A Smart Coherent Security Model (SCSM) using Intelligent Optimization and Ensemble Deep Learning Mechanisms for Healthcare-IoT Networks
Subbarayan Venkatesan, M. Ramakrishnan · 2024
Artificial Intelligence (AI) and the Internet of Things (IoT) have spearheaded the digital revolution in contemporary healthcare. Security concerns need to be taken into account from the outset of any digital transition. The security of patients is jeopardized by healthcare data breaches, which are highly sensitive. Particularly in IoT networks where devices that are connected are susceptible to exploited, since cyberattacks may have potentially fatal outcomes. A ground-breaking security approach called the Smart Coherent Security Model (SCSM) is intended to protect small healthcare IoT systems' privacy and security against contemporary hostile threats. During the data pretreatment and cleaning procedures, outlier removal and standardization are completed, which contributes to the classifier's increased efficiency. In addition, the innovative Circulator System based Optimization (CirSO) technique is used to more efficiently select the most desired attributes from the provided data. In addition, the Deep Ensemble Prediction Model (DEPL), which has a high system accuracy and performance rate, is employed to distinguish between normal and attacking events. To evaluate performance, this study makes use of the most recent and widely used cyber-attack datasets, including CIC-IDS 2017, NSL-KDD, and CCIoT2023. Additionally, the outcomes demonstrate that, with a 99% accuracy increase and a 1.5% error rate decrease, the suggested SCSM model outperforms other existing techniques.