Internet of Hearts

The Internet of Hearts (IoH) is a new cardiac telemedicine system. It leverages wearable device, wireless sensing, mobile technology and cloud computing for continuous cardiac monitoring, early identification of acute cardiac events, timely delivery of life-saving, personalized therapies, and smart management of cardiac health. It involves both cardiac patients and cardiologists, and it is a Internet of Things (IoT) technology that specific to the cardiac healthcare.
History
The IoH is proposed by Dr. Hui Yang and Chen Kan from Harold and Inge Marcus Department of Industrial and Manufacturing Engineering, the Pennsylvania State University. The core technology of IoH was firstly patented in the year of 2012, which later appeared in the paper "Spatiotemporal differentiation of myocardial infarctions" in 2013. The complete idea of IoH was presented in the paper "Mobile sensing and network analytics for realizing smart automated systems towards health Internet of Things" in 2015 IEEE International Conference on Automation Science and Engineering (CASE 2015). This work is supported in part by the National Science Foundation (CMMI-1617148 and IIP-1447289).
Motivation
Since 2000, the Internet of Things has been hailed as a revolution of automation science and information technology. The IoT system deploys a multitude of wireless sensors, mobile computing units, and physical objects in an Internet-like infrastructure. This provides an unprecedented opportunity to realize smart automated systems such as smart manufacturing, smart health, smart transportation, and smart home. As every agent is capable of sensing, computing and communicating seamlessly to others, large-scale IoT systems lead to the accumulation of big data. Notably, wearable sensing and mobile technology accelerate human-centered computing for smart health management. People are of greater importance vis-à-vis other physical objects in the IoT system. Understanding human beings and their ambient variables is critical to developing smart automated systems. Rapid advancements of sensing technology lead to more sophisticated and multi-purpose wearable sensing products.
However, most of existing products (e.g., fitbit) focus on wearable sensing and fitness applications, while are limited in the capability for cardiac sensing and clinical applications. Very little work has been done to develop advanced IoT technologies for smart monitoring and control of heart health. It is worth mentioning that heart diseases are the leading cause of death in the world. In 2012, nearly 30% of global deaths (17.5 million) were due to cardiac diseases. There is an urgent need to develop a new IoT technology specific to the heart that will facilitate early identification of acute cardiac events, timely delivery of life-saving, personalized therapies, and smart management of heart health.
Components
Wireless ECG Sensing
The portable sensing device in IoH is not only able to record hospital-grade multi-lead ECG, but also comfortable, flexible and reliable to facilitate long-term continuous monitoring. Existing electrodes are foam-made, fixed-shape and attached to the skin by electrolyte gel. They do not adhere well to the irregular body surface, thereby resulting in artifacts during body movement. A new generation of ECG sensor is adopted in IoH, which is built on stretchable substrates that can stretch, fold, twist and wrap around complex surface of the skin. A Bluetooth LE module is included in the device to transmit ECG signals wirelessly to mobile devices.
Dynamic Network for Health Analytics
IoH leverages the network structure in large-scale IoT systems to enhance the information-processing capability of big data for disease pattern recognition using space-time Vectorcardiogram (VCG) signals. It is aimed at quantifying of spatiotemporal dissimilarity between functional recordings, which provides a great opportunity for the identification of cardiovascular diseases. However, due to phase shift and discrete sampling, two VCG signals can be misaligned, e.g., both signals show a typical pattern and yet there are variations in shape, amplitude and phase between them. The spatiotemporal warping approach optimally aligns P, QRS, and T loops in the VCG signals for two subjects in both space and time.<ref name=":0" /> Such an alignment is critical to compare the corresponding electrical activity of heart chambers. Further, IoH employs a network embedding algorithm for transforming the warping matrix into feature vectors that preserve the warping distances among functional recordings. As such, each VCG loop is embedded as a node (coordinates as predictive features) in the high-dimensional network that preserves the dissimilarity distance matrix. When a new VCG recording is presented in the practice, the pattern dissimilarity will be measured against the database of patients. Then, a new row and column will be obtained in the warping matrix, and a new node will be embedded in the high-dimensional network. Finally, classification models can be used to predict cardiac conditions with node coordinates (i.e., feature vectors).
Experiments
The core technology of IoH was evaluated using real-world data. As it is reported in the paper "Spatiotemporal differentiation of myocardial infarctions",<ref name=":0" /> the algorithm successfully distinguished between healthy control and myocardial infarction subjects, as well as between subjects with infarctions in different locations of the heart. In the paper "Mobile sensing and network analytics for realizing smart automated systems towards health Internet of Things",<ref name=":1" /> patients with different types of left bundle branch blocks were effectively separated by the core technology of IoH.

Main Menu

See Also

Main Menu