HO-CER: Hybrid-optimization-based convolutional ensemble random forest for data security in healthcare applications using blockchain technology

【Author】 Badri, Sahar

【Source】ELECTRONIC RESEARCH ARCHIVE

【影响因子】1.604

【Abstract】The Internet of Things (IoT) plays a vital role in the rapid progression of healthcare diligence. In recent years, IoT has become one of the most significant sources in the medical domain, since physical devices collect essential patient information to share real-time data with medical practitioners via various sensors. Meanwhile, numerous existing intrusion detection techniques failed to meet the security needs to safeguard the patient data collected. If an attack or intrusion cannot be identified at a particular time, immeasurable damage will be developed, which will fail the system. Utilizing innovative and new technologies, namely Blockchain, edge computing, and machine learning, provides a powerful security solution to preserve the medical data of various patients. This paper proposes a modified convolutional ensemble random forest-based hybrid particle swarm (MCERF-HPS) approach to guarantee healthcare data security with the advancement of blockchain technology. The proposed MCERF-HPS-based intrusion detection system identifies and categorizes attacks and regular traffic in blockchain-based edge systems. In immediate response to the identification, the gateway devices in the network layer block the attack traffic within seconds, with fewer computing and processing abilities. Applying the detection mechanism at the edge layer close to the attack source provides a quick detection response and minimizes the workload of clouds. The proposed MCERF-HPS approach's ability to detect an intrusion is tested using the BoT-IoT database. The analytic result illustrates that the proposed MCERF-HPS approach achieves an improved attack detection accuracy of about 98.7% compared to other methods.

【Keywords】blockchain; cybersecurity; modified convolutional neural network; ensemble random forest; hybrid particle swarm optimization; attack; intrusion detection

【发表时间】2023

【收录时间】2023-08-27

【文献类型】

【主题类别】

--