{"title":"Knowledge-Based QRS Detection Performed by a Cascade of Moving Average Filters","authors":"L. Bachi, L. Billeci, M. Varanini","doi":"10.22489/CinC.2020.175","DOIUrl":"https://doi.org/10.22489/CinC.2020.175","url":null,"abstract":"The detection of QRS complexes is a crucial step since all the subsequent processing of the ECG signal is very sensitive to the accuracy of this detection. This study presents an accurate and computationally efficient approach to heartbeat detection based on preprocessing and enhancement of the QRS complexes by means of cascades of moving averages. Several derivative QRS-enhancing moving averages filters were defined which were characterized by different shapes of the impulsive response. In the initialization phase of the algorithm, the best filter for each record was selected by maximizing a specifically defined signal quality index. Detection of the QRS complex was based on a decision logic and a set of adaptive thresholds. The MIT-BIH, QTDB and EU ST-T databases were considered for performance evaluation and comparison with the output of some publicly available QRS Pan-Tompkins detectors, obtaining results comparable to the best reported in the literature (F1=99.84% and 98.46% on MIT-BIH channel 1 and 2 respectively).","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122822310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. W. Wong, Weijie Sun, S. Kalmady, P. Kaul, Abram Hindle
{"title":"Multilabel 12-Lead Electrocardiogram Classification Using Gradient Boosting Tree Ensemble","authors":"A. W. Wong, Weijie Sun, S. Kalmady, P. Kaul, Abram Hindle","doi":"10.22489/CinC.2020.128","DOIUrl":"https://doi.org/10.22489/CinC.2020.128","url":null,"abstract":"The 12-lead electrocardiogram (ECG) is a commonly used tool for detecting cardiac abnormalities such as atrial fibrillation, blocks, and irregular complexes. For the Phy-sioNet/CinC 2020 Challenge, we built an algorithm using gradient boosted tree ensembles fitted on morphology and signal processing features to classify ECG diagnosis. For each lead, we derive features from heart rate variability, PQRST template shape, and the full signal wave-form. We join the features of all 12 leads to fit an ensemble of gradient boosting decision trees to predict probabilities of ECG instances belonging to each class. We train a phase one set of feature importance determining models to isolate the top 1,000 most important features to use in our phase two diagnosis prediction models. We use repeated random sub-sampling by splitting our dataset of 43,101 records into 100 independent runs of 85:15 training/validation splits for our internal evaluation results. Our methodology generates us an official phase validation set score of 0.476 and test set score of − 0.080 under the team name, CVC, placing us 36 out of 41 in the rankings.","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122932725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chiara Celotto, C. Sánchez, Konstantinos A. Mountris, Mostafa Abdollahpur, Frida, Sandberg, P. Laguna, E. Pueyo
{"title":"Relationship Between Atrial Oscillatory Acetylcholine Release Pattern and f-wave Frequency Modulation: a Computational and Experimental Study","authors":"Chiara Celotto, C. Sánchez, Konstantinos A. Mountris, Mostafa Abdollahpur, Frida, Sandberg, P. Laguna, E. Pueyo","doi":"10.22489/CinC.2020.303","DOIUrl":"https://doi.org/10.22489/CinC.2020.303","url":null,"abstract":"The frequency of fibrillatory waves (f-waves), Ff, exhibits significant variation over time, and previous studies suggest that some of this variation is related to respiratory modulation through the autonomic nervous system. In this study, we tested the hypothesis that this variation (ΔFf) could be related to acetylcholine concentration ([ACh]) release pattern. Electrocardiograms were recorded from seven patients during controlled respiration before and after full vagal blockade, from which f-wave frequency modulation was characterized. Computational simulations in human atrial tissues were performed to assess the effects of [ACh] release pattern on Ff and compared to experimental results in humans. A cross-stimulation protocol was applied onto the tissue to initiate a rotor while cyclically varying [ACh] following a sinusoidal waveform of frequency equal to 0.125 Hz. Different mean levels (0.05, 0.075μM/l) and peak-to-peak ranges (0.1, 0.05, 0.025 μM/l) of [ACh] variation were tested. In all patients, an f-wave frequency modulation could be observed. In 57% of the patients, this modulation was significantly reduced after vagal blockade. Simulations confirmed that rotor frequency variations followed the induced [ACh] patterns. Mean Ff was dependent on mean [ACh] level, whileΔFfwas dependent on [ACh] variation range.","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123971395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
