Severe sequential organ failure scoring method and system based on machine learning

Abstract

The invention discloses a machine learning-based critical sequential organ failure scoring method and system. The method comprises the following steps of: screening 25 important influence indexes from clinical indexes, and carrying out related data extraction by utilizing the public and severe database; establishing a combined model based on a decision tree model for the extracted data for model training; performing index screening by using an algorithm, and finally retaining 13 important indexes of the eight organ systems through algorithm branch selection in combination with clinical experience knowledge; dividing an index value interval, and establishing an evaluation score system; and finally, verifying the effectiveness of the method, and proving the effectiveness of the new scoring method. The key breakthrough of the method is that a 13-index eight-system scoring method is established qualitatively and quantitatively, and the effectiveness of the method reaches 0.82 and is far higher than that of SOFA scoring.

Description

technical field [0001] The invention belongs to the technical field of medical scoring, and in particular relates to a machine learning-based scoring method and system for severe sequential organ failure. Background technique [0002] Critical illness severity score is a method for quantitatively evaluating the severity of critical illness by weighting or assigning values ​​based on some important symptoms, signs, and physiological parameters of the disease. Quantitatively assessing the severity of disease and predicting the risk of disease or patient death can help clinicians make decisions about the treatment of individual patients. In addition, it can re-evaluate treatment measures, resource utilization, quality control, ICU turnover and utilization, medical costs, quality of life after recovery, disability status, and leadership decision-making. [0003] Currently commonly used critical illness severity scores include Acute Physiological Function and Chronic Health Scor...

AI Technical Summary

Problems solved by technology

More importantly, due to the lack of infection indicators and demographic indicators, it cannot effectively evaluate the infection system and personal characteristics, and some evaluation indicators can no longer meet the needs of current clinical applications.

Images