Heart disease has become a major global health concern that is affecting millions of people worldwide. The situation is particularly critical in developing countries where the access to medical facilities is limited. This barrier to health care leads to increased fatalities from heart disease. Early diagnosis of cardiovascular conditions can be lifesaving. However, personal medicalgrade equipment can be expensive and not easily accessible for people living in these areas. It is important to expand the same level of medical care to these communities at an affordable price. Our research aims to investigate the performance of a machine learning model on a low-cost embedded system. This study will evaluate the accuracy, run time, and overall performance of the model in diagnosing cardiovascular diseases. The results will help us determine the feasibility of using machine learning models for classifying cardiovascular disease in low-cost embedded systems. A selected machine learning model has been trained, modified, and compiled into the embedded system. The model returns the classification results based on preprocessed input data. Multiple metrics are collected to measure the performance of the model and the embedded system. The preliminary results are promising with accuracy levels similar to the original model. If these results hold up in multiple trials, it is expected that the machine learning model for classifying cardiovascular diseases on the embedded system will be practical and useful in extending affordable medical care to developing countries.