Intelligent anomaly detection in medical IoT systems using Coral Reefs Optimization and Aquila enhanced CatBoost

The rapid growth of medical Internet of Things (MIoT) systems is changing modern healthcare. These systems allow real-time monitoring of patients and they support intelligent data-driven diagnostics. They help doctors and caregivers detect health issues early in life. However, the heavy use of connected medical devices also creates serious concerns. The security of patient data becomes difficult. The detection of anomalies in such large networks also remains a critical challenge. This paper presents an anomaly detection framework designed for MIoT environments. The framework uses Coral Reef Optimization (CRO) to select the most important features. CRO reduces data redundancy and improves the learning process. For classification, the model applies CatBoost. The hyperparameters of CatBoost are tuned with Aquila Optimization (AO). AO improves the model by searching for better parameter values and strengthens generalization. The results show a strong performance. The proposed model reaches higher accuracy precision recall and F1 score than traditional methods. These methods include support vector machine XGBoost gradient boosting logistic regression and random forest. In addition, the model shows lower complexity than several state-of-the-art approaches. This makes the framework efficient and suitable for practical use in real-world healthcare settings.