Cyber-attack detection in healthcare using cyber-physical system and machine learning techniques

Cyber-physical systems have been extensively utilized in healthcare domains to deliver high-quality patient treatment in multifaceted clinical scenarios. The medical device' heterogeneity involved in these systems (mobile devices and body sensor nodes) introduces enormous attack surfaces and therefore necessitates effective security solutions for these complex environments. Hence, in this study, the cognitive machine learning assisted Attack Detection Framework has been proposed to share healthcare data securely. The Healthcare Cyber-Physical Systems will be proficient in spreading the collected data to cloud storage. Machine learning models predict cyber-attack behavior, and processing this data can offer healthcare specialists decision support. This proposed approach is based on a patient-centric design that safeguards the information on a trusted device like the end-users mobile phones and end-user control data sharing access. Experimental results demonstrate that our suggested model achieves an attack prediction ratio of 96.5%, an accuracy ratio of 98.2%, an efficiency ratio of 97.8%, less delay of 21.3%, and a communication cost of 18.9% to other existing models.

Automated summary

Cyber-physical systems in healthcare utilize a cognitive machine learning assisted Attack Detection Framework to securely share healthcare data for decision support, achieving high accuracy and efficiency in predicting cyber-attack behavior.