Parametric Influence Of Intrusion Detection System In Healthcare Sector Using Deep Neural [LSTM] Network


  • Monika Khatkar, Kaushal Kumar, Brijesh Kumar, Pankaj Agarwal



With the recent advancements of computer connectivity and the number of computer-related applications, the challenge of achieving cyber security has increased. It also necessitates a strong defence system against a variety of cyber-attacks in healthcare, Manufacturing, Transportation and Government sector as well. Furthermore, the presence of trespassers with the intent to launch various attacks within the healthcare network should not be underestimated. The most focused framework intrusion detection system protects the healthcare infrastructure from potential intrusions by inspecting network traffic to ensure its confidentiality, integrity, and availability. ML techniques also offer the opportunity to detect and monitor network security issues in healthcare sectors caused by the emergence of programmable features. Recently, AI (ML) and deep learning-based interruption detection frameworks (IDS) have been used as potential solutions for proficiently recognizing network interruptions. In this research, a deep learning-based LSTM model is used on the NSL-KDD dataset. Dataset used in the study is collected from the website of New Brunswick University. Popular classification algorithms, Random Tree, Decision Tree, SVM, KNN, Decision Tree, ANN (Artificial Neural Network), and Deep Neural Network are used to detect the interferences. To justify the superiority of LSTM, performance metrics, precision, recall, f1-score, and accuracy are evaluated with Denial-of-Service cyber-attacks. The experimental result shows LSTM model outperforms in the case of an intrusion detection system.

This study first describes IDS and then provides a categorization based on the notable ML and DL methods used in the designing of Network-Based IDS (NIDS) systems.



— Updated on 2023-01-22




How to Cite

Parametric Influence Of Intrusion Detection System In Healthcare Sector Using Deep Neural [LSTM] Network. (2023). Journal of Pharmaceutical Negative Results, 685-692.