Your browser doesn't support javascript.
loading
Mostrar: 20 | 50 | 100
Resultados 1 - 20 de 180.002
Filtrar
2.
Open Heart ; 7(2)2020 10.
Artigo em Inglês | MEDLINE | ID: mdl-33020258

RESUMO

BACKGROUND: Cardiac involvement with COVID-19 is increasingly being recognised. Clinical characteristics and outcomes of patients with COVID-19 complicated by secondary Takotsubo cardiomyopathy (TC) is poorly understood. METHODS: This retrospective case series was conducted between March and April 2020 at four hospitals of Steward Health Care Network of Massachusetts, USA. Seven patients out of 169 who had echocardiogram were identified to have features of TC. Demographic, clinical, laboratory, management and outcome were gathered from their electronic medical records. We also reviewed all the published cases of COVID-19 and TC in the literature to recognise their common clinical characteristics, risk factors and outcomes. RESULTS: In our series of seven patients, three typical, two inverted, one biventricular and one global TC were recognised. Three were females and four were males. The mean age was 71±11 years. In-hospital death was observed in 57% of patients. Patients who belonged to the high-risk group and had high-risk echocardiographic features in our series had a 100% mortality rate. CONCLUSIONS: COVID-19 complicated by TC has a high mortality rate. Early identification of patients with COVID-19 who are at higher risk for developing secondary TC is important for the prevention of complications, and thus improved outcomes.


Assuntos
Causas de Morte , Infecções por Coronavirus/epidemiologia , Surtos de Doenças/estatística & dados numéricos , Pneumonia Viral/epidemiologia , Cardiomiopatia de Takotsubo/diagnóstico por imagem , Cardiomiopatia de Takotsubo/epidemiologia , Distribuição por Idade , Idoso , Estudos de Coortes , Comorbidade , Infecções por Coronavirus/diagnóstico , Ecocardiografia/métodos , Eletrocardiografia/métodos , Feminino , Coração Auxiliar , Mortalidade Hospitalar/tendências , Humanos , Masculino , Massachusetts , Pessoa de Meia-Idade , Pandemias , Pneumonia Viral/diagnóstico , Prevalência , Estudos Retrospectivos , Medição de Risco , Índice de Gravidade de Doença , Distribuição por Sexo , Cardiomiopatia de Takotsubo/terapia
3.
Mayo Clin Proc ; 95(10): 2099-2109, 2020 10.
Artigo em Inglês | MEDLINE | ID: mdl-33012341

RESUMO

OBJECTIVE: To study whether combining vital signs and electrocardiogram (ECG) analysis can improve early prognostication. METHODS: This study analyzed 1258 adults with coronavirus disease 2019 who were seen at three hospitals in New York in March and April 2020. Electrocardiograms at presentation to the emergency department were systematically read by electrophysiologists. The primary outcome was a composite of mechanical ventilation or death 48 hours from diagnosis. The prognostic value of ECG abnormalities was assessed in a model adjusted for demographics, comorbidities, and vital signs. RESULTS: At 48 hours, 73 of 1258 patients (5.8%) had died and 174 of 1258 (13.8%) were alive but receiving mechanical ventilation with 277 of 1258 (22.0%) patients dying by 30 days. Early development of respiratory failure was common, with 53% of all intubations occurring within 48 hours of presentation. In a multivariable logistic regression, atrial fibrillation/flutter (odds ratio [OR], 2.5; 95% CI, 1.1 to 6.2), right ventricular strain (OR, 2.7; 95% CI, 1.3 to 6.1), and ST segment abnormalities (OR, 2.4; 95% CI, 1.5 to 3.8) were associated with death or mechanical ventilation at 48 hours. In 108 patients without these ECG abnormalities and with normal respiratory vitals (rate <20 breaths/min and saturation >95%), only 5 (4.6%) died or required mechanical ventilation by 48 hours versus 68 of 216 patients (31.5%) having both ECG and respiratory vital sign abnormalities. CONCLUSION: The combination of abnormal respiratory vital signs and ECG findings of atrial fibrillation/flutter, right ventricular strain, or ST segment abnormalities accurately prognosticates early deterioration in patients with coronavirus disease 2019 and may assist with patient triage.


