the other waveform in the same way and separated the P-wave and the T-wave based on the location. We used the PhysioNet QT database to evaluate the performances of the algorithm, and calculate the mean and the standard deviations. The experiment results show that standard deviations are under the tolerances accepted by expert physicians, and outperform the results obtained by the other algorithms.

Research and Development of Electrocardiogram P-wave Detection Technology

In this study, a brief overview on the ECG P-wave automatic detection was presented, including the P-wave of ECG preprocessing techniques and several common detection methods. In addition, some types of analyses and compari- son of detection algorithms were carried out. Finally ECG P-wave direction was identified.

A new multi-window detection approach for P-wave boundary points in electrocardiograms based on bilateral accumulative area.

This study presented an efficient and robust multi-window detection method for P-wave boundary points in electrocardiograms on the basis of bilateral accumulative area. Through mathematical analysis, the local extreme points of bilateral accumulative area curves were respectively found in a fitting parabola, which might be regarded as a significant indicator for the morphological characteristic of boundary points. And, bilateral accumulative area curves of different window lengths had different sensitivities to the details and contours of signals. With combination of the multi-window and 12-lead synchronous detection, the proposed method could screen the optimal boundary points from all extreme points of different window lengths, and adaptively match the P-wave location. The results of the proposed method were evaluated on the dataset-3 of the standard CSE database. As a result, the value of sensitivity Se = 97.8% was obtained for the detection of P-wave, with the standard deviations of 3.9 ms and 4.8 ms respectively for the onset and offset of P-wave.

Comparison of Characteristics of P-Wave Detection in ECG with Wireless Patch Electrodes

P-wave characteristic in the human electrocardiogram (ECG) is important in the diagnosis of atrial conduction pathology. In this paper, we measured an ECG signal from patient with cardiovascular disease using one lead ECG electrode system which is based on the wireless cardiac monitoring system. And we detected a P-wave in ECG signal using the complex-valued continuous wavelet transforms (CWT) according to two kinds of patch type electrodes such as an existing narrow patch type electrode and the improved wide patch type electrode presented in this paper. Also, we compared the characteristics in detecting the P-wave in terms of the magnitude and the width of P-waves. From the results of comparison we found that the width and the magnitude of P-wave detected using the wide patch type electrode is improved to be interpreted easier compared to those using the narrow patch type electrode. Furthermore, we have also proven that the complex-valued CWT can be used as a robust detector for P-wave in ECG signal analysis.

Designing a Network-Based Earthquake Early Warning Algorithm for California: ElarmS-2

The California Integrated Seismic Network (CISN) is developing an earthquake early warning (EEW) demonstration system for the state of California. Within this CISN ShakeAlert project, three algorithms are being tested, one of which is the network-based Earthquake Alarm Systems (ElarmS) EEW system. Over the last three years, the ElarmS algorithms have undergone a large-scale reassessment and have been recoded to solve technological and methodological challenges. The improved algorithms in the new production-grade version of the ElarmS version 2 (referred to as ElarmS-2 or E2) code maximize the current seismic network's configuration, hardware, and software per- formance capabilities, improving both the speed of the early warning processing and the accuracy of the warnings. E2 is designed as a modular code and consists of a new event monitor module with an improved associator that allows for more rapid association with fewer triggers, while also adding several new alert filter checks that help minimize false alarms. Here, we outline the methodology and summarize the performance of this new online real-time system. The online performance from 2 October 2012 to 15 February 2013 shows, on average, ElarmS currently issues an alert 8:68 � 3:73 sa fter the first P-wave detection for all events across California. This time is reduced by 2 s in regions with dense station instrumentation. Standard deviations of magnitude, origin time are 0.4 magnitude units, 1.2 s, and the median location errors is 3.8 km. E2 successfully detected 26 of 29 earthquakes (MANSS >3:5) across California, while issuing two false alarms. E2 is now delivering alerts to ShakeAlert, which in turn distributes warnings to test users.

Innovative P-wave detection for discrimination between ventricular and supraventricular tachycardia in single-chamber ICDs: is the P-wave invisible during tachycardia?

