Akademik Veri Yönetim Sistemi
Diğer Uluslararası Fon Programları, 2024 - 2026
This project proposes an advanced eHealth framework that leverages Internet of Things (IoT) technologies, multi-sensor data, and artificial intelligence (AI) to improve health monitoring and emergency detection, particularly for elderly and vulnerable populations. The system integrates wearable, ambient, and multimedia sensors to enable real-time human activity recognition and anomaly detection, addressing limitations in existing approaches that rely on single-sensor data. By combining machine learning (ML) and deep learning (DL) techniques within a hybrid two-stage architecture, the framework enhances the accuracy and reliability of detecting critical events such as falls and health irregularities while maintaining user privacy.
The research introduces novel data fusion strategies, edge computing mechanisms, and intelligent anomaly detection models, including a Multivariate Anomaly Detector (MaD), to improve responsiveness and reduce energy consumption. A prototype testbed will be developed and evaluated using real and simulated data to validate system performance. The proposed solution has significant societal impact, offering a cost-effective and scalable approach to early health risk detection, improved patient monitoring, and enhanced quality of life. By addressing key challenges in multimodal data integration, interpretability, and real-time processing, this project contributes to the advancement of next-generation smart healthcare systems.