AI driven IOMT security framework for advanced malware and ransomware detection in SDN

An AI driven double-tier detection system, means this is a security framework that works in two layers to increase accuracy and reliability of threat identification, this is presented in this research. The system designs to tackle rising Software Defined Network (SDN) security threats especially in healthcare environment to the risk of malware, and ransomware attacks, is powered by Deep Learning (DL) and Machine Learning (ML) technologies to work together, as a static and dynamic scanning methods. Using this, execution traces and the Application Programming Interface (API) call sequences are integrated with Generative Adversarial Networks (GANs) and autoencoder based predictive analysis to improve the threat detection. This is a continuous training system that ensures medical data’s protection while maintaining compliance with health regulations and creates a secure operational environment. The double tier eXtreme Gradient Boosting (XGBoost) performance reaches 99.60% accuracy with an F1 score of 0.9966 and outperforming the benchmark model at 98.80% accuracy and F1 score 0.988. Further, a two-layer Light Gradient Boosting Machine (LightGBM) model achieves 99.32% accuracy with an F1 score of 0.9934, in contrast to which at 98.96% accuracy and an F1 score of 0.9895, beats its benchmark counterpart. Furthermore, the Deep Neural Network (DNN) system with bi layered also achieves 99.13% accuracy and F1 score of 0.9915. This study presents novel countermeasures for these SDN and healthcare security threats that warrant a breakthrough in the field of cybersecurity. The presented security model is agile and strong which offer real time network assessment and continue to detect emerging threats thus improving the protection of critical healthcare infrastructure.