Deep learning algorithms for detecting cardiac arrests or arrhythmias using photoplethysmography and accelerometry: a systematic review

To this day, no systematic review has specifically studied the application of deep learning algorithms for the prediction or detection of arrhythmias and cardiac arrests using photoplethysmography (PPG) and accelerometry data. In 2022, a review conducted by Lee et al. evaluated the use of artificial intelligence for detecting cardiovascular-related diseases from wearable devices [1]. However, this review included reports of studies using traditional machine learning methods, and the majority of models were trained on ECG data. This systematic review aims to fill this gap and provides an up-to-date synthesis of the distinct architectures and performance of deep learning algorithms developed in peer-reviewed reports for the prediction or detection of arrhythmias and cardiac arrest using photoplethysmography (PPG) and accelerometry (ACC). Future research can therefore make well-founded decisions regarding algorithm architectures. For example, it may guide the development of deep learning algorithms for automatic cardiac arrest or arrhytmia detection with wearable devices that possess PPG and/or ACC sensors. [1] Lee S, Chu Y, Ryu J, Park YJ, Yang S, Koh SB. Artificial Intelligence for Detection of Cardiovascular-Related Diseases from Wearable Devices: A Systematic Review and Meta-Analysis. Yonsei Med J 2022;63:S93. https://doi.org/10.3349/ymj.2022.63.S93.