E. M. Cirugeda-Roldán, S. Calero, A. Quesada, V. M. Hidalgo, J. J. Rieta, R. Alcaraz
{"title":"Limb Versus Precordial ECG Leads as Improved Predictors of Electrical Cardioversion Outcome in Persistent Atrial Fibrillation","authors":"E. M. Cirugeda-Roldán, S. Calero, A. Quesada, V. M. Hidalgo, J. J. Rieta, R. Alcaraz","doi":"10.22489/CinC.2020.373","DOIUrl":"https://doi.org/10.22489/CinC.2020.373","url":null,"abstract":"Electrical cardioversion (ECV) is an effective and low-cost rhythm control strategy for persistent atrial fibrillation (AF). Because of its limited mid- and long-term success rates, prediction of early failure could avoid patients with reduced chance to maintain sinus rhythm (SR). To this end and due to its proximity to the right atrium, several indices characterizing atrial activity have been proposed based on lead V1. However, information from other leads has been discarded to date. Hence, this work studies how effective some common indices computed over the whole set of 12 standard ECG leads are in predicting ECVout-come. Precisely, amplitude, dominant frequency, and sample entropy were computed from the fibrillatory (f -) waves extracted for each one of 12 standard leads acquired before ECV from 58 patients in persistent AF. The classification between the patients who relapsed to AF and maintained sinus rhythm after a follow-up of 4 weeks achieved by these parameters was better from limb lead II than from V1, thus reporting improvements about 6 and 12%. As a consequence, characterization of f-waves from the more accessible limb lead II has proven to be the best choice to improve AF ECV outcome prediction from the ECG.","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129071094","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Hayn, M. Sareban, Alphons Eggerth, M. Falgenhauer, Angelika Rzepka, Heimo Traninger, K. Mayr, M. Philippi, M. Porodko, C. Puelacher, S. Höfer, J. Niebauer
{"title":"Telehealth Services for Home-based Rehabilitation of Cardiac Patients","authors":"D. Hayn, M. Sareban, Alphons Eggerth, M. Falgenhauer, Angelika Rzepka, Heimo Traninger, K. Mayr, M. Philippi, M. Porodko, C. Puelacher, S. Höfer, J. Niebauer","doi":"10.22489/CinC.2020.150","DOIUrl":"https://doi.org/10.22489/CinC.2020.150","url":null,"abstract":"Cardiovascular diseases (CVD) are the leading cause of death in the Western world. Several modifiable risk factors contribute to the pathogenesis of CVD which are all addressed during cardiac rehabilitation (CR). CR is conducted in three phases: I: acute care hospital, II: subsequent in- or outpatient CR, and III: out-patient CR with focus on lifelong prevention. Despite its proven merits, the adherence to healthy lifestyle changes following completion of CR phase II is challenging. This gap is addressed in recent recommendations, suggesting that clinicians should help patients to set personal goals to i) achieve and maintain the benefits of physical activity, ii) include physical activity into their daily routine and iii) overcome barriers to exercise, to achieve behavior change more effectively and more sustainably. We have developed tele-rehabilitation services to support patients during home-based exercise training in CR phase III Our services provide a link between CR experts and patients by means of individualized exercise prescription supported by different kinds of wearables for measuring e.g. physical activity volume. The effectiveness of such services and other supportive measures regarding adherence to home training plans and changes in exercise capacity during CR phase III CR is currently evaluated in a study.","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128752166","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An End-to-end Deep Learning Scheme for Atrial