The privacy of medical data and resource restrictions in the Internet of Things (IoT) nodes prohibit medical users from utilizing disease detection (DD) services offered by the health cloud in the mobile healthcare network (MHN). Also, health clouds may need the DD procedures to be private. Therefore, the essential requirements for MHN DD services are (i) performing accurate and fast DD without jeopardizing the privacy of health clouds and medical users and (ii) reducing the computational and transmission overhead (energy-consumption) of the green IoT devices while performing privacy-preserving DD. The outsourced privacy-preserving DD is available in the literature based on popular tree-based machine learning schemes such as a random forest. However, these schemes utilize energy-hungry public-key encryption schemes in IoT nodes at medical users for privacy preservation. This work proposes an energy-efficient, fully homomorphic modified Rivest scheme (FHMRS) for the proposed privacy-preserving random forest classification (PRFC). A secure integer comparison protocol is also developed for reducing processing time and energy consumption for users while performing outsourced PRFC. The implementation results and security analysis show that the proposed schemes guarantee better energy efficiency for MHN green IoT devices without compromising privacy than the existing tree-based schemes.

Adopting mobile healthcare network (MHN) services such as disease detection is fraught with concerns about the security and privacy of the entities involved and the resource restrictions at the Internet of Things (IoT) nodes. Hence, the essential requirements for disease detection services are to (i) produce accurate and fast disease detection without jeopardizing the privacy of health clouds and medical users and (ii) reduce the computational and transmission overhead (energy consumption) of the IoT devices while maintaining the privacy. For privacy preservation of widely used neural network– (NN) based disease detection, existing literature suggests either computationally heavy public key fully homomorphic encryption (FHE), or secure multiparty computation, with a large number of interactions. Hence, the existing privacy-preserving NN schemes are energy consuming and not suitable for resource-constrained IoT nodes in MHN. This work proposes a lightweight, fully homomorphic, symmetric key FHE scheme (SkFhe) to address the issues involved in implementing privacy-preserving NN. Based on SkFhe, widely used non-linear activation functions ReLU and Leaky ReLU are implemented over the encrypted domain. Furthermore, based on the proposed privacy-preserving linear transformation and non-linear activation functions, an energy-efficient, accurate, and privacy-preserving NN is proposed. The proposed scheme guarantees privacy preservation of the health cloud’s NN model and medical user’s data. The experimental analysis demonstrates that the proposed solution dramatically reduces the overhead in communication and computation at the user side compared to the existing schemes. Moreover, the improved energy efficiency at the user is accomplished with reduced diagnosis time without sacrificing classification accuracy.

This study proposes a secure and privacy‐preserving protocol for outsourcing health data processing operations during the emergency in the mobile healthcare network. The proposed protocol provides a practical solution to utilise smartphone resources at both remote and nearby for processing the overwhelming personal health information (PHI) of a user in healthcare emergency opportunistically and securely. The patients with symptoms matching with those of the user in an emergency are considered as opportunities to minimise the privacy disclosure of the user. Opportunities at both remote and nearby are exploited with the help of a base station in the 4G network. Moreover, novel and efficient outsourced privacy access control schemes are developed to minimise the power drain of the user in an emergency without compromising his privacy. The outsourced privacy access control is facilitated through the design of innovative schemes for outsourced attribute‐based access mechanism and an outsourced privacy‐preserving scalar product computation. Detailed performance evaluations through implementations on Raspberry Pi 3B + board and simulations using NS3 network simulator and Scyther tool confirm the efficiency of the proposed protocol in providing highly reliable PHI processing and transmissions with reasonably low delay and energy consumption while maintaining user privacy.