Differentiation between supraventricular tachycardia (SVT) and ventricular tachycardia (VT) remains a substantial clinical challenge in patients with single-chamber implantable cardioverter-defibrillators (ICDs) due to absence of visible P waves. Innovative optimization of intrathoracic electrogram (EGM) configuration will facilitate P-wave detection and rhythm differentiation during tachycardia. Innovative optimization of EGM configuration was originally performed to improve patient care. In this retrospective cohort study, we examined our database for records of 140 consecutive patients undergoing single-chamber ICD implantation. During the follow-ups of 61 included patients with optimized EGM configuration, 27 patients were identified to have VT and/or SVT. EGMs in the Can (generator) to superior vena cava (Can-SVC) configuration were compared with those conventionally from the Can to right ventricular coil (Can-RV coil) source in the same patients. In Can-SVC EGMs, the ratio of P/QRS amplitude was 14-fold higher (0.57 ± 0.08 vs. 0.04 ± 0.00, P < 0.001) compared with those in Can-RV coil EGMs during sinus rhythm. With Can-SVC configuration, the odds of atrioventricular dissociation detection in patients with VT was increased 15-fold (61.9% vs. 9.5% with Can-RV coil; odds ratio, 15.4; 95% confidence interval, 2.8 to 84.7; P = 0.0009). In patients with SVT, P-waves or retrograde P-waves were markedly more identifiable in Can-SVC configuration compared with Can-RV coil (odds ratio, 40; 95% confidence interval, 3.6 to 447.1; P = 0.0010). P-wave recognition by optimizing EGM configuration provides a novel diagnostic tool for differentiation between VT and SVT in single-chamber ICDs. A potential discrimination algorithm would provide a cost-effective approach to improving the qualitative outcomes.

A modified left thoracotomy approach for enhancing ECG P-wave detection in the cynomolgus monkey

A modified left thoracotomy approach for enhancing ECG P-wave detection in the cynomolgus monkey

시간-주파수 누적 변화량과 가변 임계값을 이용한 지진 이벤트 자동 검출 알고리즘

This paper presents a new approach for the detection of seismic events using accumulated changes on time-frequency domain and variable threshold. To detect seismic P-wave arrivals with rapidness and accuracy, it is that the changes on the time and the frequency domains are simultaneously used. Their changes are parameters appropriated to reflect characteristics of earthquakes over moderate magnitude( magnitude 4.0) and microearthquakes. In addition, adaptively controlled threshold values can prevent false P-wave detections due to low SNR. We tested our method on real earthquakes those have various magnitudes. The proposed algorithm gives a good detection performance and it is also comparable to STA/LTA algorithm in computational complexity. Computer simulation results shows that the proposed algorithm is superior to the conventional popular algorithm (STA/LTA) in the seismic P-wave detection.

High-Accuracy Characterization of Ambulatory Holter Electrocardiogram Events

In this paper, three noise-robust high-accuracy methods aiming at the detection and delineation of the electrocardiogram (ECG) events (QRS complex, P-wave, T-wave) were developed. The ECG signal was initially appropriately preprocessed by application of a bandpass FIR filter and Discrete Wavelet Transform (DWT). The first detection-delineation method was the Walsh-Hadamard Transform (WHT). The WHT coefficients were divided into two groups and the signal was reconstructed using the second group coefficients. By this reconstruction, the values of first derivative of events are made stronger rather than the values of other parts of signal. In the second method, a feed forward artificial neural network was implemented to detect all events of the ECG signal. In the third method, the first derivative of signal was computed using a new signal smoothing algorithm with corresponding statistical properties. For decreasing False Positive (FP) errors associated with P-wave detection, a discriminating border was introduced as the post processing stage specified by three QRS parameters: the duration of a QRS complex, the time distance from the former and latter QRS complexes, and the potential difference from former QRS complex J-location and the latter QRS complex fiducial location. The proposed methods were applied to DAY general hospital high resolution holter data.

Automatic Detection of QRS Complex, P-Wave and T-Wave in the Electrocardiogram

Automatic Detection of QRS Complex, P-Wave and T-Wave in the Electrocardiogram