Abstract
Millions of people die every year from causes that can be prevented. With the global ageing population and unhealthy practices, such as sedentary lifestyle, high-calorie diets, and chronic stress, the incidence of chronic diseases is increasing. 5G technologies and sensor networks are emerging as key transformative technologies in the health domain. They promise new opportunities for continuous monitoring of health, including cardiac health and chronic diseases or conditions (obesity, diabetes, and others). Healthcare delivery is changing emphasis from hospital to ambulatory care, and to home healthcare. This transition involves monitoring vital signs using multiple sensors and the Internet of Things (IoT) devices that may generate large data streams.
We developed a multi-sensor system for personalized heart health monitoring. Multi-sensor systems including sensor-based wearable devices and Internet of Things (IoTs) devices. They produce massive amounts of health-related data including vital signs, environment variables, and activity-based variables. The health monitoring system based on sensors and sensor network produces health-related ‘big data’ that leads to challenges in data management and analysis. The major data issues we met are: (i) Data collection and cleaning for adequate accurate data; (ii) Data interpretation. For example, definitions of the normal and abnormal patterns of heart rate; (iii) Data interoperability and the lack of eHealth data standards applied to consumer devices; (iv) Regulatory compliance, legal, and ethical issues for data ownership; (v) Knowledge management and decision-making support.
This framework involves heart rate (HR) data streams captured by sensors (ECG or PPG) that are transferred to mobile device for collecting and edge computing device for processing. Other measured entities in our framework include breathing rate, weight, activity, sleep, blood pressure, and blood sugar monitoring. Reference databases and charts are used for the comparison of observed vital signs with patterns representing healthy heart function or various cardiac pathologies and comorbidities. All data processing is performed locally using edge computing devices, and regular reports are generated to form personal health record. Smart algorithms continuously analyze data streams. Data sharing with health care providers is facilitated by blockchain. Smart algorithms based on artificial intelligence and big data analytics might offer certain advantages than the traditional approach of health monitoring, such as continuous monitoring, early identification of abnormal functions, and low additional burdens on the healthcare services.
Rapid advancement of sensor systems, information and communication technologies, and knowledge management abilities facilitate the transformation of healthcare from curative (focus on the treatment of illness) to preventative (prevention of disease, early diagnosis, and treatment at early stages). The proposed framework and solutions have great potentials for promoting and changing the healthcare model for personalized health monitoring with consumer devices.
We developed a multi-sensor system for personalized heart health monitoring. Multi-sensor systems including sensor-based wearable devices and Internet of Things (IoTs) devices. They produce massive amounts of health-related data including vital signs, environment variables, and activity-based variables. The health monitoring system based on sensors and sensor network produces health-related ‘big data’ that leads to challenges in data management and analysis. The major data issues we met are: (i) Data collection and cleaning for adequate accurate data; (ii) Data interpretation. For example, definitions of the normal and abnormal patterns of heart rate; (iii) Data interoperability and the lack of eHealth data standards applied to consumer devices; (iv) Regulatory compliance, legal, and ethical issues for data ownership; (v) Knowledge management and decision-making support.
This framework involves heart rate (HR) data streams captured by sensors (ECG or PPG) that are transferred to mobile device for collecting and edge computing device for processing. Other measured entities in our framework include breathing rate, weight, activity, sleep, blood pressure, and blood sugar monitoring. Reference databases and charts are used for the comparison of observed vital signs with patterns representing healthy heart function or various cardiac pathologies and comorbidities. All data processing is performed locally using edge computing devices, and regular reports are generated to form personal health record. Smart algorithms continuously analyze data streams. Data sharing with health care providers is facilitated by blockchain. Smart algorithms based on artificial intelligence and big data analytics might offer certain advantages than the traditional approach of health monitoring, such as continuous monitoring, early identification of abnormal functions, and low additional burdens on the healthcare services.
Rapid advancement of sensor systems, information and communication technologies, and knowledge management abilities facilitate the transformation of healthcare from curative (focus on the treatment of illness) to preventative (prevention of disease, early diagnosis, and treatment at early stages). The proposed framework and solutions have great potentials for promoting and changing the healthcare model for personalized health monitoring with consumer devices.
| Original language | English |
|---|---|
| Publication status | Published - 5 Jul 2021 |
ASJC Scopus subject areas
- Health Informatics
- Computer Science (all)
