scholarly journals Detecting Cardiomyopathies in Pregnancy and the Postpartum Period with an Electrocardiogram-Based Deep Learning Model

2021 ◽  
Author(s):  
Demilade A Adedinsewo ◽  
Patrick W Johnson ◽  
Erika J Douglass ◽  
Itzhak Zachi Attia ◽  
Sabrina D Phillips ◽  
...  
Keyword(s):  
Deep Learning ◽  
Natriuretic Peptides ◽  
Postpartum Period ◽  
Learning Model ◽  
Left Ventricular ◽  
Left Ventricular Ejection ◽  
Clinical Parameters ◽  
Multivariable Model ◽  
Ventricular Ejection Fraction ◽  
Deep Learning Model

Abstract Aims Cardiovascular disease is a major threat to maternal health, with cardiomyopathy being among the most common acquired cardiovascular diseases during pregnancy and the postpartum period. The aim of our study was to evaluate the effectiveness of an electrocardiogram (ECG)-based deep learning model in identifying cardiomyopathy during pregnancy and the postpartum period. Methods and Results We used an ECG-based deep learning model to detect cardiomyopathy in a cohort of women who were pregnant or in the postpartum period seen at Mayo Clinic. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. We compared the diagnostic probabilities of the deep learning model with natriuretic peptides and a multivariable model consisting of demographic and clinical parameters. The study cohort included 1,807 women; 7%, 10%, and 13% had left ventricular ejection fraction (LVEF) of 35% or less, less than 45%, and less than 50%, respectively. The ECG-based deep learning model identified cardiomyopathy with AUCs of 0.92 (LVEF ≤35%), 0.89 (LVEF <45%), and 0.87 (LVEF <50%). For LVEF of 35% or less, AUC was higher in Black (0.95) and Hispanic (0.98) women compared to White (0.91). Natriuretic peptides and the multivariable model had AUCs of 0.85 to 0.86 and 0.72, respectively. Conclusions An ECG-based deep learning model effectively identifies cardiomyopathy during pregnancy and the postpartum period and outperforms natriuretic peptides and traditional clinical parameters with the potential to become a powerful initial screening tool for cardiomyopathy in the obstetric care setting.

scholarly journals Detecting cardiomyopathies in pregnancy and the postpartum period using ECG

2021 ◽  
Vol 42 (Supplement_1) ◽  
Author(s):  
D Adedinsewo ◽  
P W Johnson ◽  
E J Douglass ◽  
Z I Attia ◽  
S D Phillips ◽  
...  
Keyword(s):  
Deep Learning ◽  
Natriuretic Peptides ◽  
Mayo Clinic ◽  
Learning Model ◽  
National Institutes Of Health ◽  
Postpartum Women ◽  
Clinical Parameters ◽  
Multivariable Model ◽  
Deep Learning Model ◽  
In Pregnancy

Abstract Background Cardiovascular disease (CVD) has been identified as a major threat to maternal health in the US and UK with cardiomyopathy being one of the most common acquired CVD in the pregnant and postpartum period. Diagnosing cardiomyopathy in pregnancy is challenging due to an overlap of cardiovascular symptoms with normal pregnancy symptoms. Purpose The purpose of this study was to evaluate the effectiveness of an ECG based deep learning model in identifying cardiomyopathy among pregnant and postpartum women. Methods We utilized an ECG based deep learning model to detect cardiomyopathy in a cohort of pregnant or postpartum women seen at multiple hospital sites. Model performance was evaluated using area under the curve (AUC), accuracy, sensitivity, and specificity. We compared the diagnostic probabilities of the deep learning model with natriuretic peptides and a multivariable model consisting of demographic and clinical parameters. Results 1,807 women were included. 7%, 10% and 13% had LVEF ≤35%, <45% and <50% respectively. The ECG based deep learning model identified cardiomyopathy with an AUC of 0.92 for left ventricular ejection fraction (LVEF) ≤35%, 0.89 for LVEF <45% and 0.87 for LVEF <50%. For LVEF ≤35%, AUC was higher in Black (0.95) and Hispanic (0.98) women compared to white (0.91). Natriuretic peptides and the multivariable model had AUCs of 0.85 and 0.72 respectively. Conclusions A deep learning model effectively identifies cardiomyopathy in pregnant or postpartum women, outperforms natriuretic peptides and traditional clinical parameters with the potential to become a powerful initial screening tool for cardiomyopathy in the obstetric care setting. FUNDunding Acknowledgement Type of funding sources: Other. Main funding source(s): This study was made possible using resources supported by the Mayo Clinic Women's Health Research Center and the Mayo Clinic Building Interdisciplinary Research Careers in Women's Health (BIRCWH) Program funded by the National Institutes of Health (NIH), grant number K12 HD065987. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

scholarly journals Detecting cardiomyopathies in pregnancy and the postpartum period using ECG

