IoT and 6G-Enabled Dual-Target Voting Ensemble Framework for Predictive Healthcare Network Interference and Resource Utilization Management

Abstract

The healthcare sector in India is experiencing rapidly increasing demand while operating under limited medical resources. With national healthcare expenditure at approximately 3% of Gross Domestic Product (GDP), the system faces shortages in hospital beds, medical professionals, and essential equipment. The widespread adoption of Internet of Things (IoT) devices for health and lifestyle monitoring has transformed modern healthcare by enabling continuous, real-time data collection. Additionally, sixth-generation (6G) cellular networks support IoT-based healthcare services through enhanced ultra-reliable low-latency communication (eURLLC), ensuring timely and reliable transmission of critical health data. This research proposes an intelligent and scalable framework to optimize healthcare resource management using advanced machine learning techniques. The system utilizes real-time data from IoT-enabled devices within a 6G environment to perform dual-task prediction: classification of resource utilization levels and regression-based estimation of efficiency scores. A Classification and Regression Tree (CART) approach is applied to enhance structured decision-making and model interpretability. The framework integrates multiple machine learning models, including Random Forest (RF) and Support Vector Machine (SVM) combined as the Hybrid Kernel Forest Network (HFKN), along with Gradient Boosting (GB), Logistic Regression (LR), and KNearest Neighbors (KNN) forming the TriVector Intelligent Fusion Model (TVIFM). Extreme Gradient Boosting (XGBoost) is proposed as the primary model due to its strong boosting capability and generalization performance. Implemented using a Flask-based architecture, the system supports data upload, training, evaluation, real-time prediction, and visualization of metrics such as accuracy, precision, recall, F1-score, and R², improving resource allocation and operational efficiency.