The increasing elderly population in Indonesia has contributed to a higher prevalence of non-communicable diseases, including hypertension, cardiovascular disease, diabetes mellitus, and chronic obstructive pulmonary disease (COPD). This trend highlights the need for continuous, real-time health monitoring rather than periodic medical assessments. This study aims to design an Internet of Things (IoT)-based elderly health monitoring information system integrated with Federated Learning (FL) to enable early health-risk prediction while preserving the privacy of users' medical data. The research adopts the Design Science Research (DSR) methodology, consisting of problem identification, requirements analysis, system design, implementation, and evaluation. The proposed prototype employs an ESP32 microcontroller integrated with MAX30102, blood pressure, DS18B20 body temperature, and MPU6050 sensors to collect vital-sign data and transmit it to a web-based dashboard via the MQTT protocol. Health-risk classification is performed using the Federated Averaging (FedAvg) algorithm, allowing model training across user devices without transferring raw data. Experimental results demonstrate that the system provides real-time vital-sign monitoring and achieves health-risk classification performance comparable to centralized machine learning, indicating its potential as a privacy-preserving telehealth platform for elderly care in households, community health centers, and nursing homes.
Ilham Satya Nugraha, Erwin Halim· Zenodo (CERN European Organi...· 0 citations
The increasing elderly population in Indonesia has contributed to a higher prevalence of non-communicable diseases, including hypertension, cardiovascular disease, diabetes mellitus, and chronic obstructive pulmonary disease (COPD). This trend highlights the need for continuous, real-time health monitoring rather than periodic medical assessments. This study aims to design an Internet of Things (IoT)-based elderly health monitoring information system integrated with Federated Learning (FL) to enable early health-risk prediction while preserving the privacy of users' medical data. The research adopts the Design Science Research (DSR) methodology, consisting of problem identification, requirements analysis, system design, implementation, and evaluation. The proposed prototype employs an ESP32 microcontroller integrated with MAX30102, blood pressure, DS18B20 body temperature, and MPU6050 sensors to collect vital-sign data and transmit it to a web-based dashboard via the MQTT protocol. Health-risk classification is performed using the Federated Averaging (FedAvg) algorithm, allowing model training across user devices without transferring raw data. Experimental results demonstrate that the system provides real-time vital-sign monitoring and achieves health-risk classification performance comparable to centralized machine learning, indicating its potential as a privacy-preserving telehealth platform for elderly care in households, community health centers, and nursing homes.
Ilham Satya Nugraha, Erwin Halim· Zenodo (CERN European Organi...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