2021 ◽  
Vol 2 (4) ◽  
Author(s):  
D Adedinsewo ◽  
P W Johnson ◽  
E J Douglass ◽  
Z I Attia ◽  
S D Phillips ◽  
...  
Keyword(s):  
Deep Learning ◽  
Natriuretic Peptides ◽  
Mayo Clinic ◽  
Learning Model ◽  
National Institutes Of Health ◽  
Postpartum Women ◽  
Clinical Parameters ◽  
Multivariable Model ◽  
Deep Learning Model ◽  
In Pregnancy

Abstract Background Cardiovascular disease (CVD) has been identified as a major threat to maternal health in the US and UK with cardiomyopathy being one of the most common acquired CVD in the pregnant and postpartum period. Diagnosing cardiomyopathy in pregnancy is challenging due to an overlap of cardiovascular symptoms with normal pregnancy symptoms. Purpose The purpose of this study was to evaluate the effectiveness of an ECG based deep learning model in identifying cardiomyopathy among pregnant and postpartum women. Methods We utilized an ECG based deep learning model to detect cardiomyopathy in a cohort of pregnant or postpartum women seen at multiple hospital sites. Model performance was evaluated using area under the curve (AUC), accuracy, sensitivity, and specificity. We compared the diagnostic probabilities of the deep learning model with natriuretic peptides and a multivariable model consisting of demographic and clinical parameters. Results 1,807 women were included. 7%, 10% and 13% had LVEF ≤35%, <45% and <50% respectively. The ECG based deep learning model identified cardiomyopathy with an AUC of 0.92 for left ventricular ejection fraction (LVEF) ≤35%, 0.89 for LVEF <45% and 0.87 for LVEF <50%. For LVEF ≤35%, AUC was higher in Black (0.95) and Hispanic (0.98) women compared to white (0.91). Natriuretic peptides and the multivariable model had AUCs of 0.85 and 0.72 respectively. Conclusions A deep learning model effectively identifies cardiomyopathy in pregnant or postpartum women, outperforms natriuretic peptides and traditional clinical parameters with the potential to become a powerful initial screening tool for cardiomyopathy in the obstetric care setting. Funding Acknowledgement Type of funding sources: Other. Main funding source(s): This study was made possible using resources supported by the Mayo Clinic Women's Health Research Center and the Mayo Clinic Building Interdisciplinary Research Careers in Women's Health (BIRCWH) Program funded by the National Institutes of Health (NIH), grant number K12 HD065987. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

scholarly journals 265 Incidence, determinants and prognostic relevance of HS-troponin and natriuretic peptides elevation at admission in hospitalized COVID-19 pneumonia patients

2020 ◽  
Vol 22 (Supplement_N) ◽  
pp. N65-N79
Author(s):  
Luca Arcari ◽  
Michelangelo Luciano ◽  
Luca Cacciotti ◽  
Maria Beatrice Musumeci ◽  
Valerio Spuntarelli ◽  
...  
Keyword(s):  
Hospital Mortality ◽  
Natriuretic Peptides ◽  
Multivariable Analysis ◽  
Left Ventricular ◽  
Cardiac Biomarkers ◽  
Left Ventricular Ejection ◽  
Ventricular Ejection Fraction ◽  
Good Ability ◽  
Myocardial Involvement ◽  
D Dimer

Abstract Aims myocardial involvement in the course of Coronavirus disease 2019 (COVID-19) pneumonia has been reported, though not fully characterized yet. Aim of the present study is to undertake a joint evaluation of hs-Troponin and natriuretic peptides (NP) in patients hospitalized for COVID-19 pneumonia. Methods and results in this multicenter observational study, we analyzed data from n = 111 COVID-19 patients admitted to dedicated “COVID-19” medical units. Hs-Troponin was assessed in n = 103 patients and NP in n = 82 patients on admission; subgroups were identified according to values beyond reference range. increased hs-Troponin and NP were found in 38% and 56% of the cases respectively. As compared to those with normal cardiac biomarkers, these patients were older, had higher prevalence of cardiovascular diseases (CVD) and more severe COVID-19 pneumonia by higher CRP and D-dimer and lower PaO2/FIO2. Two-dimensional echocardiography performed in a subset of patients (n = 24) showed significantly reduced left ventricular ejection fraction in patients with elevated NP only (p = 0.02), whereas right ventricular systolic function (tricuspid annular plane systolic excursion) was significantly reduced both in patients with high hs-Troponin and NP (p = 0.022 and p = 0.03 respectively). On multivariable analysis, independent associations were found of hs-Troponin with age, PaO2/FIO2 and D-dimer (B = 0.419, p = 0.001; B=-0.212, p = 0.013 and B = 0.179, p = 0.037 respectively), and of NP with age and previous CVD (B = 0.480, p < 0.001 and B = 0.253, p = 0.001 respectively). In patients with in-hospital mortality (n = 23, 21%) hs-Troponin and NP were both higher (p = 0.001 and p = 0.002 respectively), while increasing hs-troponin and NP were associated with worse in-hospital prognosis [OR 4.88 (95% CI 1.9-12.2), p = 0.001 (adjusted OR 3.1 (95% CI 1.2-8.5), p = 0.025) and OR 4.67 (95% CI 2-10.8), p < 0.001 (adjusted OR 2.89 (95% CI 1.1-7.9), p = 0.04) respectively]. Receiver operator characteristic curves showed good ability of hs-Troponin and NP in predicting in-hospital mortality (AUC = 0.869 p < 0.001 and AUC = 0.810, p < 0.001 respectively). Conclusion myocardial involvement at admission is common in COVID-19 pneumonia and associated to worse prognosis, suggesting a role for cardiac biomarkers assessment in COVID-19 risk stratification. Independent associations of hs-Troponin with markers of disease severity and of NP with underlying CVD might point towards existing different mechanisms leading to their elevation in this setting.

