Abstract
In the modern world, due to changing lifestyles, more people are infected by deadly diseases, especially heart disorders. An early prognosis is the only possible way of increasing survival rate. One of the key elements of this context is the healthcare digitalization that is carried out through Internet of Things (IoTs) and cloud computing. However, main issue in cloud with IoT is false diagnosis, which leads to cause a major impact on patient’s life. Moreover, data communication through cloud servers can be compromised by attackers due to security issues. Therefore, to overcome these issues, this paper proposes a novel secure e-healthcare system with dual objectives- accurate disease prediction and improved cloud security. A novel diagnosis approach ‘Hybrid Binary Particle Firefly Optimized Extreme Learning Machine classifier (HybBPF-ELM)’ to predict HD is designed, that recognize the subject with HD abnormality from the normal ones. The prediction accuracy and time efficiency of training process are improved by the adoption of a Hybrid Binary Particle Firefly Optimizer (HBPFO) through fine-tuning weight and bias of Extreme Learning Machine (ELM) classifier. Additionally, a new intelligent encryption and decryption framework ‘IEDF’ is introduced for cloud security. It combines four encryption algorithms, including Advanced Encryption Standard (AES), Data Encryption Standard (DES), Rivest- Shamir-Adleman (RSA), and Modified Blow Fish (MBF) to enhance cryptographic strength and key security. Along with this, an Automatic Sequence Encryption (ASC) for data block encryption is employed to ensure strong security and network performance. The integration of HD prediction module and cloud security framework within e-healthcare system assists healthcare experts in securely storing and transferring data for advanced diagnosis. The results show that our proposed system achieves an overall prediction accuracy of 99.36 % and less processing time for encryption and decryption of 127.55 s and 452.01 s at 2.2 GB file size respectively.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have