An explainable framework for hypertension prediction using uncertainty-based Monte Carlo dropout and firefly optimization

Abstract Hypertension is a major health challenge that causes premature deaths in the world. Although there has been progress in machine learning and deep learning to predict hypertension, the current approaches still have a number of critical limitations such as extreme class imbalance, suboptimal classification error, high annotation, and poor hyperparameter optimization tools. Most investigations make use of traditional gradient-based or manually-tuned optimization methods, which in most cases lead to local optimal solutions, slow training behavior, inadequate search space exploration, and higher computational costs. These weaknesses associated with optimization reduce the robustness of the models, their stability, and their generalization capabilities in different clinical environments. To mitigate these challenges, an advanced synthetic sampling strategy was employed to address class imbalance by generating representative minority class samples based on distribution-aware weighting mechanisms. This approach reduces bias toward the majority class, improves the structural representation of minority instances within the feature space, and enhances overall classification performance and generalization capability. Building on this balanced dataset, we proposed three novel architectures: Model 1: HighCeptNet, Model 2: HighCeptALNet, and Model 3: FireCeptNet. Model 1 integrates Highway and Inception networks to jointly exploit deep residual learning and multi-scale feature extraction, improving discriminative capacity. Model 2 extends this by incorporating uncertainty-based active learning with Monte Carlo Dropout active learning, enabling selective sampling of the most informative unlabeled instances, which substantially reduces labeling costs while retaining high predictive performance. Model 3 further enhances stability and convergence by integrating firefly optimization for adaptive hyperparameter tuning, thus addressing training instability and improving robustness. Simulation results demonstrate that all three models significantly outperform existing and baseline methods across multiple evaluation metrics. Specifically, Model 1 achieves relative improvements of 3.30% in accuracy, 3.33% in F1-score, 3.30% in recall, 3.26% in precision, 10.13% in Matthews Correlation Coefficient (MCC), 11.39% in Cohen’s kappa, and 28.95% in Receiver Operating Characteristic – Area Under the Curve (ROC-AUC). Similarly, Model 2 achieves gains of 6.59% in accuracy, 7.78% in F1-score, 7.69% in recall, 3.26% in precision, 16.46% in MCC, 17.72% in Cohen’s kappa, and 28.95% in ROC-AUC. Moreover, Model 3 achieves the highest performance improvements with 7.69% in accuracy, 7.78% in F1-score, 7.69% in recall, 5.43% in precision, 21.52% in MCC, 20.25% in Cohen’s kappa, and 28.95% in ROC-AUC, consistently outperforming all the baselines and the other proposed architectures under diverse optimization settings. To ensure reliable performance estimation, 10-fold cross validation is implemented, complemented by confidence interval estimation for all metrics and statistical significance validation using the Mann–Whitney U test. Finally, Shapley additive attributing of features globally and instance-level interpretable attributing of models locally is used to create model interpretability that gives transparency and clinical-level trust. The findings validate the fact that the presented framework provides a sound, statistically stable, and interpretable solution to the task of early hypertension prediction and fills the gap between predictive accuracy, annotation efficiency, and clinical interpretability.