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.

scholarly journals Deep Learning to Predict Cardiac Magnetic Resonance–Derived Left Ventricular Mass and Hypertrophy From 12-Lead ECGs

2021 ◽  
Vol 14 (6) ◽  
Author(s):  
Shaan Khurshid ◽  
Samuel Friedman ◽  
James P. Pirruccello ◽  
Paolo Di Achille ◽  
Nathaniel Diamant ◽  
...  
Keyword(s):  
Deep Learning ◽  
Cardiac Magnetic Resonance ◽  
Magnetic Resonance ◽  
Left Ventricular Mass ◽  
Left Ventricular ◽  
Uk Biobank ◽  
C Statistic ◽  
Lv Mass ◽  
Test Sets ◽  
The Uk

Background: Classical methods for detecting left ventricular (LV) hypertrophy (LVH) using 12-lead ECGs are insensitive. Deep learning models using ECG to infer cardiac magnetic resonance (CMR)-derived LV mass may improve LVH detection. Methods: Within 32 239 individuals of the UK Biobank prospective cohort who underwent CMR and 12-lead ECG, we trained a convolutional neural network to predict CMR-derived LV mass using 12-lead ECGs (left ventricular mass-artificial intelligence [LVM-AI]). In independent test sets (UK Biobank [n=4903] and Mass General Brigham [MGB, n=1371]), we assessed correlation between LVM-AI predicted and CMR-derived LV mass and compared LVH discrimination using LVM-AI versus traditional ECG-based rules (ie, Sokolow-Lyon, Cornell, lead aVL rule, or any ECG rule). In the UK Biobank and an ambulatory MGB cohort (MGB outcomes, n=28 612), we assessed associations between LVM-AI predicted LVH and incident cardiovascular outcomes using age- and sex-adjusted Cox regression. Results: LVM-AI predicted LV mass correlated with CMR-derived LV mass in both test sets, although correlation was greater in the UK Biobank (r=0.79) versus MGB (r=0.60, P<0.001 for both). When compared with any ECG rule, LVM-AI demonstrated similar LVH discrimination in the UK Biobank (LVM-AI c-statistic 0.653 [95% CI, 0.608 -0.698] versus any ECG rule c-statistic 0.618 [95% CI, 0.574 -0.663], P=0.11) and superior discrimination in MGB (0.621; 95% CI, 0.592 -0.649 versus 0.588; 95% CI, 0.564 -0.611, P=0.02). LVM-AI-predicted LVH was associated with incident atrial fibrillation, myocardial infarction, heart failure, and ventricular arrhythmias. Conclusions: Deep learning-inferred LV mass estimates from 12-lead ECGs correlate with CMR-derived LV mass, associate with incident cardiovascular disease, and may improve LVH discrimination compared to traditional ECG rules.

scholarly journals Non-invasive cuff-less blood pressure estimation using a hybrid deep learning model

2021 ◽  
Vol 53 (2) ◽  
Author(s):  
Sen Yang ◽  
Yaping Zhang ◽  
Siu-Yeung Cho ◽  
Ricardo Correia ◽  
Stephen P. Morgan
Keyword(s):  
Machine Learning ◽  
Blood Pressure ◽  
Deep Learning ◽  
Absolute Error ◽  
Learning Model ◽  
Physiological Measurement ◽  
British Hypertension Society ◽  
Non Invasive ◽  
Blood Pressure Estimation ◽  
Deep Learning Model

AbstractConventional blood pressure (BP) measurement methods have different drawbacks such as being invasive, cuff-based or requiring manual operations. There is significant interest in the development of non-invasive, cuff-less and continual BP measurement based on physiological measurement. However, in these methods, extracting features from signals is challenging in the presence of noise or signal distortion. When using machine learning, errors in feature extraction result in errors in BP estimation, therefore, this study explores the use of raw signals as a direct input to a deep learning model. To enable comparison with the traditional machine learning models which use features from the photoplethysmogram and electrocardiogram, a hybrid deep learning model that utilises both raw signals and physical characteristics (age, height, weight and gender) is developed. This hybrid model performs best in terms of both diastolic BP (DBP) and systolic BP (SBP) with the mean absolute error being 3.23 ± 4.75 mmHg and 4.43 ± 6.09 mmHg respectively. DBP and SBP meet the Grade A and Grade B performance requirements of the British Hypertension Society respectively.