Assuntos
Arritmias Cardíacas/diagnóstico por imagem , Infecções por Coronavirus/fisiopatologia , Eletrocardiografia/estatística & dados numéricos , Serviço Hospitalar de Emergência/estatística & dados numéricos , Pneumonia Viral/fisiopatologia , Tempo para o Tratamento/estatística & dados numéricos , Adulto , Betacoronavirus , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Pandemias , Prognóstico
4.
Annu Int Conf IEEE Eng Med Biol Soc ; 2020: 1548-1551, 2020 07.
Artigo em Inglês | MEDLINE | ID: mdl-33018287

RESUMO

This paper proposes an automatic method for classifying Aortic valvular stenosis (AS) using ECG (Electrocardiogram) images by the deep learning whose training ECG images are annotated by the diagnoses given by the medical doctor who observes the echocardiograms. Besides, it explores the relationship between the trained deep learning network and its determinations, using the Grad-CAM.In this study, one-beat ECG images for 12-leads and 4-leads are generated from ECG's and train CNN's (Convolutional neural network). By applying the Grad-CAM to the trained CNN's, feature areas are detected in the early time range of the one-beat ECG image. Also, by limiting the time range of the ECG image to that of the feature area, the CNN for the 4-lead achieves the best classification performance, which is close to expert medical doctors' diagnoses.Clinical Relevance-This paper achieves as high AS classification performance as medical doctors' diagnoses based on echocardiograms by proposing an automatic method for detecting AS only using ECG.


Assuntos
Estenose da Valva Aórtica , Aprendizado Profundo , Eletrocardiografia , Estenose da Valva Aórtica/diagnóstico , Ecocardiografia , Humanos , Redes Neurais de Computação
5.
BMC Infect Dis ; 20(1): 730, 2020 Oct 07.
Artigo em Inglês | MEDLINE | ID: mdl-33028242

RESUMO

BACKGROUND: The incidence of Lyme disease (LD) in North America has increased substantially in the past two decades. Concomitant with the increased incidence of infection has been an enhancement in the recognition of LD complications. Here, we report a case of Lyme carditis complicated by heart block in a pediatric patient admitted to our children's hospital. What is unique about this case is that the complaint of chest palpitations is an infrequent presentation of LD, and what it adds to the scientific literature is an improved understanding of LD in the pediatric population. CASE PRESENTATION: The patient was a 16-year-old male who presented with the main concerns of acute onset of palpitations and chest pain. An important clinical finding was Erythema migrans (EM) on physical exam. The primary diagnoses were LD with associated Lyme carditis, based on the finding of 1st degree atrioventricular heart block (AVB) and positive IgM and IgG antibodies to Borrelia burgdorferi. Interventions included echocardiography, electrocardiography (EKG), and intravenous antibiotics. The hospital course was further remarkable for transition to 2nd degree heart block and transient episodes of complete heart block. A normal sinus rhythm and PR interval were restored after antibiotic therapy and the primary outcome was that of an uneventful recovery. CONCLUSIONS: Lyme carditis occurs in < 5% of LD cases, but the "take-away" lesson of this case is that carditis can be the presenting manifestation of B. burgdorferi infection in pediatric patients. Any patient with suspected Lyme carditis manifesting cardiac symptoms such as syncope, chest pain, or EKG changes should be admitted for parenteral antibiotic therapy and cardiac monitoring. The most common manifestation of Lyme carditis is AVB. AVB may manifest as first-degree block, or may present as high-grade second or third-degree block. Other manifestations of Lyme carditis may include myopericarditis, left ventricular dysfunction, and cardiomegaly. Resolution of carditis is typically achieved through antibiotic administration, although pacemaker placement should be considered if the PR interval fails to normalize or if higher degrees of heart block, with accompanying symptoms, are encountered. With the rising incidence of LD, providers must maintain a high level of suspicion in order to promptly diagnose and treat Lyme carditis.


