Answers are grounded in this paper and cite their sources.
Data security is an important parameter to be maintained in the healthcare sector as there is an abundant amount of sensitive data such as the information about patient’s health, medication provided, and the medical history of the patients. Data security measures need to be taken to be implemented to secure the availability, integrity, and confidentiality of the private data of the patients and hospitals. Securing electronic health records (EHRs), data related to health, and patient information is the objective of data security. High complexity, privacy concerns, and slow adoption of security standards are the issues faced by conventional security approaches in the healthcare sector. Hence, we propose a Modified Extreme Learning Machine with an Ensemble support Vector machine (MELSVM) for data acquisition and intrusion detection in healthcare. In this technique, the support vector machine (SVM) is used to detect the patterns related to security breaches, and the extreme learning machine (ELM) is used to categorize the intrusions according to their types. Then, to enhance the performance of the proposed technique by merging the performance of SVM and ELM, an ensemble approach is used. Further, to identify the maximum local/global optima or the entire problem space, the Multi-Modal Bacterial Foraging Optimizer (M2BFO) is used. By utilizing the combined advantages of the MELM and ESVM, the enhancement of the security of MELSVM is done. Utilizing the blockchain technology, the data obtained is safely stored in the cloud. By employing smart contracts, the data exchange between system administrators and clients is ensured. The validation of the proposed technique is done using the datasets UNSW-NB15 and CIDDS-001. In the evaluations done, our proposed technique outperforms the existing techniques by attaining 99.52% accuracy, 98.52% precision, 99.52% recall, and 99.52% F1 score. After performing all comprehensive analyses, the proposed model demonstrates its effectiveness in terms of using security and privacy measures, intrusion detection and classification measures. The proposed method provided promising solutions for the loud security that helps the model for extending this innovation to future investigations.
No takes yet. Share an insight, caveat, or question.
Johnson et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: