Secure Cloud Storage for Medical IoT Data using Adaptive Neuro-Fuzzy Inference System

Abstract

Medical Internet of Things (MIoT) plays a vital role in improving people’s health, protection, and treatment. Rather than going to the hospital, it is possible to remotely track patients’ health-related criteria and move them to medical data centers using cloud storage. There is also an exponential rise in the amount of data handled by MIoT devices. This means higher disclosure of confidential data; consequently, data protection and privacy obtained from MIoT devices are major unresolved concerns. As classification systems are growing rapidly, the need to apply machine learning algorithms to large-scale industrial data is also expanding. In this paper, we classify medical data into disease-free individuals from affected ones. To securely store this information in the cloud, we introduce a novel Adaptive Neuro-Fuzzy Inference System (ANFIS) for data protection to improve and determine the degree of security like breaches, data integrity, etc. Results indicate that with the 6.2506e−06 training error and 0.3429 testing error, ANFIS performs extremely well in improving the degree of security parameters for a secure path to the cloud.