Detecting Anomalous Patterns in IoT Healthcare Systems through ContextAware Intelligence

Abstract

The rapid growth of smart healthcare systems and interconnected environments has increased the demand for secure, scalable, and intelligent data analysis mechanisms. With the emergence of distributed architectures and cloud-based services, achieving accurate prediction and efficient anomaly detection has become a major challenge. This research presents a cloud-driven intelligent system that integrates machine learning (ML) techniques with a client–server architecture for secure and real-time prediction. The primary problem addressed in this work is the difficulty in obtaining high prediction accuracy from large and imbalanced datasets in distributed environments. Many traditional systems rely on standalone models and manual analysis, resulting in limited accuracy, poor scalability, and ineffective handling of class imbalance. These limitations reduce their suitability for real-time and large-scale applications. There is a strong need for an automated, robust, and scalable system capable of performing efficient data preprocessing, handling imbalance, supporting multiple ML models, and enabling secure remote access. The system should also ensure seamless communication between distributed components. To address these challenges, the proposed system adopts a client–server architecture consisting of Laptop 1 (LP1) as the server and Laptop 2 (LP2) as the client. LP1 integrates a desktop graphical user interface with a cloud-based prediction engine using the Flask framework. The system performs preprocessing using label encoding (LE) and applies the Synthetic Minority Oversampling Technique (SMOTE) for class balancing. Multiple models including Ridge Classifier (RC), Quadratic Discriminant Analysis (QDA), and Perceptron (PC) are implemented. Furthermore, a Locally Deep Ensemble Classifier (LDEC) combines Histogram Gradient Boosting Classifier (HGB) and Light Gradient Boosting Machine Classifier (LGBM) using soft voting (SV). Secure authentication uses Redis. The system improves accuracy, scalability, reliability, and security for modern applications.