scholarly journals Prognostic significance of cardiac magnetic resonance-based markers in patients with hypertrophic cardiomyopathy

2021 ◽  
Author(s):  
Zsofia Dohy ◽  
Liliana Szabo ◽  
Attila Toth ◽  
Csilla Czimbalmos ◽  
Rebeka Horvath ◽  
...  
Keyword(s):  
Cardiac Magnetic Resonance ◽  
Hypertrophic Cardiomyopathy ◽  
Magnetic Resonance ◽  
Myocardial Fibrosis ◽  
Ventricular Arrhythmias ◽  
Prognostic Significance ◽  
Strain Analysis ◽  
Left Ventricular ◽  
Icd Therapy ◽  
Lv Mass

AbstractThe prognosis of patients with hypertrophic cardiomyopathy (HCM) varies greatly. Cardiac magnetic resonance (CMR) is the gold standard method for assessing left ventricular (LV) mass and volumes. Myocardial fibrosis can be noninvasively detected using CMR. Moreover, feature-tracking (FT) strain analysis provides information about LV deformation. We aimed to investigate the prognostic significance of standard CMR parameters, myocardial fibrosis, and LV strain parameters in HCM patients. We investigated 187 HCM patients who underwent CMR with late gadolinium enhancement and were followed up. LV mass (LVM) was evaluated with the exclusion and inclusion of the trabeculae and papillary muscles (TPM). Global LV strain parameters and mechanical dispersion (MD) were calculated. Myocardial fibrosis was quantified. The combined endpoint of our study was all-cause mortality, heart transplantation, malignant ventricular arrhythmias and appropriate implantable cardioverter defibrillator (ICD) therapy. The arrhythmia endpoint was malignant ventricular arrhythmias and appropriate ICD therapy. The LVM index (LVMi) was an independent CMR predictor of the combined endpoint independent of the quantification method (p < 0.01). The univariate predictors of the combined endpoint were LVMi, global longitudinal (GLS) and radial strain and longitudinal MD (MDL). The univariate predictors of arrhythmia events included LVMi and myocardial fibrosis. More pronounced LV hypertrophy was associated with impaired GLS and increased MDL. More extensive myocardial fibrosis correlated with impaired GLS (p < 0.001). LVMi was an independent CMR predictor of major events, and myocardial fibrosis predicted arrhythmia events in HCM patients. FT strain analysis provided additional information for risk stratification in HCM patients.

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 &lt;45%), and 0.87 (LVEF &lt;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.

Relation of B-type natriuretic peptide to left ventricular wall stress as assessed by cardiac magnetic resonance imaging in patients with dilated cardiomyopathy

2007 ◽  
Vol 85 (8) ◽  
pp. 790-799 ◽  
Author(s):  
P. Alter ◽  
H. Rupp ◽  
M.B. Rominger ◽  
A. Vollrath ◽  
F. Czerny ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Cardiac Magnetic Resonance ◽  
Magnetic Resonance ◽  
Wall Stress ◽  
Left Ventricular ◽  
Ventricular Wall ◽  
Resonance Imaging ◽  
Sphere Model ◽  
Lv Mass ◽  
Systolic Wall Stress

Ventricular loading conditions are crucial determinants of cardiac function and prognosis in heart failure. B-type natriuretic peptide (BNP) is mainly stored in the ventricular myocardium and is released in response to an increased ventricular filling pressure. We examined, therefore, the hypothesis that BNP serum concentrations are related to ventricular wall stress. Cardiac magnetic resonance imaging (MRI) was used to assess left ventricular (LV) mass and cardiac function of 29 patients with dilated cardiomyopathy and 5 controls. Left ventricular wall stress was calculated by using a thick-walled sphere model, and BNP was assessed by immunoassay. LV mass (r = 0.73, p < 0.001) and both LV end-diastolic (r = 0.54, p = 0.001) and end-systolic wall stress (r = 0.66, p < 0.001) were positively correlated with end-diastolic volume. LV end-systolic wall stress was negatively related to LV ejection fraction (EF), whereas end-diastolic wall stress was not related to LVEF. BNP concentration correlated positively with LV end-diastolic wall stress (r = 0.50, p = 0.002). Analysis of variance revealed LV end-diastolic wall stress as the only independent hemodynamic parameter influencing BNP (p < 0.001). The present approach using a thick-walled sphere model permits determination of mechanical wall stress in a clinical routine setting using standard cardiac MRI protocols. A correlation of BNP concentration with calculated LV stress was observed in vivo. Measurement of BNP seems to be sufficient to assess cardiac loading conditions. Other relations of BNP with various hemodynamic parameters (e.g., EF) appear to be secondary. Since an increased wall stress is associated with cardiac dilatation, early diagnosis and treatment could potentially prevent worsening of the outcome.

