An Accurate Early Mortality Risk Prediction Model Based on ICU over MIMIC III using Machine Learning

Mortality risk prediction of ICU patients is a valuable and challenging task due to limited clinical data. Accurate mortality risk prediction can improve the utilization of resources. In this work, we explore the use of feature engineering followed by applying a machine learning approach for predicting mortality risk in ICU patients using the MIMIC III dataset. The results of this study show that our prediction accuracy is 99.99% using a neural network.