1Professor, HOD, Department of CSE, KGISL Institute of Technology, Coimbatore, 641035, TamilNadu, India. Email: hodcse@kgkite.ac.in
2Department of CSE, KGISL Institute of Technology, Coimbatore, 641035, TamilNadu, India. Email: makeshl023@gmail.com
3Department of CSE, KGISL Institute of Technology, Coimbatore, 641035, TamilNadu, India. Email: sabarishkovalan@gmail.com
4Department of CSE, KGISL Institute of Technology, Coimbatore, 641035, TamilNadu, India. Email: sreedharsancbe@gmail.com
5Department of CSE, KGISL Institute of Technology, Coimbatore, 641035, TamilNadu, India. Email: vasanth8447@gmail.com
Internet of Medical Things (IoMT) continues to be increasingly integrated in modern healthcare systems as a form of real-time specialist healthcare along with data exchange among medical equipment. Nevertheless, the high rate of interconnected medical equipment development also preconditions the emergence of a range of cybersecurity threats that can alarm patient safety and confidentiality of their data. This study suggests a hybrid intrusion detection agenda to solve these concerns, a pool of machine learning and deep learning processes. The suggested system combines the feature selection based on a Grey Wolf Optimizer (GWO), a Random Forest classifier to determine known attacks, and an autoencoder model to detect unknown attacks or zero-day attacks. GWO algorithm eliminates redundant features in the network traffic dataset without losing significant information needed to do the accurate detection. Random Forest model is applied to classify known patterns of attack with categorized data, and the autoencoder is trained to learn the normal work of the IoMT traffic and identify anomalies without the necessity of labelled attack samples. On the whole, the hybrid framework advances the reliability and effectiveness of intrusion detection on the IoMT setting through the fusion of the feature optimization and machine learning and anomaly detection algorithms. The suggested method can assist in enhancing the security of healthcare networks and assist in safer disposition of IoMT systems.
Keywords: Internet of Medical Things, Intrusion Detection System, Grey Wolf Optimizer, Hybrid Machine Learning and Deep Learning, Autoencoder, Anomaly Detection, Zero-Day Attacks, healthcare security, Feature Selection, random forest.
How to cite this article: Thenmozhi T, Makesh L, Gopal KS, Dharsan SS, Vasanth M., A Hybrid Machine Learning and Deep Learning Framework for Intrusion Detection in IoMT Using Grey Wolf Optimizer...Int J Drug Deliv Technol. 2026; 16(11s): 745-768; DOI: 10.25258/ijddt.16.11s.76
Source of support: Nil.
Conflict of interest: None