The integration of machine learning (ML) in medical diagnostics has significantly advanced the accuracy and efficiency of disease detection. However, conventional deep learning and deterministic models often face challenges such as high computational overhead and lack of interpretability, limiting their deployment in resource-constrained healthcare environments. This study presents a comprehensive analysis of randomized explainable machine learning models for efficient medical diagnosis. The proposed framework leverages ensemble-based classifiers—Random Forest, XGBoost Classifier, Stacking Classifier, and Voting Classifier—to balance predictive accuracy, computational efficiency, and interpretability. Randomization within these algorithms enhances generalization, reduces overfitting, and accelerates model training. Furthermore, explainability methods are incorporated to ensure transparent decision-making, aiding clinicians in understanding key diagnostic features influencing predictions. The comparative evaluation demonstrates that randomized ensemble classifiers achieve near state-of-the-art diagnostic performance while significantly lowering computational costs. This approach enables faster, reliable, and interpretable predictions suitable for time-sensitive medical applications, offering a promising pathway toward sustainable, explainable, and efficient healthcare AI systems.

Keywords: Randomized Machine Learning, Explainable AI, Medical Diagnosis, Random Forest, XGBoost, Ensemble Learning, Stacking Classifier, Voting Classifier, Computational Efficiency, Healthcare Analytics.