Cardiovascular diseases (CVDs) are still the leading cause of death in the worldwide. Traditional machine learning models often have difficulty in determine how to capture the complex links between disease risk factors and disease occurrence. This article discusses a hybrid machine learning approach for cardiovascular risk prediction (HMLCRP) to address this problem. This approach combines logistic regression (LR), support vector machines (SVMs) and neural networks (NNs) to make predictions more correct and reliable. The proposed model looks at important coronary heart sickness risk factors, including excessive blood pressure, a record of coronary heart disorder within the family, pressure, age, sex, levels of cholesterol, body mass index (BMI) and poor dwelling choices. The hybrid technique makes use of the nice functions of LR for clean understanding, SVM for dealing with large amounts of facts and NNs for finding developments. By integrating these models together, the HMLCRP makes positive that type is correct and that danger predictions are accurate. In this study, benchmark datasets used, which include the cardio statistics set, heart ailment dataset and Framingham heart examination dataset, are used to train and test the version. Popular parameter measures, such as accuracy, precision, recall and the F1-score, are used to determine overall performance. The results of the experiments indicate that the HMLCRP is better at predicting effects than individual models. The suggested combination model is a major step forward in personalised healthcare because it allows proactive risk management and early intervention methods to stop CVD.