Mapping critical illness early signs to priority alert transmission on wireless networks
2017 IEEE 9th Latin-American Conference on Communications (LATINCOM)(2017)
摘要
The prediction of adverse events in patient health has attracted researchers' attention for many years. Sensors in or on the body allow the continuous monitoring of patient's vital signs and their accurate analysis, supporting the generation of medical alerts. Exploring adverse event prediction only makes sense if medical alerts can be promptly transmitted to a hospital emergency response team. However, promptly transmitting them through heterogeneous wireless networks is still a challenge because of wireless communication features, i.e., interferences and collisions, and the current medium access control (MAC) protocol design, that still produces competition between medical alerts, and video, voice and other types of data. Differently from current mechanisms that either are separately concerned to the prediction of adverse events or to give priority to different type of data transmission, this work presents SANTE, a System for Anticipated identificatioN and Transmission of mEdical alerts on wireless networks. Based on trends about the imminence of adverse events on patient health, the system generates medical alerts and promptly transmit them. It presents a novel proposal to medium access control for medical alerts, reducing contention window and Arbitration Inter-Frame Spacing (AIFS) time for them. Simulation results show a reduction of 39% in the average latency for alert transmissions and 8% in losses.
更多查看译文
关键词
patient health,medical alerts,priority alert transmission,adverse event prediction,heterogeneous wireless networks,current medium access control protocol design,critical illness early sign mapping,hospital emergency response team,wireless communication features,interferences-and-collisions,data transmission,SANTE,system-for-anticipated identification-and-transmission-of-medical alerts,reducing contention window,arbitration interframe spacing time,AIFS time
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络