E-healthcare application cyber security analysis using quantum machine learning in malicious user detection

In the medical field, it is crucial to manage visual and auditory data generated by Internet of Things (IoT) devices. Cloud servers are often used to manage the massive amounts of data generated by these IoT devices. Current improvements in electronic and communication technology have greatly impacted the e-healthcare sector owing to the effective exchange of patient data. IoMTs, or the Internet of Medical Things, are a relatively recent development in the field of remote health monitoring. They are used in patient-centric systems for the transmission and tracking of patient data. Authentication and anomaly detection are two areas where modern medical systems make extensive use of encryption, biometrics, and machine learning (ML) technology. This study suggests a new method for assessing the cyber security of e-healthcare apps; one that makes use of quantum machine learning. Users of e-healthcare applications have been tracked and analysed to identify risky behaviours. A deep variational adversarial encoder network and a fuzzy Gaussian quantile neural network classify the characteristics of observed user activity data, leading to the identification of malicious users and an increase in network security. Recreation aftereffects of the proposed engineering show vigor with regards to proficient execution, including prescient misfortune = 7%, learning rate = goldilocks (0.5), record advancement = 23%, transmission influence = − 18 dBm, jitter = 32 ms, delay = 90 ms, throughput = 170 bytes, obligation cycle and conveyance = 10%, and dynamic serverless reactions. Proposed technique attained Random accuracy of 98%, F-1 Score of 75%, mean average Precision (mAP) of 65%, Specificity of 66%, kappa Co-efficient of 69%.