A Secured IoT Healthcare Architecture-Based Access Control System Using Deep Learning

Much medical data is processed and stored in the current healthcare system based on traditional administration. Internet of Things (IoT)-enabled healthcare has evolved with remarkable data processing and significant data storage capabilities due to its integration and progressive development. The result is an advanced medical monitoring system that can use a wearable gadget to remotely check a patient's condition because of developments in the Industrial Internet of Things (IIoT). Unauthorized users and attackers may exploit the server where the wearable IoT module stores its collected data, compromising privacy. This paper proposes an IoT-based deep learning-based data analytics solution to overcome this security vulnerability. Users provide data, which is gathered, with sensitive data being kept apart. The health-related data is analyzed without revealing personal information about the consumers using an artificial neural network (ANN). As a result, a healthcare system's user attributes are utilized to operate a secure access control module. The accuracy, F1 measure, recall, and precision of the proposed ANN classifier are achieved at 96.67, 94.82, 93.73, and 92.98%. More excellent performance is obtained as the training set size grows. The system works better with data augmentation added than without it. The existing systems used for comparison are SVM, BPNN, and CNN algorithms. To assess the effectiveness of the suggested approach, Matlab 2013A is utilized to create a deep learning algorithm.