Key result
Nearly 40% of adult congenital heart disease patients were ineligible for subcutaneous implantable cardioverter-defibrillators, and a ridge machine learning model predicted standing left-sided eligibility with an ROC AUC of 0.687.
Why the study?
The study was conducted to assess left- and right-sided S-ICD eligibility in ACHD patients, use machine learning to predict eligibility, and transform 12-lead ECG to S-ICD 3-lead ECG and vice versa.
Can machine learning models and 12-lead ECG transformations accurately predict S-ICD eligibility in adult congenital heart disease patients?
Cross-Sectional (n=101)
No
Can machine learning models and 12-lead ECG transformations accurately predict S-ICD eligibility in adult congenital heart disease patients?
Nearly half of adult congenital heart disease patients are ineligible for S-ICD based on standard screening, but machine learning applied to standard 12-lead ECGs can help predict eligibility and streamline screening.
Answers are grounded in this paper and cite their sources.
ECG-based ML screening may aid ACHD eligibility assessment; leaves open prospective validation before clinical adoption.
Background The goals of this study were: assess left-and right-sided subcutaneous implantable cardioverter-defibrillator (S-ICD) eligibility in adult congenital heart disease (ACHD) patients, use machine learning to predict S-ICD eligibility in ACHD patients, and transform 12-lead ECG to S-ICD 3-lead ECG, and vice versa. Methods ACHD outpatients (n=101; age 42±14 y; 52% female; 85% white; left ventricular ejection fraction (LVEF) 56±9%) were enrolled in a prospective study. Supine and standing 12-lead ECG was recorded simultaneously with a right- and left-sided S-ICD 3-lead ECG. Peak-to-peak QRS and T amplitudes, RR, PR, QT, QTc, QRS intervals, T max , and R/T max (31 predictor variables) were tested. Model selection, training, and testing were performed using supine ECG datasets. Validation was performed using standing ECG datasets and out-of-sample non-ACHD population (n=68; age 54±16 y; 54% female; 94% white; LVEF 61±8%). Results A 40% of participants were ineligible for S-ICD. Tetralogy of Fallot patients passed right-sided screening (57%) more often than left-sided (21%; McNemar’s χ 2 P =0.025). The ridge model demonstrated the best cross-validation function. Validation of the ridge models was satisfactory for standing left-sided [ROC AUC 0.687 (95%CI 0.582-0.791)] and right-sided [ROC AUC 0.655(95%CI 0.549-0.762)] S-ICD eligibility prediction. Out-of-sample validation in the non-ACHD population yielded a 100% sensitivity of the pre-selected threshold for the elastic net model. Validation of the transformation matrices showed satisfactory agreement (<0.1 mV difference). Conclusion Nearly half of the contemporary ACHD population is ineligible for S-ICD. Machine-learning prediction of S-ICD eligibility can be used for screening of S-ICD candidates.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2019) conducted a cross-sectional in Adult congenital heart disease (n=101). Machine learning prediction of S-ICD eligibility was evaluated on Standing left-sided S-ICD eligibility prediction (ROC AUC) (ROC AUC 0.687, 95% CI 0.582-0.791). Nearly 40% of adult congenital heart disease patients were ineligible for subcutaneous implantable cardioverter-defibrillators, and a ridge machine learning model predicted standing left-sided eligibility with an ROC AUC of 0.687.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: