Key result
Random Forest algorithm predicts hypertension best, achieving ~92% accuracy among tested machine learning models.
Why the study?
Little or no research on hypertension prediction using local clinical data from Nigeria has been done despite abundant data availability.
Do machine learning algorithms utilizing clinical data accurately predict the likelihood of hypertension?
Do machine learning algorithms utilizing clinical data accurately predict the likelihood of hypertension?
Machine learning algorithms, particularly Random Forest, can accurately predict the likelihood of hypertension using clinical data, offering a potential tool for early detection and preventive interventions.
Answers are grounded in this paper and cite their sources.
May support ML hypertension prediction in research; leaves open prospective validation before clinical use.
Machine learning classifiers usage in medicine has been on a rise over the years due to increase in available clinical data. Hypertension, one of the cardiovascular heart diseases is a global health concern with several risk factors where early detection could prevent related heart diseases complications. With abundant clinical data piling within our health facilities here in Nigeria, little or non-research on hypertension prediction using local dataset has been done. This research work is aimed at developing models for predicting individuals with likelihood of hypertension even without necessary going through medical procedures. A sample of 294 clinical data from Specialist Hospital Gashua was used. Random Forest, Decision tree, Artificial neural network, Support vector machine, Logistic regression and Naïve bayes algorithms were used for classification. To minimised issues of over-fitting due to the number of available data, ADASYN balancing technique was used for data balancing. Using Root Means Square Error (RMSE), Accuracy, Recall, F1-score and ROC values as parameter matrix for evaluating the models, our results showed that machine learning can be utilised in predicting hypertension diseases for quick preventive measures. Furthermore, this research work further illustrate that Random Forest has proven to be the most efficient and best algorithms for hypertension prediction having performance supremacy over all others five algorithms with 92% prediction accuracy, 100% Precision, Recall values of 91%, 91% F1-score value, RMSE of 29% and AUC value of 97%. Thereby, stressing the significant of balancing dataset for enhanced model accuracy (and all other metrics).
No takes yet. Share an insight, caveat, or question.
Egwa et al. (2026) studied Hypertension (n=294). Machine learning algorithms (Random Forest) vs. Other supervised learning algorithms was evaluated on Hypertension prediction performance (Accuracy, Precision, Recall, F1-score, RMSE, AUC). The Random Forest algorithm was the most efficient for predicting hypertension, achieving 92% accuracy, 100% precision, 91% recall, and an AUC of 97%.
Synapse has enriched 10 related papers. Consider them for comparative context: