Cardiovascular Disease Prediction using Relief-SVM Approach

cardiovascular diseases (CVD) present an important healthiness mission. The early prediction of CVDs is paramount for efficient prevention and intervention efforts. In our research, we present a novel method for identifying the most relevant clinical features by optimizing features using the Relief determinant technique. In order to forecast CVD, an additional technique based on the Support Vector Machine (SVM) is used to the chosen relevant features. This proposed approach effectively determines the state health of a patient: healthy or affected by heart-related conditions. The results emphasize that the proposed hybrid method (Relief-SVM) significantly achieve the highest evaluation metrics; Accuracy ‘Ac=99.86%%’, Precision ‘Pr=97.36%’, Sensitivity ‘SE=98.52%’ and Specificity SP=95.84’. Compared to previous works, this procedure exhibits substantial promise in ensuring precise predictions regarding heart diseases.