Answers are grounded in this paper and cite their sources.
This paper introduces a novel Di-CVD Tri-Layer CX Classifier, an IoT-integrated and machine learning (ML)-driven framework, to predict the individual and joint risk of diabetes (DB) and heart disease (HD). The proposed model comprises three phases: secure IoT-based data collection using Enhanced BGV encryption with Dynamic Distributed Hashing (DDH); a feature extraction (FE) phase leveraging (IGO) Information Gain Ratio and disease-specific ranking and a three-step classifier—Cm-Ro (FS) feature selection, hierarchical XGBoost classification, and synergistic prioritized risk scoring. By integrating multi-attribute features, rule-free optimization, and enhanced interoperability, the model addresses critical challenges such as heterogeneous data formats, poor feature relevance, and low interoperability in previous studies. When compared to conventional classifiers such as SVM and standard XGBoost, experimental evaluation on the NHANES dataset shows improved performance in terms of accuracy (ACC), recall (R), precision (P), and F1-score. The outcomes validate the framework’s effectiveness in early, secure, and individualized risk prediction, offering substantial support for timely interventions and enhanced patient care.
No takes yet. Share an insight, caveat, or question.
Thumilvannan et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: