Establishment method of severe spinal cord injury prognosis prediction model

Abstract

The invention discloses a method for establishing a severe spinal cord injury prognosis prediction model, and the method is characterized by comprising the following steps: extracting clinical data of cases of patients diagnosed as spinal cord injury: 1) incorporating the following clinical characteristics; 2) preprocessing the clinical features: processing missing data through different filling methods according to the types of the clinical features; 3) incorporating an algorithm combination of a feature selection method and a machine learning classification algorithm, wherein the feature selection method is used for screening clinical features with significant prediction values, and the selected clinical features are used for training the machine learning classification algorithm; 4) selecting the algorithm combination with the maximum area AUC under the micro average curve from the prediction performance of the algorithm combination in the step 3) in the training data set, and integrating the algorithm combination by using a stacking method to obtain a prediction model. The invention has accurate and objective performance for predicting the prognosis of the patient with severe spinal cord injury.

Description

technical field [0001] The invention relates to a method for establishing a prediction model for the prognosis of severe spinal cord injury. Background technique [0002] Patients with spinal cord injury are often admitted to the intensive care unit (ICU) due to major trauma or serious complications, so their prognosis is of great concern to clinicians and patients' families. However, how to accurately predict the prognosis of severe spinal cord injury is a clinical problem. Clinically, doctors often judge the prognosis of patients based on experience to formulate a diagnosis and treatment plan. However, it is often impossible to give an objective and quantifiable probability of prognosis when explaining the patient's condition to the patient's family. Therefore, an accurate and objective system for predicting the prognosis of patients with severe spinal cord injury is needed to assist clinicians, and how to establish a prediction model for this system for predicting the p...

AI Technical Summary

Problems solved by technology

However, it is often impossible to give an objective and quantifiable prognostic probability when explaining the patient's condition to the patient's family.

Images