Trust‐Driven ECC With Random Ensemble and Bat Optimization for Advanced Healthcare Cybersecurity

ABSTRACT In healthcare, cybersecurity breaches have increased significantly, with recent reports indicating that over 78% of hospitals experienced at least one cyberattack in the past year, exposing millions of patient records. Existing cybersecurity models often pose drawbacks such as high computational cost, poor energy efficiency, and limited capability in detecting sophisticated anomalies in IoT‐enabled healthcare environments. To tackle these issues, this study introduces a Trust‐based Elliptical Curve Cryptography with Random Ensemble model that combines an energy‐efficient anomaly detection mechanism based on the Pearson correlation coefficient with lightweight cryptographic security. The method includes Pearson Correlation Coefficient‐based feature selection (reducing dimensionality by up to 35%), and class imbalance is tackled by Synthetic Minority Over‐sampling Technique‐based data augmentation. A Random Forest ensemble model integrating Gaussian Naïve Bayes and Decision Tree to identify anomalies is enhanced. Elliptical Curve Cryptography ensures efficient encryption and decryption, while a Trust Agreement Model and Body Area Network logic verify secure communication between healthcare servers and patients. In addition, the Bat Optimization Algorithm optimizes authentication parameters, and computational overhead is reduced. The Trust‐based Elliptical Curve Cryptography with Random Ensemble model is validated on three datasets, including BoT‐IoT, UNSW‐NB15, and CSE‐CIC‐IDS2018, and demonstrates superior performance, attaining 98.89% accuracy, 98.60% precision, and 98.45% F1‐score outperforming existing methods. The results demonstrate that the proposed method provides a more reliable, energy‐efficient, and scalable cybersecurity solution for modern healthcare IoT systems.