scholarly journals Deep Learning to Estimate Cardiac Magnetic Resonance-Derived Left Ventricular Mass

2020 ◽  
Author(s):  
Shaan Khurshid ◽  
Samuel Friedman ◽  
James P. Pirruccello ◽  
Paolo Di Achille ◽  
Nathaniel Diamant ◽  
...  
Keyword(s):  
Deep Learning ◽  
Cardiac Magnetic Resonance ◽  
Magnetic Resonance ◽  
Open Source ◽  
Pearson Correlation ◽  
Absolute Error ◽  
Learning Model ◽  
Left Ventricular ◽  
Lv Mass ◽  
Deep Learning Model

ABSTRACTBackgroundCardiac magnetic resonance (CMR) is the gold standard for left ventricular hypertrophy (LVH) diagnosis. CMR-derived LV mass can be estimated using proprietary algorithms (e.g., inlineVF), but their accuracy and availability may be limited.ObjectiveTo develop an open-source deep learning model to estimate CMR-derived LV mass.MethodsWithin participants of the UK Biobank prospective cohort undergoing CMR, we trained two convolutional neural networks to estimate LV mass. The first (ML4Hreg) performed regression informed by manually labeled LV mass (available in 5,065 individuals), while the second (ML4Hseg) performed LV segmentation informed by inlineVF contours. We compared ML4Hreg, ML4Hseg, and inlineVF against manually labeled LV mass within an independent holdout set using Pearson correlation and mean absolute error (MAE). We assessed associations between CMR-derived LVH and prevalent cardiovascular disease using logistic regression adjusted for age and sex.ResultsWe generated CMR-derived LV mass estimates within 38,574 individuals. Among 891 individuals in the holdout set, ML4Hseg reproduced manually labeled LV mass more accurately (r=0.864, 95% CI 0.847-0.880; MAE 10.41g, 95% CI 9.82-10.99) than ML4Hreg (r=0.843, 95% CI 0.823-0.861; MAE 10.51, 95% CI 9.86-11.15, p=0.01) and inlineVF (r=0.795, 95% CI 0.770-0.818; MAE 14.30, 95% CI 13.46-11.01, p<0.01). LVH defined using ML4Hseg demonstrated the strongest associations with hypertension (odds ratio 2.76, 95% CI 2.51-3.04), atrial fibrillation (1.75, 95% CI 1.37-2.20), and heart failure (4.53, 95% CI 3.16-6.33).ConclusionsML4Hseg is an open-source deep learning model providing automated quantification of CMR-derived LV mass. Deep learning models characterizing cardiac structure may facilitate broad cardiovascular discovery.

P5415Sleep disordered breathing in patients before and after heart transplantation: association with biomarkers and clinical parameters

2019 ◽  
Vol 40 (Supplement_1) ◽  
Author(s):  
E Medvedeva ◽  
L S Korostovtseva ◽  
M A Simonenko ◽  
Y V Sazonova ◽  
Y V Sviryaev
Keyword(s):  
Heart Failure ◽  
Sleep Apnea ◽  
Heart Transplantation ◽  
Left Ventricular ◽  
Left Ventricular Ejection ◽  
Clinical Parameters ◽  
Ventricular Ejection Fraction ◽  
Obstructive Sleep ◽  
Before And After ◽  
Disordered Breathing

