Deep learning methods for screening patients' S-ICD implantation eligibility

作者：

Highlights：

• We propose an accurate, reliable, and automated method to better scrutinise patients’ S-ICD eligibility.

• Our method is based on a CNN framework for predicting the T:R ratio from PSR images of filtered 10-second ECG segments.

• The method is designed for easy integration into a clinical tool accessing leads having low risk of inappropriate shocks.

摘要

•We propose an accurate, reliable, and automated method to better scrutinise patients’ S-ICD eligibility.•Our method is based on a CNN framework for predicting the T:R ratio from PSR images of filtered 10-second ECG segments.•The method is designed for easy integration into a clinical tool accessing leads having low risk of inappropriate shocks.

论文关键词：Subcutaneous implantable cardioverter-defibrillators,Sudden cardiac death,Ventricular arrhythmia,Electrocardiogram,Deep learning,Convolutional neural networks,Phase space reconstruction,Patient screening

论文评审过程：Received 9 March 2021, Revised 20 June 2021, Accepted 3 August 2021, Available online 9 August 2021, Version of Record 20 August 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102139