A Hierarchical Intrusion Detection System for Secured Healthcare System using Hybrid Machine Learning and Encryption System

Security is a major concern in the healthcare industry, especially regarding the legal and ethical prospects of patients’ medical data. Ensuring that medical data is transmitted securely is critical to providing timely and accurate insight into a patient’s treatment. This paper proposes a highly secure healthcare data transmission methodology covering two security tasks: an intrusion detection system (IDS) based on optimized machine learning (ML) and healthcare data encryption based on chaotic elliptic curve cryptography (C-ECC). Initially, in the data collection phase, the input data collected from Kaggle online resources includes medical data and node information. The Hybrid Extreme Learning Machine (HELM) model for IDS is introduced, which is a combination of support vector machine (SVM) and random forest (RF). The HELM model has been improved by using a salt-enhanced zebra optimization algorithm (SE-ZOA), which checks nodes to check whether the collected data is malicious throughout transmission. Any malicious nodes detected are removed to ensure that only benign nodes are used to render medical data. Medical data is further secured using a chaos-based encryption scheme with ECC and Distance Aware Secretary Bird Optimization (DA-SBO), which encrypts data through optimal key generation. The encrypted data is then placed on the blockchain to ensure secure contact with the cloud server. The same optimum key is utilized for the decryption procedure.