Machine Learning Based Systolic and Diastolic Pressure Prediction Model

Hypertension is the medical term for elevated blood pressure. Blood pressure has a significant impact on hypertension and cardiovascular disease. A sphygmomanometer that resembles a cuff is frequently used to measure a patient's blood pressure. Other newly developed signal-based cuffless methods include photoplethysmography (PPG) and electrocardiogram (ECG). Due to the need to connect several sensors, this method is both expensive and challenging. A totally new machine learning-based model for forecasting systolic and diastolic blood pressure is suggested in this study. It takes advantage of the Kaggle website's dataset of 70,000 blood pressure readings.To increase model accuracy for the four different machine learning algorithms utilized, various percentages of training, validation, and testing were computed (KNN, logistic regression, decision trees, and random forest). Accuracy, recall, precision, and F1 scores are calculated for the aforementioned algorithms. The accuracy and robustness of the model were strengthened by the results, which were validated. Of the four algorithms, Random Forest has the best accuracy and F1 score.