Assuntos
Borrelia burgdorferi/isolamento & purificação , Doença de Lyme/diagnóstico , Administração Intravenosa , Adolescente , Antibacterianos/uso terapêutico , Borrelia burgdorferi/imunologia , Eletrocardiografia , Bloqueio Cardíaco/diagnóstico , Bloqueio Cardíaco/etiologia , Humanos , Imunoglobulina M/sangue , Doença de Lyme/complicações , Doença de Lyme/tratamento farmacológico , Doença de Lyme/microbiologia , Masculino , Miocardite/diagnóstico
6.
Medicine (Baltimore) ; 99(41): e22491, 2020 Oct 09.
Artigo em Inglês | MEDLINE | ID: mdl-33031283

RESUMO

RATIONALE: Coronary chest pain is usually ischemic in etiology and has various electrocardiographic presentations. Lately, it has been recognized that myocardial bridging (MB) with severe externally mechanical compression of an epicardial coronary artery during systole may result in myocardial ischemia. Such a phenomenon can be associated with chronic angina pectoris, acute coronary syndromes (ACS), coronary spasm, ventricular septal rupture, arrhythmias, exercise-induced atrioventricular conduction blocks, transient ventricular dysfunction, and sudden death. PATIENT CONCERNS: We report the case of a 58-year-old woman presenting with recurrent episodes of constrictive chest pain during exercise within the last 2 weeks. Except for obesity, general and cardiovascular clinical examination on admission were normal. DIAGNOSES: The resting 12 lead electrocardiogram (ECG) revealed changes typically for Wellens syndrome. High-sensitive cardiac troponin I was normal. We established the diagnosis of low-risk non-ST-segment elevation acute coronary syndrome with a Global Registry of Acute Coronary Events risk score of 92 points. INTERVENTIONS: The patient underwent coronary angiography, who showed subocclusive dynamic obstruction of the left anterior descending artery due to MB. OUTCOMES: The patient was managed conservatively. Her hospital course was uneventful and she was discharged on pharmacological therapy (clopidogrel, bisoprolol, amlodipine, atorvastatin, and metformin) with well-controlled symptoms on followup. LESSONS: MB is an unusual cause of myocardial ischemia. Wellens syndrome is an unusual presentation of ACS. We present herein a rare case of Wellens syndrome caused by MB. This case highlights the importance of subtle and frequently overseen ECG findings when assessing patients with chest pain and second, the importance of considering nonatherosclerotic causes for ACS.


Assuntos
Síndrome Coronariana Aguda/diagnóstico , Oclusão Coronária/diagnóstico por imagem , Ponte Miocárdica/diagnóstico por imagem , Dor no Peito/etiologia , Angiografia Coronária , Eletrocardiografia , Feminino , Humanos , Pessoa de Meia-Idade , Ponte Miocárdica/fisiopatologia , Síndrome
7.
Praxis (Bern 1994) ; 109(13): 1035-1038, 2020.
Artigo em Alemão | MEDLINE | ID: mdl-33050812

RESUMO

CME ECG 66/Answers: Torsade de Pointes: The Danger of a Rotating Heart Axis Abstract. Torsade de pointes tachycardia is a potentially life-threatening heart rhythm disorder, caused by prolongation of the QT interval resulting in triggered activity. This QT prolongation can be congenital or acquired. If acquired, it is mainly caused by pharmacological therapy. The hallmark of torsade de pointes is an undulating QRS axis with a twist of the QRS complex around the ECG's baseline. Often, this polymorphic ventricular tachycardia is self-limiting, but degeneration into ventricular fibrillation is possible, which makes torsade de pointes tachycardia dangerous. This article aims to provide insights into etiology, diagnostics, prevention and management of this heart rhythm disorder.


