Dataset Information

Electrocardiogram machine learning for detection of cardiovascular disease in African Americans: the Jackson Heart Study.

ABSTRACT:

Aims

Almost half of African American (AA) men and women have cardiovascular disease (CVD). Detection of prevalent CVD in community settings would facilitate secondary prevention of CVD. We sought to develop a tool for automated CVD detection.

Methods and results

Participants from the Jackson Heart Study (JHS) with analysable electrocardiograms (ECGs) (n=3679; age, 6212 years; 36% men) were included. Vectorcardiographic (VCG) metrics QRS, T, and spatial ventricular gradient vectors magnitude and direction, and traditional ECG metrics were measured on 12-lead ECG. Random forests, convolutional neural network (CNN), lasso, adaptive lasso, plugin lasso, elastic net, ridge, and logistic regression models were developed in 80% and validated in 20% samples. We compared models with demographic, clinical, and VCG input (43 predictors) and those after the addition of ECG metrics (695 predictors). Prevalent CVD was diagnosed in 411 out of 3679 participants (11.2%). Machine learning models detected CVD with the area under the receiver operator curve (ROC AUC) 0.690.74. There was no difference in CVD detection accuracy between models with VCG and VCG + ECG input. Models with VCG input were better calibrated than models with ECG input. Plugin-based lasso model consisting of only two predictors (age and peak QRS-T angle) detected CVD with AUC 0.687 [95% confidence interval (CI) 0.6250.749], which was similar (P=0.394) to the CNN (0.660; 95% CI 0.5970.722) and better (P<0.0001) than random forests (0.512; 95% CI 0.4930.530).

Conclusions

Simple model (age and QRS-T angle) can be used for prevalent CVD detection in limited-resources community settings, which opens an avenue for secondary prevention of CVD in underserved communities.

SUBMITTER: Pollard JD

PROVIDER: S-EPMC8139412 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundStudies consisting mostly of whites have shown that the prevalence of masked hypertension differs by prehypertension status. Using data from the Jackson Heart Study, an exclusively African American population-based cohort, we evaluated the association of masked hypertension and prehypertension with left ventricular mass index and common carotid intima media thickness.Methods and resultsAt the baseline visit, clinic blood pressure (CBP) measurement and 24-hour ambulatory blood pressure monitoring were performed. Masked hypertension was defined as mean systolic/diastolic CBP <140/90 mm Hg and mean daytime systolic/diastolic ambulatory blood pressure ≥135/85 mm Hg. Clinic hypertension was defined as mean systolic/diastolic CBP ≥140/90 mm Hg. Normal CBP was defined as mean systolic/diastolic CBP <120/80 mm Hg and prehypertension as mean systolic/diastolic CBP 120 to 139/80 to 89 mm Hg. The analytic sample included 909 participants. Among participants with systolic/diastolic CBP <140/90 mm Hg, the prevalence of masked hypertension and prehypertension was 27.5% and 62.4%, respectively. The prevalence of masked hypertension among those with normal CBP and prehypertension was 12.9% and 36.3%, respectively. In a fully adjusted model, which included prehypertension status and antihypertensive medication use as covariates, left ventricular mass index was 7.94 g/m(2) lower among those without masked hypertension compared to participants with masked hypertension (P<0.001). Left ventricular mass index was also 4.77 g/m(2) lower among those with clinic hypertension, but this difference was not statistically significant (P=0.068). There were no significant differences in left ventricular mass index between participants with and without masked hypertension, or clinic hypertension.ConclusionsMasked hypertension was common among African Americans with prehypertension and also normal CBP, and was associated with subclinical cardiovascular disease.