The Relation Between Respiratory & Acute Coronary Syndrome Using Data Mining Techniques
The British University in Dubai (BUiD)
In healthcare world, data science is one of the most important sciences that helps in predicting diseases, and despite the availability of medical data from laboratory tests, most medical institutions in middle east region still do not benefit from these data in diseases analysis and prediction. The purpose of this study is to diagnose acute coronary syndrome using the widely available respiratory diagnosing tools and laboratory test results using data mining classification techniques (Decision tree, Gradient boosted tree, Neural Network, and Naïve Bias). In this study I’ve split one dataset of patients who have attended to emergency departments in Abu Dhabi hospitals to two datasets (Respiratory and Cardiac), then applied the data mining algorithms on each dataset and one time on the original dataset. This study found that respiratory features such as sO2(Oxygen saturation), O2Hb (Oxyhemoglobin), pH and HHb (Deoxyhemoglobin) values can predict if the patient is an acute coronary syndrome or there is a possibility be affected by this disease.
data science, data mining techniques, acute coronary syndrome