P5270Impaired myocardial deformation assessed by cardiac magnetic resonance is associated with increased arrhythmic risk in hypertrophic cardiomyopathy

2019 ◽  
Vol 40 (Supplement_1) ◽  
Author(s):  
A R Barbosa ◽  
C M O'neill ◽  
C Ruivo ◽  
I Cruz ◽  
O Sousa ◽  
...  
Keyword(s):  
Cardiac Magnetic Resonance ◽  
Hypertrophic Cardiomyopathy ◽  
Magnetic Resonance ◽  
Sudden Cardiac Arrest ◽  
Heart Association ◽  
Left Ventricular ◽  
Myocardial Deformation ◽  
Clinical Severity ◽  
Lv Mass ◽  
Estimated Risk

Abstract Background Strain techniques, such as feature tracking cardiac magnetic resonance (FT-CMR), have emerged as a promise for more accurate evaluation of cardiac function compared to ejection fraction. In hypertrophic cardiomyopathy (HCM) patients, impaired myocardial deformation measured by FT-CMR has been associated with severity of hypertrophy and presence of late gadolinium enhancement (LGE) but associations with clinical severity and prognosis are scarce. Purpose To analyse the association between left ventricular strain measured by FT-CMR, morphologic features and prognostic markers in patients with HCM. Methods Retrospective analysis of clinical, echocardiography, Holter and CMR data of HCM patients aged ≥16 years followed at two referral centres. Ventricular arrhythmias (VA) were defined as non-sustained or sustained ventricular tachycardia or sudden cardiac arrest. Sudden cardiac death (SCD) risk was evaluated using the score proposed by the European Society of Cardiology. LGE extension was evaluated using the American Heart Association 17-segment model. FT-CMR was used to evaluate global peak systolic longitudinal (GLS), radial (GRS) and circumferential (GCS) strains - GLS was averaged from three standard longitudinal views while GRS and GCS were averaged from the basal, mid and apical LV short-axis planes. Results A total of 109 HCM patients (59.2±16.2 years old; 60.6% males) were included; mean follow-up was 39±25 months. Mean LV mass was 170.6±70.3g, LVEF was 63.7±10.0% and the number of segments with LGE was 3.14±3.32. Mean GLS, GRS and GCS were −14.8±4.0%, 34.4±13.3% and −17.5±4.8%, respectively. Impaired strain was associated with higher LV mass (GLS: r=0.46, GRS: r=−0.46, GCS: r=0.47, p<0.001 for all), reduced LVEF (GLS: r=−0,33, GRS: r=0,44, GCS: r=−0.41, p<0.003 for all) and LGE extension (GLS: r=0.26, GRS: r=−0.38, GCS: r=0.38, p<0.01 for all). SCD risk score was 3.12%±2.98 (8 patients scored as high risk) and VA were documented in 26 patients (26%). Patients with VA had worse strain values than those without (GLS −13.2±4.12 vs −15.5±3.71, p=0.011; GCS −15,8±5.22 vs −18.3±4.24, p=0.017). Patients with high estimated risk of SCD also had worse strain values than those at low/intermediate risk (GLS −12.2±3.57 vs −15.1±3.83, p=0.048; GCS −14.5±4.26 vs −17.9±4.54, p=0.047). A correlation between SCD risk and GLS and GCS was observed (r=0.32, p=0.004; r=0.23, p=0.03, respectively). Conclusions In our population, worse strain measurements were associated with a more severe HCM phenotype, presence of VA and a higher estimated risk of SCD. Strain assessed by FT-CMR may improve risk stratification in HCM patients.

Sign in / Sign up

Export Citation Format

Share Document