Assuntos
Síndrome do QT Longo , Taquicardia Ventricular , Torsades de Pointes , Arritmias Cardíacas , Eletrocardiografia , Humanos , Síndrome do QT Longo/diagnóstico , Taquicardia Ventricular/diagnóstico , Torsades de Pointes/diagnóstico
8.
Praxis (Bern 1994) ; 109(13): 1017-1025, 2020.
Artigo em Alemão | MEDLINE | ID: mdl-33050815

RESUMO

CME: Left Bundle Branch Block and Painful Left Bundle Branch Block Syndrome Abstract. Left bundle branch block (LBBB) is the electrocardiographic correlate of a pathologic transmission of the electrical signals in the myocardium which can lead to a dyssynchronous left ventricular activation and thus to an inefficient contraction of the ventricles. It is usually the expression of an underlying cardiopathy and represents an independent risk factor of cardiovascular mortality, therefore further examination is indicated in each case. Besides the treatment of an underlying disease, a specific therapy has been available since the introduction of cardiac resynchronization therapy (CRT). A rarer phenomenon is the painful left bundle branch block in structurally healthy hearts.


Assuntos
Bloqueio de Ramo , Terapia de Ressincronização Cardíaca , Eletrocardiografia , Ventrículos do Coração , Humanos , Síndrome
9.
Ther Umsch ; 77(8): 385-389, 2020.
Artigo em Alemão | MEDLINE | ID: mdl-33054647

RESUMO

Arrhythmia as an Incidental Finding Abstract. An arrhythmic pulse can be determined using clinical or technical examinations such as palpitation and ECG. Due to the rapid spread of wearables, more and more people have the opportunity to derive pulse curves or even ECGs themselves before seeking professional medical care, which increases the number of randomly detected arrhythmic pulse. Incidental findings naturally lead to relevant diagnoses in some of the people affected, but on the other hand they are also psychologically stressful for some individuals. Therefore, it is important to differentiate frequently found common benign causes from the causes with therapeutic consequence and to adequately inform patients about their rhythm. In particular bradyarrhythmias often have no therapeutic consequence as long as the patient remains asymptomatic. Pacemakers are usually only indicated for symptomatic bradycardia. Atrial fibrillation deserves special attention due to its frequency and the fact that, if undetected, this is associated with significantly increased morbidity and mortality. Supraventricular and ventricular extrasystoles increase with age. They are often "idiopathic", but they also can be an expression of still subclinical heart disease.


Assuntos
Fibrilação Atrial , Achados Incidentais , Fibrilação Atrial/diagnóstico , Fibrilação Atrial/terapia , Bradicardia/diagnóstico , Bradicardia/terapia , Eletrocardiografia , Humanos
10.
Acta Med Indones ; 52(3): 290-296, 2020 Jul.
Artigo em Inglês | MEDLINE | ID: mdl-33020340

RESUMO

Since the first case was reported at the end of 2019, COVID-19 has spread throughout the world and has become a pandemic. The high transmission rate of the virus has made it a threat to public health globally. Viral infections may trigger acute coronary syndromes, arrhythmias, and exacerbation of heart failure, due to a combination of effects including significant systemic inflammatory responses and localized vascular inflammation at the arterial plaque level. Indonesian clinical practice guideline stated that (hydroxy)chloroquine alone or in combination with azithromycin may be used to treat for COVID-19. However, chloroquine, hydroxychloroquine, and azithromycin all prolong the QT interval, raising concerns about the risk of arrhythmic death from individual or concurrent use of these medications. To date, there is still no vaccine or specific antiviral treatment for COVID-19. Therefore, prevention of infection in people with cardiovascular risk and mitigation of the adverse effects of treatment is necessary.


Assuntos
Antiarrítmicos/uso terapêutico , Betacoronavirus , Infecções por Coronavirus/complicações , Morte Súbita Cardíaca/prevenção & controle , Pneumonia Viral/complicações , Taquicardia Ventricular/prevenção & controle , Infecções por Coronavirus/epidemiologia , Morte Súbita Cardíaca/etiologia , Eletrocardiografia , Humanos , Pandemias , Pneumonia Viral/epidemiologia , Prognóstico , Taquicardia Ventricular/etiologia
12.
Acta Med Indones ; 52(3): 290-296, 2020 Jul.
Artigo em Inglês | MEDLINE | ID: covidwho-833791

