Deepthy Varghese, MSN, ACNP, FNP, Northside Hospital is joined by Tina Baykaner, MD, MPH Stanford University, and Gurukripa N Kowlgi, MBBS, MSci, Mayo Clinic–Rochester to discuss; the multicenter study investigated the potential of machine learning (ML) models to improve risk stratification for implantable cardioverter-defibrillator (ICD) implantation in patients at risk of sudden cardiac death (SCD). By combining clinical variables with 12-lead electrocardiogram (ECG) time-series features, the models aimed to predict non-arrhythmic mortality within three years after device implantation. Results showed that ML models identified patients at risk with high accuracy, demonstrating robust performance in both the development and external validation cohorts. This suggests that ML-based approaches could enhance risk assessment for SCD prevention in primary prevention populations.

https://www.hrsonline.org/education/TheLead https://academic.oup.com/europace/article/25/9/euad271/7274626

Host Disclosure(s): D. Varghese: Nothing to disclose

Contributor Disclosure(s): G. Kowlgi: Nothing to disclose T. Baykaner: Honoraria, Speaking, and Consulting: Medtronic Inc., Pacemate, Research: NIH

This episode has .25 ACE credits associated with it. If you want credit for listening to this episode, please visit the episode page on HRS365 https://www.heartrhythm365.org/URL/TheLeadEpisode67