Fibrillation Detection","authors":"Yingjie Jia, Haoyu Jiang, Ping Yang, Xianliang He","doi":"10.22489/CinC.2020.106","DOIUrl":"https://doi.org/10.22489/CinC.2020.106","url":null,"abstract":"The aim of this study was the detection of atrial fibrillation (AF) from continuous ECG analysis. In this study, an end-to-end deep learning scheme was proposed. When the scheme was applied, 30-second multi-lead ECG data segments with an overlapping window of 5 seconds were preprocessed and sequentially fed into a multi-layer residual convolutional neural network (CNN) to extract ECG's multi-scale local morphological (spatial) features, The generated local spatial features were then processed by the following two bidirectional long short-term memory (LSTM) layers, and the output sequences of the LSTM layers were then weighted by an attention module and processed by a following dense network to complete AF detection. Finally, the sequential detection results were further processed to improve accuracy. To demonstrate its effectiveness, the proposed scheme was trained and tested on multiple ECG databases annotated by cardiologists. Episode and duration accuracies were calculated according to the performance evaluation method of atrial fibrillation detection defined in the EC57 standard [1]. An episode F1 score of 85.7% and a duration F1 score of 95.5% were achieved on the independent testing dataset.","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115984899","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mohamadamin Forouzandehmehr, Nicolò Cogno, Jussi T. Koivumäki, J. Hyttinen, M. Paci
{"title":"The Comparison Between Two Mathematical Contractile Elements Integrated into an hiPSC-CM In-silico Model","authors":"Mohamadamin Forouzandehmehr, Nicolò Cogno, Jussi T. Koivumäki, J. Hyttinen, M. Paci","doi":"10.22489/CinC.2020.055","DOIUrl":"https://doi.org/10.22489/CinC.2020.055","url":null,"abstract":"Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) are a valuable tool for in vitro drug testing and disease studies. As contractility has become one of the main experimental outputs, hiPSC-CMs in silico models should also feature the mechanisms of force generation. Thus, we integrated two contractile elements (CE), Rice2008 and Negroni2015, into Paci2020 hiPSC-CM model. The simulated force-Ca2+ relationships from skinned versions of the CEs revealed rather close pCa50 values for both CEs: 6.17 and 6.10, respectively for Rice2008 and Negroni2015. However, Hill's coefficients for the two curves were 7.30 and 3.6. The relationships agreed with in vitro data from human engineered heart tissues. Most of the biomarkers measured from simulated spontaneous action potentials (APs) and Ca2+ transients (CaTs) showed good agreement with in vitro data for both CEs. The active peak force observed in paced conditions (1 Hz) and at extracellular Ca2+ concentration ([Ca2+]o) of 1.8 mM was 0.011 nM/mm2 for Paci2020+Rice2008 and 0.57 mN/mm2 for Paci2020+Negroni2015. These values match, qualitatively with the 0.26 mN/mm2 peak force reported previously in vitro at [Ca2+]o=1.8 mM Our results set an opening to develop more sophisticated hiPSC-CM models featuring both electrophysiology and biomechanics.","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116030701","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. Qureshi, Aditi Roy, H. Chubb, A. Vecchi, O. Aslanidi
{"title":"Investigating Strain as a Biomarker for Atrial Fibrosis Quantified by Patient Cine MRI Data","authors":"A. Qureshi, Aditi Roy, H. Chubb, A. Vecchi, O. Aslanidi","doi":"10.22489/CinC.2020.212","DOIUrl":"https://doi.org/10.22489/CinC.2020.212","url":null,"abstract":"Atrial fibrillation (AF) is responsible for deterioration of left atrial (LA) mechanical function. Sinus rhythm (SR) can be restored by terminating AF using catheter ablation (CA) therapy. CA often targets fibrotic tissue by creating scar tissue which is similar to fibrosis. We propose the use of myocardial strain to identify regions of fibrosis and understand its role in atrial mechanics. Patient-specific LA models were reconstructed from Cine and Late Gadolinium Enhanced (LGE) MRI data for two groups of patients: AF and SR pre-CA. LGE intensities represented atrial fibrosis and feature-tracking was applied on the Cine images to produce a series of 3D deforming LA meshes. The myocardial area strain (MAS) was calculated as a measure of regional contractile ability. 