Comprehensive Summary
This study by Lee et al. examines the application of several artificial intelligence (AI) models for the detection and prediction of cardiovascular-related diseases through wearable and mobile devices. A meta-analysis was conducted following PRISMA 2020 guidelines using 102 studies from databases including Medline, Embase, and Cochrane Library. AI models showed strong performance. Arrhythmia was the main focus of the studies, and AF detection was the most common task. Meta-analysis showed that sensitivity was 94.80% and specificity was 96.96%.Deep neural networks outperformed conventional machine learning methods; however, AI models trained on data from wearable devices performed worse than those evaluated on public datasets.
Outcomes and Implications
Cardiovascular diseases are the leading causes of death around the world, and early detection using wearable AI-enabled devices could reduce hospitalizations, lower healthcare costs, and prevent severe complications. There are limitations to widespread use for these wearable devices, such as poor signal quality and false alarms, resulting in unnecessary testing. Nonetheless, wearable device-enabled detection of cardiovascular-related diseases is likely to become more common clinically as healthcare technology is more prevalent. The FDA has already approved some wearable monitoring devices, but more testing will be required for widespread use.