Towards Secure Healthcare Systems: An Evolutionary Optimized Ensemble Learning Model for Threat Detection in IoHT

Ensuring security is vital in the Internet of Health Things (IoHT) as it protects personal patient data, maintains device integrity, and minimizes danger to patients. Malware can infiltrate medical devices, create chaos with health services, and risk lives. Enforcing critical security is a method of safeguarding the integrity of data, ensuring the right diagnoses and treatments are provided, and maintaining a patient's privacy, promoting patients' confidence in their health systems. This study makes contributions to enhancing cybersecurity in IoHT. This study constructs and implements a machine learning (ML)-enabled platform capable of classification and detection of various cyberattacks with a formidable defense tool for IoHT systems. In addition, the paper focuses on Differential Evolution optimized Random Forest (RF+DE) since it shows enhancement in accuracy, precision, recall and F1-score compared to traditional methods. An average score of 99% is noticed in all performance metrics. The paper presents an evaluation based on a benchmark IoHT continuation process, as well as finding the ability to identify detailed patterns in attack methodology. Lastly, the study presents valuable insights regarding the utilization of the proposed solution and supports further deployment options in healthcare systems for identifying possible threats and interventions in real-time to mitigate and ensure integrity of the IoHT ecosystem.