An intelligent healthcare system with data priority based on multi vital biosignals

Area of Science:

• Biomedical Engineering
• Health Informatics
• Internet of Things (IoT)

Background:

• Home-based healthcare systems increasingly use Internet of Things (IoT) devices for continuous patient monitoring.
• Large volumes of multi-biosignal data from numerous patients cause transmission overhead and high latency, hindering clinical event prediction.
• Existing systems face challenges in efficiently storing and processing the enormous data generated by IoT healthcare devices.

Purpose of the Study:

• To introduce a data assessment method to enhance efficiency in data collection and prediction for IoT-based healthcare systems.
• To address the inefficiencies caused by transmission overhead and high latency in processing large volumes of patient data.
• To optimize the performance of clinical event prediction in remote healthcare settings.

Main Methods:

• Developed a data assessment algorithm inspired by the National Early Warning Score (NEWS).
• Implemented a system that transmits only abnormal time-sequence biosignal data to the server for analysis.
• Prioritized data processing based on urgency, ensuring critical patient information is handled promptly.

Main Results:

• The proposed model demonstrated a potential data volume reduction of 20% during collection.
• Achieved a 75% reduction in waiting time for high-priority data before prediction.
• Optimized overall data waiting time for further analysis compared to standard processing flows.

Conclusions:

• Introduced an enhanced healthcare system utilizing data priority assessment.
• Successfully optimized data collection and prediction by reducing data size and minimizing waiting times.
• The findings suggest a more efficient approach to managing and analyzing patient data in IoT healthcare.