A Framework for Remote Interaction and Management of Home Care Elderly Adults

Growing aging population highlights the importance of managing chronic diseases. The rapid development of Internet of Things (IoT) and big data analysis makes it feasible and affordable to monitor and manage chronic diseases for caring at home. The process of chronic diseases management involves mon...

Full description

Saved in:
Bibliographic Details
Published inIEEE sensors journal Vol. 22; no. 11; pp. 11034 - 11044
Main Authors Zhang, Bin, Zhu, Leqi, Pei, Zichen, Zhai, Qian, Zhu, Junhong, Zhong, Xiang, Yi, Jingang, Liu, Tao
Format Journal Article
LanguageEnglish
Published New York IEEE 01.06.2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Adults
Artificial neural networks
Big Data
Biomedical monitoring
Cameras
Chronic illnesses
Data acquisition
Data analysis
Decision trees
Diagnosis
Diseases
Heart diseases
Internet of Things
millimeter wave sensors
Monitoring
Neural networks
Older people
Radar
Rehabilitation
sensor data processing
Sensor system integration
sensor system networks
Sensors
Signal transmission
Support vector machines
Telemedicine
Wearable sensors
Online AccessGet full text

Cover

More Information
Summary:Growing aging population highlights the importance of managing chronic diseases. The rapid development of Internet of Things (IoT) and big data analysis makes it feasible and affordable to monitor and manage chronic diseases for caring at home. The process of chronic diseases management involves monitoring rehabilitation and recovery, tracking physiological and behavioral status, and health condition classification and diagnosis. In this paper, we proposed a framework for monitoring and management of home care elderly adults. A three-level architecture, which is IoT-Intelligent Terminal(IT)-Cloud, is established to achieve data acquisition, signal transmission, remote interaction and diagnosis. Five diagnosis methods including Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), K Near Neighbor (KNN), and Back Process Neural Network (BPNN) are implemented to evaluate the risk of suffering from heart disease and experiments showed the accuracy is over 95%. Experimental result reveals that the data flow and remote interaction in this system are effective and primary diagnosis is validated.
ISSN:1530-437X
1558-1748
DOI:10.1109/JSEN.2022.3170295