Measurement: Sensors (Feb 2024)
Deep learning enabled blockchain based electronic heathcare data attack detection for smart health systems
Abstract
The idea of networked personal medical devices is a component of contemporary Smart Health Systems (SHS). These gadgets offer remote observing and the exchange of wellbeing information, which enormously further develop the patient's personal satisfaction while at the same time reducing treatment expenses for both the patient and the medical care suppliers (telemedicine). For individuals' wellbeing, a cutting edge individual wellbeing record framework is fundamental. Here, there are still difficulties with information mix from different EHRs, information interoperability, and guaranteeing that admittance to information is totally under the power of the patient. Some security issues are caused by the Network. To settle these issues, we propose a novel profound learning-based framework that circuits state of the art decentralized innovations like IPFS and blockchain with wellbeing information interoperability principles and advances like FHIR's APIs. In this review, we show that correspondence between private clinical gadgets is as a matter of fact powerless to different cyber attacks. We show how an outer assailant could involve man-in-the-center, replay, bogus information infusion, and refusal of-administration assaults to block delicate wellbeing information stream by capturing the correspondence of the individual clinical gadget. We likewise suggest an Interruption Recognition Framework (IDS), GAN, to additional screen traffic on private clinical gear and spot attacks against them. Our extensive investigation shows that GAN, with an F1-score of 98 % and an accuracy of 98.7 %, can successfully and accurately recognise numerous assaults on personal medical equipment.