RESUMO

Since the first case was reported at the end of 2019, COVID-19 has spread throughout the world and has become a pandemic. The high transmission rate of the virus has made it a threat to public health globally. Viral infections may trigger acute coronary syndromes, arrhythmias, and exacerbation of heart failure, due to a combination of effects including significant systemic inflammatory responses and localized vascular inflammation at the arterial plaque level. Indonesian clinical practice guideline stated that (hydroxy)chloroquine alone or in combination with azithromycin may be used to treat for COVID-19. However, chloroquine, hydroxychloroquine, and azithromycin all prolong the QT interval, raising concerns about the risk of arrhythmic death from individual or concurrent use of these medications. To date, there is still no vaccine or specific antiviral treatment for COVID-19. Therefore, prevention of infection in people with cardiovascular risk and mitigation of the adverse effects of treatment is necessary.


Assuntos
Antiarrítmicos/uso terapêutico , Betacoronavirus , Infecções por Coronavirus/complicações , Morte Súbita Cardíaca/prevenção & controle , Pneumonia Viral/complicações , Taquicardia Ventricular/prevenção & controle , Infecções por Coronavirus/epidemiologia , Morte Súbita Cardíaca/etiologia , Eletrocardiografia , Humanos , Pandemias , Pneumonia Viral/epidemiologia , Prognóstico , Taquicardia Ventricular/etiologia
13.
Annu Int Conf IEEE Eng Med Biol Soc ; 2020: 608-611, 2020 07.
Artigo em Inglês | MEDLINE | ID: mdl-33017915

RESUMO

Fetal electrocardiography is a valuable alternative to standard fetal monitoring. Suppression of the maternal electrocardiogram (ECG) in the abdominal measurements, results in fetal ECG signals, from which the fetal heart rate (HR) can be determined. This HR detection typically requires fetal R-peak detection, which is challenging, especially during low signal-to-noise ratio periods, caused for example by uterine activity. In this paper, we propose the combination of a convolutional neural network and a long short-term memory network that directly predicts the fetal HR from multichannel fetal ECG. The network is trained on a dataset, recorded during labor, while the performance of the method is evaluated both on a test dataset and on set-A of the 2013 Physionet /Computing in Cardiology Challenge. The algorithm achieved a positive percent agreement of 92.1% and 98.1% for the two datasets respectively, outperforming a top-performing state-of-the-art signal processing algorithm.


Assuntos
Frequência Cardíaca Fetal , Memória de Curto Prazo , Eletrocardiografia , Feminino , Monitorização Fetal , Humanos , Gravidez , Processamento de Sinais Assistido por Computador
14.
Artigo em Inglês | MEDLINE | ID: mdl-33017920

RESUMO

Cardiography enables diagnostic and preventive care in hospitals and outpatient scenarios. However, most heart monitors do not distinguish the phases of the cardiac cycle. The transition between phases is indicated by the primary heart sounds. OBJECTIVE: Automatically identify the vibrations corresponding to both heart sounds. METHODS: Cardiac activity was monitored for 15 subjects while at rest, during exertion, and while performing static breath holds. The subjects consisted of 6 males and 9 females between the ages of 18-39 years with no known cardiorespiratory ailments. Motion corresponding to the heart sounds was identified using vibrational cardiography (VCG). The waveforms were processed to obtain quantities associated with their linear jerk and rotational kinetic energy. RESULTS: The ability to identity the first vibration was evaluated using the heart rate as a figure of merit. Its correlation with electrocardiography (ECG) measurements produced a r2 coefficient of 0.9887. The second vibration was compared with impedance cardiography (ICG) based on its delay from the ECG R-peak, and the fraction of the beat duration occupied by left ventricular ejection time. The comparisons produced r2 values of 0.251 and 0.2797, respectively. CONCLUSION: The vibrations corresponding to both primary heart sounds have the potential to be analyzed using VCG. SIGNIFICANCE: This study provides evidence of the feasibility of using VCG in identifying mechanical cardiovascular function. It facilitates non-invasive cardiac health monitoring in daily life.


