A Secure and Intelligent IoT-Based Remote Patient Monitoring Framework for Early Detection of Alzheimer’s Disease Using Edge AI and Federated Learning

Alzheimer's disease is a progressive neurodegenerative disorder and is widely recognized as a major burden on healthcare systems. Symptoms are often delayed; therefore, timely diagnosis is essential to improve patients' quality of life. However, traditional periodic clinical assessments may not accurately capture continuous changes in behavior and physiology characteristic of early-stage Alzheimer's. We present a secure Internet of Things (IoT)-based remote patient monitoring (RPM) framework. This model utilizes environmental and wearable sensors to collect physiological signals and behavioral data. Real-time data processing at the device level, combined with anomaly detection using Edge AI, minimizes latency and enables rapid response. To address privacy concerns, Federated Learning (FL) is incorporated to enable decentralized model training without transferring raw patient data. A hybrid anomaly detection model based on time-series behavioral analysis identifies patterns of early cognitive decline. Simulation-based evaluation suggests that the proposed framework improves detection accuracy and reduces latency by up to 40-50% compared to traditional cloud-based systems. This architecture provides a scalable and privacy-aware solution for intelligent healthcare systems. In addition, such intelligent monitoring systems can support healthcare professionals in making timely and informed clinical decisions. The integration of advanced data analytics with real-time monitoring provides a promising direction for improving early diagnosis and long-term patient management in neurodegenerative diseases.