The healthcare industry is rapidly adopting new technologies such as the Internet of Things (IoT), which are dropping costs and improving healthcare outcomes. Such IoT systems typically include edge devices (glucose monitors, ventilators, pacemakers), gateway devices that aggregate the data from the edge devices and transmit it to the cloud, and cloud-based systems which analyze the device data to draw conclusions, display information, or direct the connected devices to take action. This process can lead to communication lags and delayed responses to patient conditions/treatment. The aim of this proposal is to overcome these delays with IoT technology and allow for prompt urgent treatment to patients. The solution proposed includes a model to monitor and process the data disseminated by wearable devices related to the patients’ health issues and connect the data to IoT cloud platforms. Analysis of the patients’ health data to identify anomalies will be performed at the device level by developing an offline machine learning model using specific algorithms for anomaly detection and deploying them on the IoT devices or IoT gateway. Processing of the real-time health data will be performed at the device level and the prediction of anomalous data will be sent to the third-party cloud for implementing any necessary actions.
Paper
Full text
Health Monitoring with Low Power IoT Devices using Anomaly Detection Algorithm
Semantic Scholar · Engineering · 2019
Abstract
The healthcare industry is rapidly adopting new technologies such as the Internet of Things (IoT), which are dropping costs and improving healthcare outcomes. Such IoT systems typically include edge devices (glucose monitors, ventilators, pacemakers), gateway devices that aggregate the data from the edge devices and transmit it to the cloud, and cloud-based systems which analyze the device data to draw conclusions, display information, or direct the connected devices to take action. This process can lead to communication lags and delayed responses to patient conditions/treatment. The aim of this proposal is to overcome these delays with IoT technology and allow for prompt urgent treatment to patients. The solution proposed includes a model to monitor and process the data disseminated by wearable devices related to the patients’ health issues and connect the data to IoT cloud platforms. Analysis of the patients’ health data to identify anomalies will be performed at the device level by developing an offline machine learning model using specific algorithms for anomaly detection and deploying them on the IoT devices or IoT gateway. Processing of the real-time health data will be performed at the device level and the prediction of anomalous data will be sent to the third-party cloud for implementing any necessary actions.