Key result
Hybrid RF-ANN model predicts heart disease with ~90% accuracy, outperforming baseline HRFLM.
Why the study?
Cardiovascular disease is the leading cause of death worldwide, motivating non-invasive machine learning screening using routinely collected clinical parameters.
Does a Hybrid RF-ANN model improve heart disease prediction accuracy compared to HRFLM?
Does a Hybrid RF-ANN model improve heart disease prediction accuracy compared to HRFLM?
A novel Hybrid RF-ANN machine learning model significantly improves the accuracy and ROC-AUC of heart disease prediction using routine clinical parameters compared to existing baseline models.
Improved hybrid RF-ANN may enhance CVD prediction; leaves open prospective validation before any clinical use.
The leading cause of death worldwide is cardiovascular disease, which is motivating screening using non-invasive machine learning on clinical parameters collected routinely. The present work paper reviews eight studies representative and heart disease prediction most relevant, highlights the major research gaps they leave open, and replicates a well-known hybrid baseline the Hybrid Random Forest with Linear Model (HRFLM) proposed by Mohan et al., and finally presents an improved Hybrid RF-ANN. The proposed model rescaling clinical input feature by their importance score derived from Random-Forest and feeding those features to two hidden layers NN. The Hybrid RF-ANN, whose code is published, attained an accuracy of 90.16% in the study, a significant 9.83 percentage point improvement over the accuracy of HRFLM, while the Hybrid RF-ANN achieved the highest ROC-AUC of all ten studied models (0.977), which is an enhancement of 0.0787 in ROC-AUC against HRFLM.
No takes yet. Share an insight, caveat, or question.
Jyoti et al. (2026) studied Heart disease. Hybrid Random Forest-Artificial Neural Network (RF-ANN) model vs. Hybrid Random Forest with Linear Model (HRFLM) was evaluated on Model accuracy for heart disease prediction (9.83 percentage point improvement). The Hybrid RF-ANN model achieved an accuracy of 90.16% and a ROC-AUC of 0.977 for heart disease prediction, outperforming the baseline HRFLM model by 9.83 percentage points in accuracy.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: