Key points are not available for this paper at this time.
Real-time health monitoring technologies, such as fog computing (FC) and Internet of Things (IoT) sensors, have brought in a new era of healthcare. Healthcare services have accepted IoT with great ease, in accordance with Industry 4.0 goals which helped by strong fundamental components such as FC. This research uses cutting-edge technologies like fog computing and IoT to present a novel framework for meeting the changing demands of the healthcare monitoring system. With the use of machine learning, this work seeks to improve crucial communication characteristics and further the research by identifying the best security method based on the occupations of the patients. For optimisation, the framework makes use of the Firefly (FFLY) and Grey Wolf Optimisation (GWO) algorithms. Furthermore, Elliptic Curve Cryptography (ECC) and Rivest-Shamir-Adleman (RSA) encryption techniques are taken into consideration to improve data security in healthcare simulations. This security selection is powered by machine learning-based classification algorithms, where the primary goal is to maintain security while preserving energy resources. In summary, the amalgamation of the RSA security algorithm with the Firefly (FFLY) and Grey Wolf Optimization (GWO) algorithms yielded substantial enhancements in several critical Quality of Service (QoS) attributes. The proposed improved healthcare system obtains significant results in terms of QoS parameters and security selection using machine learning classification methods, surpassing the basic findings. Significantly, reliability experienced notable improvements of 17.32% and 22.69%, convergence achieved optimizations of 9.64% and 16.02%, and interoperability demonstrated improvements of 6.61% and 8.71%. Notably, when it comes to energy consumption, a vital consideration for resource-limited sensor configurations, FFLY and GWO with RSA showcased optimizations of 11.03% and 13.16%. The choice of a security algorithm is determined through machine learning techniques, where the Support Vector Machine (SVM) algorithm outperformed alternative methods. In the evaluation of classification techniques, SVM and Random Forest (RF) exhibited accuracy and F-Measure values of 0.999 and 0.993, respectively. These results underscore SVM’s effectiveness in managing medical data.
Lalit et al. (Mon,) studied this question.