Predicting patient mortality

Doctors and medical practitioners frequently face life-or-death decisions based on a patient's medical condition upon arrival at the hospital. In this project, we used the MIMIC-III (Medical Information Mart for Intensive Care III) dataset to develop a predictive model that estimates the probability of patient mortality within a specific timeframe. This tool aims to aid medical professionals in risk stratification, thereby facilitating more informed clinical decision-making.

A predictive model was created using K-Nearest Neighbor and SVM algorithms and achieved an accuracy rate of 0.88.