AEGISCARE: A MULTIMODAL AI ASSISTANT FOR INTELLIGENT ELDERLY SAFETY AND HOME MONITORING

The existing elderly care systems are often limited by fragmented functionalities and a reactive approach, where assistance is provided only after an incident occurs. This lack of continuous monitoring and integrated support can lead to delayed responses during critical situations such as falls or prolonged inactivity, ultimately affecting the safety and well-being of elderly individuals. Additionally, many systems rely on wearable devices or manual supervision, which may not always be practical or reliable.   The AegisCare system addresses these limitations by providing an integrated and intelligent solution for real-time monitoring and assistance. By combining technologies such as computer vision, YOLOv8-based fall detection, inactivity monitoring, and AI-based object detection, the system ensures continuous observation and timely response to emergencies. The inclusion of voice interaction and automated alert mechanisms, including email notifications, further enhances usability and caregiver support.   Overall, AegisCare demonstrates the effectiveness of integrating multiple functionalities into a single platform to improve elderly safety, independence, and quality of life. The system provides a scalable and efficient solution that can be further extended to support advanced healthcare applications in the future.  

[1] Rieke, N., et al. (2020). The future of digital health with federated learning. NPJ Digital Medicine, 3, 119. [2] Sittig, D. F., et al. (2020). Precision medicine, AI, and the future of personalized health care. Yearbook of Medical Informatics, 29(1), 10–18. [3] Cao, Z., Hidalgo, G., Simon, T., Wei, S. E., & Sheikh, Y. (2021). OpenPose: Realtime multi-person 2D pose estimation using part affinity fields. IEEE Transactions on Pattern Analysis and Machine Intelligence. [4] Jocher, G., et al. (2023). Ultralytics YOLOv8: State-of-the-art object detection and pose estimation. Available: https://docs.ultralytics.com [5] Wang, L., et al. (2023). Blockchain and IoT enabled privacy-preserving medical data sharing for smart healthcare. Computers & Electrical Engineering, 107, 108605. [6] Thaker, K., et al. (2024). Enhancing health recommendations through patient metadata integration: A persona-based evaluation approach. HealthRecSys 2024. [7] Ultralytics. (2024). YOLOv8 Pose Estimation Documentation. Available: https://docs.ultralytics.com/tasks/pose/ [8] OpenCV. (2024). Open Source Computer Vision Library Documentation. Available: https://opencv.org