A Gradient Boosting-Based Approach for Intrusion Detection in IoT-Enabled Healthcare Systems

This study develops a sophisticated intrusion detection system to secure IoT-based healthcare networks against cyber-attacks. While traditional signature and anomaly-based methods struggle with low recall and adaptability, modern machine learning (ML) approaches still require improvement in accuracy and efficiency in response to emerging threats. To address this gap, we propose an intrusion detection technique based on Gradient Boosting (GB). The proposed model is comprehensively evaluated against several baseline algorithms, including Random Forest, Extra Trees, Decision Tree, and Naïve Bayes, using performance metrics such as precision, recall, F1score, and ROC-AUC. The results indicate that the GB-based model demonstrates superior performance compared to the other tested techniques on the experimental dataset, suggesting its strong potential for enhancing security in IoT-enabled healthcare environments.