Patient Survival Prediction with LightGBM achieved 0.69 F1-score and 0.89 AUC.
Dataset from kaggle. In this data set, there are over 90k rows and 85 columns, which means it has more than 90k patients featured by 84 attributes(the remaining one is the label ‘hospital_death’) including gender, height, weight and diagnosis.
The predictors of in-hospital mortality for admitted patients remain poorly characterized. We aimed to develop and validate a prediction model for all-cause in-hospital mortality among admitted patients.
In this dataset, there are various factors given, which are involved when a patient is hospitalized. On the basis of these factors, predict whether the patient will survive or not.