Cardiovascular Machine Learning: Interpreting Maximum Heart Rate Values from Patient Biometrics Using Linear Regression and Random Forest Models

Leveraging machine learning, this study analyzes and interprets patient biometrics to predict heart rates and identify disease trends. Currently, machine learning is limited in its abilities to create models based on specific types of patient biometrics, especially when it comes to heart disease related topics like cholesterol monitoring and the resting rate of blood pressure. Our goal is to create a quantitative approach to this, allowing clinicians to have an easier understanding of varied heart rate values and their causes and effects. We intend to create a machine learning algorithm, which we will train and feed with a dataset from an open-source network of patient studies, allowing us to create models for visualization and analysis. The models used in this experiment will be linear regression and random forest models, paving the path for prediction of specific variables based on other variables. More specifically, in this scenario we will be predicting the maximum heart rate through other variables that are given. The goal is to find an effective and efficient route using ML to identify direct causes and possible treatment plans. The success/results of this model will allow cardiologists to have prominent results when diagnosing heart-related conditions, allowing patients to receive the care they need quicker and from a more individualized perspective.