The invention discloses a
sepsis early prediction method based on
machine learning. The method comprises the following steps: firstly, extracting clinical data, including
demographic statistics (suchas
age and gender), vital sign variables (such as
heart rate and systolic pressure) and laboratory measurement indexes (such as
creatinine and
platelet count), of a patient within 24 hours after a patient enters an ICU by utilizing an
electronic medical record, preprocessing the data, inputting the preprocessed data into an improved deep forest
algorithm model for training, and outputting the
disease probability of the patient after training optimization. And meanwhile, the
algorithm model can also sort characteristic variables and output an early warning factor which has great influence on
early prediction of
sepsis. Finally, corresponding variables of the patient needing to be predicted are input into the trained model, so that
early prediction of
sepsis can be carried out on the patient. According to the invention, early prediction is carried out on sepsis based on a
machine learning method, doctors can be assisted in making clinical decisions, and the prediction accuracy is improved.