24 regions of clinical interest were assigned for inter- and intra-patient comparisons on the effects of CA and fibrosis on LA mechanical function. Correlation was found between low strain and dense fibrosis in the LA posterior wall for both patient groups (rs = −0.74). MAS increased (8.9%) after CA in the AF group but decreased (10%) in the SR group. This study suggests that myocardial strain can be used as a biomarker for atrial fibrosis and also identifies the detrimental effect of intentional CA-induced damage to the LA on its mechanical function.","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114286200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jorge Sánchez, Mark Nothstein, A. Neic, Yung-Lin Huang, A. Prassl, J. Klar, Robert Ulrich, Felix Bach, Philipp Zschumme, M. Selzer, G. Plank, E. Vigmond, G. Seemann, A. Loewe
{"title":"openCARP: An Open Sustainable Framework for In-Silico Cardiac Electrophysiology Research","authors":"Jorge Sánchez, Mark Nothstein, A. Neic, Yung-Lin Huang, A. Prassl, J. Klar, Robert Ulrich, Felix Bach, Philipp Zschumme, M. Selzer, G. Plank, E. Vigmond, G. Seemann, A. Loewe","doi":"10.22489/CinC.2020.111","DOIUrl":"https://doi.org/10.22489/CinC.2020.111","url":null,"abstract":"openCARP is an open cardiac electrophysiology simulator, released to the community to advance the computational cardiology field by making state-of-the-art simulations accessible. It aims to achieve this by supporting self-driven learning. To this end, an online platform is available containing educational video tutorials, user and developer-oriented documentation, detailed examples, and a question-and-answer system. The software is written in C++. We provide binary packages, a Docker container, and a CMake-based compilation workflow, making the installation process simple. The software can fully scale from desktop to high-performance computers. openCARP runs nightly tests to ensure the consistency of the simulator based on predefined reference solutions, keeping a high standard of quality for all of its components. Additionally, sustainability is achieved through automated continuous integration to generate not only software packages, but also documentation and content for the community platform. Furthermore, carputils provides an environment for users to create complex, multi-scale simulations that are shareable and reproducible. In conclusion, openCARP is a tailored software solution for the scientific community in the cardiac electrophysiology field and contributes to increasing use and reproducibility of in-silico experiments.","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115303017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Classification of 12-lead ECGs Using Intra-Heartbeat Discrete-time Fourier Transform and Inter-Heartbeat Attention","authors":"Ibrahim Hammoud, I. Ramakrishnan, P. Djurić","doi":"10.22489/CinC.2020.307","DOIUrl":"https://doi.org/10.22489/CinC.2020.307","url":null,"abstract":"In this work, we built a model to classify 12-lead ECGs using attention for the PhysioNet/Computing in Cardiology Challenge 2020. Since information about different classification outcomes might be present only in specific segments, we tune our feature representation to show the frequency distribution shift as we move through time. This is done by first representing the original signal as a spectrogram, which shows the signal's frequency spectrum during different time windows (heartbeats). The frequency spectrum at each heartbeat is extracted using discrete-time Fourier transform. The spectrogram is then inputted to a bidirectional LSTM network where each heartbeat vector represents a time step. The outputs of the bidirectional LSTM network at each stage are then used as attention vectors. The attention vectors are then multiplied with the original signal window embeddings, which are used to generate the final output. Our approach achieved a challenge validation score of 0.416 and a test score of 0.024 but were not ranked due to omissions in the submission (team name: SBU_AI).","PeriodicalId":407282,"journal":{"name":"2020 Computing in Cardiology","volume":"727 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116063409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}