Attack detection and mitigation scheme through novel authentication model enabled optimized neural network in smart healthcare. Issue 1 (2nd January 2023)

• Record Type: Journal Article
• Authors: Patel, Sagarkumar K.
• Is Part Of: Computer methods in biomechanics and biomedical engineering. Volume 26:Issue 1(2023)
• Page Start: 38
• Page End: 64
• Publication Date: 2023-01-02
• Subjects: Internet-of-things -- electrocardiography -- authentication -- healthcare -- hybrid optimization; Biomechanics -- Data processing -- Periodicals; Biomedical engineering -- Periodicals; Biomechanics -- Periodicals; Biomedical Engineering -- methods -- Periodicals; Computing Methodologies -- Periodicals; 612.7
• DOI: 10.1080/10255842.2022.2045585
• Languages: English
• ISSNs: 1025-5842

Abstract

Abstract: The Internet of Things (IoT) have become an important part of human in day-to-day life as it permits accesses and manages data flawlessly, the security of data in cloud storage is of great concern in healthcare applications. This paper proposes a secure authentication to protect the sensible data related to healthcare through IoT network. Initially, the Electrocardiography (ECG) signal from the patients are stored in the cloud in encrypted form, where the proposed modified Elliptic-curve Diffie–Hellman (ECDH) encryption is applied to ensure secure access to the stored data to be used for the analysis of arrhythmia. The obtained data for the arrhythmia diagnosis is subjected to classify the attack using the neural network (NN). The weights of the NN are tuned using the proposed hybrid tempest brain optimization algorithm, which integrates the characteristic features of collaborative search agents and the hybrid search agents. The proposed method obtained the values of 95%, 7150, and 111 of detection accuracy, number of genuine users, and information loss of the respectively, which shows the superiority of the proposed method in attack detection and mitigation.