REAL-TIME ANOMALY DETECTION IN IOMT NETWORKS USING STACKING MODEL

Abstract

ijerstThe rapid growth of the Internet of Medical Things (IoMT) has revolutionized modern healthcare by enabling continuous patient monitoring, remote diagnostics, and intelligent medical decision-making. However, the integration of interconnected medical devices and cloud-based platforms introduces significant security vulnerabilities, making IoMT networks highly susceptible to cyberattacks and anomalous activities. Ensuring real-time anomaly detection is critical to maintaining patient safety, data privacy, and system reliability. This paper proposes a real-time anomaly detection framework for IoMT networks using a stacking ensemble learning model. The proposed approach integrates multiple base classifiers such as Random Forest, Support Vector Machine, and Gradient Boosting to enhance detection accuracy and robustness. These base models independently analyze network traffic patterns and device communication behaviors, while a meta-learner combines their predictions to produce a final optimized decision. The stacking model leverages the strengths of individual algorithms, reduces overfitting, and improves generalization performance compared to traditional single-model approaches. The system utilizes real-time network traffic data collected from IoMT devices, applies preprocessing techniques including feature extraction and normalization, and performs classification to distinguish between normal and anomalous activities. Experimental results demonstrate that the proposed stacking-based framework achieves higher accuracy, precision, recall, and F1-score compared to conventional machine learning methods. Furthermore, the model supports low-latency processing, making it suitable for deployment in real-time healthcare environments. Overall, the proposed solution enhances cybersecurity resilience in IoMT ecosystems by providing an intelligent, scalable, and efficient anomaly detection mechanism capable of safeguarding sensitive medical infrastructures.