Assuntos
Ruídos Cardíacos , Adolescente , Adulto , Cardiografia de Impedância , Eletrocardiografia , Feminino , Frequência Cardíaca , Humanos , Masculino , Vibração , Adulto Jovem
15.
Annu Int Conf IEEE Eng Med Biol Soc ; 2020: 288-291, 2020 07.
Artigo em Inglês | MEDLINE | ID: mdl-33017985

RESUMO

Machine learning has become increasingly useful in various medical applications. One such case is the automatic categorization of ECG voltage data. A method of categorization is proposed that works in real time to provide fast and accurate classifications of heart beats. This proposed method uses machine learning principles to allow for results to be determined based on a training dataset. The goal of this project is to develop a method of automatically classifying heartbeats that can be done on a low level and run on portable hardware.


Assuntos
Eletrocardiografia , Processamento de Sinais Assistido por Computador , Arritmias Cardíacas/diagnóstico , Doença do Sistema de Condução Cardíaco , Humanos , Redes Neurais de Computação
16.
Annu Int Conf IEEE Eng Med Biol Soc ; 2020: 292-295, 2020 07.
Artigo em Inglês | MEDLINE | ID: mdl-33017986

RESUMO

Arrhythmia is a serious cardiovascular disease, and early diagnosis of arrhythmia is critical. In this study, we present a waveform-based signal processing (WBSP) method to produce state-of-the-art performance in arrhythmia classification. When performing WBSP, we first filtered ECG signals, searched local minima, and removed baseline wandering. Subsequently, we fit the processed ECG signals with Gaussians and extracted the parameters. Afterwards, we exploited the products of WBSP to accomplish arrhythmia classification with our proposed machine learning-based and deep learning-based classifiers. We utilized MIT-BIH Arrhythmia Database to validate WBSP. Our best classifier achieved 98.8% accuracy. Moreover, it reached 96.3% sensitivity in class V and 98.6% sensitivity in class Q, which both share one of the best among the related works. In addition, our machine learning-based classifier accomplished identifying four waveform components essential for automated arrhythmia classification: the similarity of QRS complex to a Gaussian curve, the sharpness of the QRS complex, the duration of and the area enclosed by P-wave.Clinical relevance- Early diagnosis and automated classification of arrhythmia is clinically essential.


Assuntos
Aprendizado Profundo , Eletrocardiografia , Arritmias Cardíacas/diagnóstico , Humanos , Aprendizado de Máquina , Processamento de Sinais Assistido por Computador
17.
Annu Int Conf IEEE Eng Med Biol Soc ; 2020: 296-299, 2020 07.
Artigo em Inglês | MEDLINE | ID: mdl-33017987

RESUMO

Recent developments in the field of deep learning has shown a rise in its use for clinical applications such as electrocardiogram (ECG) analysis and cardiac arrhythmia classification. Such systems are essential in the early detection and management of cardiovascular diseases. However, due to privacy concerns and also the lack of resources, there is a gap in the data available to run such powerful and data-intensive models. To address the lack of annotated, high-quality ECG data for heart disease research, ECG data generation from a small set of ECG to obtain huge annotated data is seen as an effective solution. Generative Feature Matching Network (GFMN) was shown to resolve few drawbacks of commonly used generative adversarial networks (GAN). Based on this, we developed a deep learning model to generate ECGs that resembles real ECG by feature matching with the existing data.Clinical relevance- This work addresses the lack of a large quantity of good quality, publicly available annotated ECG data required to build deep learning models for cardiac signal processing research. We can use the model presented in this paper to generate ECG signals of a target rhythm pattern and also subject-specific ECG morphology that could improve their cardiac health monitoring while maintaining privacy.


Assuntos
Arritmias Cardíacas , Cardiopatias , Arritmias Cardíacas/diagnóstico , Doença do Sistema de Condução Cardíaco , Eletrocardiografia , Humanos , Processamento de Sinais Assistido por Computador
18.
Annu Int Conf IEEE Eng Med Biol Soc ; 2020: 300-303, 2020 07.
Artigo em Inglês | MEDLINE | ID: mdl-33017988

RESUMO

Cardiac arrhythmia is a prevalent and significant cause of morbidity and mortality among cardiac ailments. Early diagnosis is crucial in providing intervention for patients suffering from cardiac arrhythmia. Traditionally, diagnosis is performed by examination of the Electrocardiogram (ECG) by a cardiologist. This method of diagnosis is hampered by the lack of accessibility to expert cardiologists. For quite some time, signal processing methods had been used to automate arrhythmia diagnosis. However, these traditional methods require expert knowledge and are unable to model a wide range of arrhythmia. Recently, Deep Learning methods have provided solutions to performing arrhythmia diagnosis at scale. However, the black-box nature of these models prohibit clinical interpretation of cardiac arrhythmia. There is a dire need to correlate the obtained model outputs to the corresponding segments of the ECG. To this end, two methods are proposed to provide interpretability to the models. The first method is a novel application of Gradient-weighted Class Activation Map (Grad-CAM) for visualizing the saliency of the CNN model. In the second approach, saliency is derived by learning the input deletion mask for the LSTM model. The visualizations are provided on a model whose competence is established by comparisons against baselines. The results of model saliency not only provide insight into the prediction capability of the model but also aligns with the medical literature for the classification of cardiac arrhythmia.Clinical relevance- Adapts interpretability modules for deep learning networks in ECG arrhythmia classfication, allowing for better clinical interpretation.


Assuntos
Algoritmos , Arritmias Cardíacas , Arritmias Cardíacas/diagnóstico , Eletrocardiografia , Humanos , Redes Neurais de Computação , Processamento de Sinais Assistido por Computador
19.
Annu Int Conf IEEE Eng Med Biol Soc ; 2020: 304-307, 2020 07.
Artigo em Inglês | MEDLINE | ID: mdl-33017989

RESUMO

Electrocardiograph (ECG) is one of the most critical physiological signals for arrhythmia diagnosis in clinical practice. In recent years, various algorithms based on deep learning have been proposed to solve the heartbeat classification problem and achieved saturated accuracy in intrapatient paradigm, but encountered performance degradation in inter-patient paradigm due to the drastic variation of ECG signals among different individuals. In this paper, we propose a novel unsupervised domain adaptation scheme to address this problem. Specifically, we first propose a robust baseline model called Multi-path Atrous Convolutional Network (MACN) to tackle ECG heartbeat classification. Further, we introduce Cluster-aligning loss and Cluster-separating loss to align the distributions of training and test data and increase the discriminability, respectively. The proposed method requires no expert annotations but a short period of unlabelled data in new records. Experimental results on the MIT-BIH database demonstrate that our scheme effectively intensifies the baseline model and achieves competitive performance with other state-of-the-arts.


Assuntos
Eletrocardiografia , Processamento de Sinais Assistido por Computador , Algoritmos , Arritmias Cardíacas/diagnóstico , Frequência Cardíaca , Humanos
20.
Annu Int Conf IEEE Eng Med Biol Soc ; 2020: 308-311, 2020 07.
Artigo em Inglês | MEDLINE | ID: mdl-33017990

RESUMO

Lacking sufficient training samples of different heart rhythms is a common bottleneck to obtain arrhythmias classification models with high accuracy using artificial neural networks. To solve this problem, we propose a novel data augmentation method based on short-time Fourier transform (STFT) and generative adversarial network (GAN) to obtain evenly distributed samples in the training dataset. Firstly, the one-dimensional electrocardiogram (ECG) signals with a fixed length of 6 s are subjected to STFT to obtain the coefficient matrices, and then the matrices of different heart rhythm samples are used to train GAN models respectively. The generated matrices are later employed to augment the training dataset of classification models based on four convolutional neural networks (CNNs). The result shows that the performances of the classification networks are all improved after we adopt the data enhancement strategy. The proposed method is helpful in augmentation and classification of biomedical signals, especially in detecting multiple arrhythmias, since adequate training samples are usually inaccessible in these studies.


Assuntos
Arritmias Cardíacas , Redes Neurais de Computação , Arritmias Cardíacas/diagnóstico , Eletrocardiografia , Análise de Fourier , Humanos
SELEÇÃO DE REFERÊNCIAS
DETALHE DA PESQUISA