IoT-enabled Intelligent Dynamic Risk Assessment of Acute Mountain Sickness Based on Data from Wearable Devices
- Resource Type
- Conference
- Authors
- Chen, Jing; Tian, Yuan; Zhang, Guangbo; Cao, Zhengtao; Zhu, Lingling; Shi, Dawei
- Source
- 2021 4th IEEE International Conference on Industrial Cyber-Physical Systems (ICPS) Industrial Cyber-Physical Systems (ICPS), 2021 4th IEEE International Conference on. :132-137 May, 2021
- Subject
- Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Robotics and Control Systems
Signal Processing and Analysis
Performance evaluation
Correlation
Wearable computers
Benchmark testing
Real-time systems
Indexes
Risk management
Medical IoT
Acute mountain sickness
Hypoxia training
Performance assessment
Process monitoring
- Language
Advances in wearable devices and medical internet of things (IoT) have enabled the intelligent measurement and monitoring of key physiological variables. In this work, we aim to build an acute mountain sickness (AMS) risk evaluation index based on data from wearable devices from a performance monitoring perspective, made possible by the medical IoT architecture. Through exploring the dynamic properties of real-time data and exploiting the underlying relationship between the AMS risk and the bandwidth of the hypoxic stress response, a dynamic SpO 2 (peripheral oxygen saturation) index (DSI) with AMS risk evaluation potential is proposed and a robust index evaluation procedure is developed to rule out the effect of measurement noises and deep-breath related disturbances. The effectiveness of DSI was assessed based on physiological data from a proof-of-the-concept clinical study ($\mathrm{N}=12$). The relationship of DSI with existing AMS metrics is analyzed through correlation analysis. Statistically significant correlation between DSI and AMS metrics (Lake Louise Score (LLS); deep sleep duration; deep sleep ratio; and mean SpO 2 ) was observed. A benchmark value for DSI was determined based on the critical value of LLS. The proposed method and experimental results indicate the feasibility of improving AMS risk evaluation performance using intelligent monitoring techniques.