Intrusion Detection in Healthcare using Sand-Cat Optimization based Long-Short Term Memory

The Internet of Medical Things (IoMT) has become an attractive environment for cybercriminals due to its rapid development. These devices have finite computational capabilities which establish minimal power absorption. Security is a significant problem in IoMT systems due to the services being accessed via the Internet by a variety of users therefore, the patient’s health information needs to be kept confidential, secure, and accurate. In this research, the Sand-Cat optimization based Long Short-Term Memory (SCO-LSTM) is proposed for intrusion detection to secure the entire network in the healthcare system using deep learning. Initially, the data is obtained by the CIC-IDS2017 dataset and then min-max normalization is performed to normalize the acquired data. The SCO approach is employed for feature selection which examines appropriate features in the healthcare system. Finally, the LSTM classification is performed to identify and classify intrusion detection accurately and effectively in the healthcare system. When compared to the existing methods, the proposed SCO-LSTM achieves a better accuracy of 99.48% respectively.