Improving intrusion detection in cloud-based healthcare using neural network

• Published in: Biomedical signal processing and control Vol. 83; p. 104680
• Main Author: Patel, Sagarkumar K.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.05.2023
• Subjects: Arrhythmia classification; Cloud; Intrusion detection; Neural network; Optimization; Arrhythmia classification; Neural network; Intrusion detection; Optimization; Cloud
• ISSN: 1746-8094
• DOI: 10.1016/j.bspc.2023.104680

•Proposed secured and effective classification of Arrhythmia using the ECG signals.•The proposed Hybrid tempest optimization algorithm is designed by integrating the characteristic behavior of the particle search agent and the collaborative search agent helps to obtain a more accurate global solution.•The intrusion detection is employed using the neural network (NN), which is trained using the proposed Hybrid tempest optimization algorithm.•The Arrhythmia classification is employed using the NN, which is performed using the ECG signal retrieved from the cloud. This research introduces an efficient smart security solution with the effective healthcare technique for Arrhythmia classification using the cloud computing scenario. Initially, the ECG of the patient in both the static and dynamic environment is collected using sensors and is stored in cloud, in which the clustering of the network and cluster head (CH) selection is devised using the Hybrid tempest optimization algorithm. In order to ensure the security for the collected healthcare data, the intrusion detection model is devised and implemented in the cloud storage for secure data access using the Neural network (NN), where the NN is trained using the Hybrid tempest optimization algorithm. The access is granted to the legitimate users for acquiring the data for diagnosis. The hybrid tempest-NN method is analyzed through accuracy, sensitivity, and specificity and obtained the maximal values of 95.72%, 96.78%, and 95.29% respectively.