Published online Sep 16, 2022. doi: 10.12998/wjcc.v10.i26.9207
Peer-review started: February 11, 2022
First decision: June 7, 2022
Revised: June 19, 2022
Accepted: July 25, 2022
Article in press: July 25, 2022
Published online: September 16, 2022
The coronavirus disease 2019 (COVID-19) has currently caused the mortality of millions of people around the world. Aside from the direct mortality from the COVID-19, the indirect effects of the pandemic have also led to an increase in the mortality rate of other non-COVID patients. Evidence indicates that novel COVID-19 pandemic has caused an inflation in acute cardiovascular mortality, which did not relate to COVID-19 infection. It has in fact increased the risk of death in cardiovascular disease (CVD) patients. For this purpose, it is dramatically inevitable to monitor CVD patients’ vital signs and to detect abnormal events before the occurrence of any critical conditions resulted in death. Internet of things (IoT) and health monitoring sensors have improved the medical care systems by enabling latency-sensitive surveillance and computing of large amounts of patients’ data. The major challenge being faced currently in this problem is its limited scalability and late detection of cardiovascular events in IoT-based computing environments. To this end, this paper proposes a novel framework to early detection of cardiovascular events based on a deep learning architecture in IoT environments. Experimental results showed that the proposed method was able to detect cardiovascular events with better performance (95.30% average sensitivity and 95.94% mean prediction values).
Core Tip: This paper has focused on presenting a health monitoring system for cardiovascular disease patients during coronavirus disease 2019 pandemic. For this purpose, a new framework for early detection of cardiovascular events was proposed based on a deep learning architecture in internet of things environments. The proposed method has provided a peaceful solution for limited scalability and late detection of cardiovascular events by enabling latency-sensitive surveillance and computing of large amounts of patients’ data.