ShockSurv: A machine learning model to accurately predict 28-day mortality for septic shock patients in the intensive care unit

•Septic shock is a subtype of sepsis with a high mortality rate.•Septic shock patients’ mortality can be reduced by early prediction and interventions.•We develop a model to predict septic shock patients’ mortality based on clinical data.•The model has high prediction accuracy and can be used to assist clinical diagnosis.•This is the first research to predict septic shock patient’s mortality via clinic available data.