Process Mining/ Deep Learning Model to Predict Mortality in Coronary Artery Disease Patients

EasyChair Preprint 13772

Abstract

Patients with Coronary Artery Disease (CAD) are at high risk of death. CAD is the third leading cause of mortality worldwide. However, there is a lack of research concerning CAD patient mortality prediction; thus, more accurate prediction modeling is needed to predict the mortality of patients diagnosed with CAD. This paper demonstrates performance improvements in predicting the mortality of CAD patients. The proposed framework is a modification of the work used for the prediction of 30-day readmission for ICU patients with heart failure. Our framework demonstrates better performance with an Area Under the ROC Curve (AUC) score of 0.871 for the Neural Network (NN) model compared to traditional baseline machine learning models that we developed. Our framework uses the medical history of patients, the time related to the variables, and patients’ demographic information for prediction. This framework has the potential to be used by medical teams to make more accurate decisions for treatment and care for patients with CAD, increasing their life expectancy.

Keyphrases: Mortality Prediction, Process Mining, coronary artery disease

@booklet{EasyChair:13772,
  author    = {Negin Ashrafi and Armin Abdollahi and Greg Placencia and Maryam Pishgar},
  title     = {Process Mining/ Deep Learning Model to Predict Mortality in Coronary Artery Disease Patients},
  howpublished = {EasyChair Preprint 13772},
  year      = {EasyChair, 2024}}