Soft Transducer for Patient's Vitals Telemonitoring with Deep Learning-Based Personalized Anomaly Detection

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Healthcare
• Wearable Technology

Background:

• Patient-centered vitals telemonitoring is crucial for managing chronic conditions like hypertension.
• Existing systems often lack personalization and robust identity verification.
• Integration of Industry 4.0 technologies enables advanced remote healthcare solutions.

Purpose of the Study:

• To design, develop, and implement an Industry 4.0-based wearable soft transducer for patient-centered vitals telemonitoring.
• To create a deep learning algorithm for personalized vital sign anomaly detection and alerts.
• To develop a mobile application for data management and patient identity verification.

Main Methods:

• A wearable soft transducer was developed to measure heart rate, oxygen saturation, and blood pressure.
• A Long-Short-Term-Memory Autoencoder deep learning model was implemented for anomaly detection.
• A mobile application with face-detection was created for data flow management and user authentication.

Main Results:

• The system demonstrated high accuracy in anomaly detection, exceeding 93%.
• A true positive rate of over 94% was achieved in identifying critical vital sign deviations.
• Experimental validation on five individuals over 30 days confirmed the system's robustness.

Conclusions:

• The developed wearable soft transducer and AI algorithm offer a robust solution for personalized remote patient monitoring.
• The system effectively detects hypertension-related vital sign anomalies, enabling timely alerts.
• The integrated mobile application enhances data accessibility and ensures patient identity verification for secure telehealth.