Enhancing IoT-based healthcare security with grey filter bayesian CNN and optimization algorithms

Abstract

The integration of Internet of Things (IoT) devices in healthcare systems has significantly enhanced patient monitoring and intervention capabilities, offering real-time data for improved decision-making. However, the widespread deployment of IoT devices has introduced critical security vulnerabilities, making healthcare systems vulnerable to cyberattacks that compromise patient safety and data privacy. This paper aims to address the security challenges in IoT-enabled healthcare by developing an advanced anomaly detection system that integrates the Grey Filter Bayesian Convolutional Neural Network (GFB-CNN), the Crow Search Algorithm (CSA), and the Binary Grey Wolf Optimizer (BGWO). The GFB-CNN is employed to filter out noise and uncertainties in the data, thereby improving the accuracy of anomaly detection. Meanwhile, CSA and BGWO optimize feature selection and tackle class imbalance, thereby enhancing computational efficiency. An experimental evaluation of the CIC IoT dataset demonstrates that the proposed method significantly outperforms existing security solutions, providing a robust and real-time anomaly detection framework. The results confirm that integrating GFB-CNN, CSA, and BGWO offers a promising approach for enhancing security in IoT-based healthcare systems, thereby ensuring better protection against cyber threats. This study contributes to the advancement of secure IoT healthcare solutions, paving the way for more reliable and scalable health system access in future healthcare applications.