Updated: May 2, 2020
Our manuscript is accepted for publication and published in the Heart Rhythm Journal 5/1/2020 (post updated).
We used machine learning (ML) to predict left-and right-sided S-ICD eligibility in ACHD patients, and developed transformation matrix to transform 12-lead ECG to S-ICD 3-lead ECG, and vice versa.
We studied a large group of complex OHSU adult congenital heart disease patients. Importantly, half of participants were female patients. Strikingly, 40% of participants were ineligible for S-ICD. Tetralogy of Fallot patients passed right-sided screening (57%) more often than left-sided screening (21%; McNemar's χ2 P=0.025). Female participants had greater odds of eligibility (adjusted OR 5.9 (95%CI 1.6-21.7); P=0.008). We developed and validated ML-based tools for S-ICD eligibility prediction and transformation matrices to transform 12-lead ECG to S-ICD 3-lead ECG, and vice versa.
We freely provide fully de-identified digital ECG data for all study participants, and open-source software code developed for the data analyses in GitHub.
Nearly half of the contemporary ACHD population is ineligible for S-ICD.
The odds of S-ICD eligibility are greater for female than male ACHD patients.
Machine-learning prediction of S-ICD eligibility can be used for screening of S-ICD candidates.