Abstract Background Sleep-disordered breathing (SDB) is highly frequent in patients with severe heart failure (HF). SDB, and predominantly central sleep apnea (CSA), may improve after recovery of cardiac function, but available data are limited and inconclusive, especially in patients who have undergone heart transplantation. The assessment of the severity of sleep apnea is mainly based on the apnea-hypopnea index (AHI), but this event-based parameter alone may not sufficiently reflect the complex pathophysiological mechanisms underlying SDB potentially contributing to adverse outcomes in patients with heart failure. Purpose To assess SDB in patients with severe HF before and after heart transplantation, their relationship with biomarkers and clinical parameters. Methods We included 117 patients (mean age 52.4±4.7 years) with HF NYHA class II-IV in the prospective cohort study, follow-up period was 5 years. The left ventricular ejection fraction (LVEF) was 28.05±9.57%. All patients underwent a comprehensive clinical examination, echocardiography, polysomnography (PSG, Embla N7000, Natus, USA). The plasma level of NT-proBNP, was analyzed by immunoassay (ELISA). The SPSS statistical software (version 23.0) was used. Results PSG showed the following types of SDB in the studied cohort: obstructive sleep apnoea (OSA) was diagnosed in 48 patients (41%), central - in 20 (17%), mixed - in 26 (22%). Among them mild SDB was diagnosed in 29 cases, moderate in 32 and severe in 33 patients. SDB was not found in 23 patients. The following correlations were identified: NT-proBNP and obstructive apnea index (OAI) (r=−0.44, p=0.007), NT-proBNP and sleep efficiency (r=−0.71, p=0.006), AHI and body mass index (BMI) (r=0.32, p=0.01), OAI and BMI index (r=0.34, p<0.001), desaturation index and BMI (r=0.43, p<0.001), average saturation oxygen and BMI (r=−0,6, p<0,001). Twenty-three patients underwent heart transplantation. According PSG-data 1 year after transplantation we observed decrease of central apnea index (CAI) (p=0,04). On the other hand, OAI increased (p=0,01) independently of the significant change in BMI (p=0,08). Conclusion We found very high rate of SDB (80%) in patients with severe HF, the predominant type was OSA. AHI, OAI and indicators of oxygen saturation correlate with BMI and biomarkers before heart transplantation. After 1 year after transplantation CAI decreased, assessment of the dynamics of obstructive sleep apnea requires further study.

Abstract 17575: Looking Beyond Left Ventricular Ejection Fraction: Investigating the Utility of Multi-factorial Computational Modeling to Predict Sudden Cardiac Death Following Acute Coronary Syndrome in the MERLIN-TIMI36 Trial

Circulation ◽  
2015 ◽  
Vol 132 (suppl_3) ◽  
Author(s):  
Alex Van Esbroeck ◽  
Mohammed Saeed ◽  
Benjamin M Scirica ◽  
Collin M Stultz ◽  
John V Guttag ◽  
...  
Keyword(s):  
Acute Coronary Syndrome ◽  
Sudden Cardiac Death ◽  
Left Ventricular Ejection Fraction ◽  
Cardiac Death ◽  
Ventricular Dysfunction ◽  
Left Ventricular ◽  
Left Ventricular Ejection ◽  
Clinical Parameters ◽  
Ventricular Ejection Fraction ◽  
Coronary Syndrome

Introduction: Guidelines to prevent sudden cardiac death (SCD) following acute coronary syndrome (ACS) are widely based on cutoffs defined on left ventricular ejection fraction (LVEF) with limited use of other available data. Methods: We investigated the improvement in predicting post-ACS SCD using a multi-factorial model that integrates an assessment of left ventricular dysfunction through echocardiography with a broader set of other parameters collected routinely during hospitalization for ACS. Patients in the MERLIN-TIMI36 trial were admitted within 48 hours of ischemic symptoms for non-ST-elevation ACS and followed for a median of 348 days. SCD was adjudicated by a blinded clinical events committee. Data from 4,404 patients with LVEF and other clinical parameters routinely collected during index hospitalization (demographic, comorbidity, history, laboratory, electrocardiographic, and medication variables) were used to train and validate a logistic regression model to predict SCD using stepwise backward elimination and leave-one-out cross-validation. Results: The stepwise elimination process retained age, history of congestive heart failure, ST depression, beta blocker use, BNP, LVEF, and ischemia and ventricular tachycardia on continuous ECG as variables in the model. The model achieved significant improvements in discrimination, calibration and reclassification relative to LVEF, and demonstrated further utility in stratifying patients with mild/moderate left ventricular dysfunction or normal systolic function (Table 1). The model also resulted in higher sensitivity without increasing false positives relative to the LVEF<=30% rule (38% increase in correct predictions of SCD). Conclusions: Risk stratification for post-ACS SCD is significantly improved using multi-factorial models to integrate information in LVEF with other clinical parameters routinely collected during hospitalization.

Sign in / Sign up

Export Citation Format

Share Document