Operation blood transfusion risk prediction model and construction method of network calculator thereof

Abstract

The invention discloses a surgical blood transfusion risk prediction model and a construction method of a network calculator thereof, and the method comprises the steps: calculating the correlation between each clinical index and a blood transfusion risk through a single-factor logistic regression model, screening out the clinical indexes related to the blood transfusion risk, and carrying out the multivariate logistic regression calculation; screening out independent prediction factors to construct a blood transfusion prediction model; generating a network calculator according to a weight coefficient score formula of each prediction factor in the prediction model; and collecting a key information data set of the surgical patient, and inputting the key information data set into the network calculator to obtain the predicted blood transfusion probability. The blood transfusion prediction model constructed by the invention can predict the probability of perioperative blood transfusion of an operative patient before an operation, a network calculator is generated according to the model, the model can be applied to medical electronic cases, and automatic electronic decisions are provided for preoperative plans and blood management.

Description

technical field [0001] The invention relates to the technical field of medical informatics, in particular to a surgical blood transfusion risk prediction model and a method for constructing a network calculator. Background technique [0002] Blood transfusion is an important means of saving lives in surgical operations. Preoperative identification of patients at high risk of blood transfusion is very important to rationally arrange limited blood resources. Adverse reactions to blood transfusion caused by blood transfusion, such as fever, allergic reaction, hemolysis reaction, circulatory system overload related to blood transfusion, acute lung injury related to blood transfusion and blood-borne infection, etc., endanger the health and even life of patients. Early identification of high-risk transfusion patients and early patient blood management are important for the prevention of adverse transfusion reactions. Blood group identification, blood group antibody screening, and...

AI Technical Summary

Problems solved by technology

[0003] At present, the blood transfusion prediction model is in the exploratory stage, and most of the blood transfusion prediction models are used for single-disease blood transfusion prediction, which will be inconvenient in clinical use

At present, there is no general-purpose blood transfusion prediction model applied in various surgical patients, and there is no design of network calculator or APP based on this type of model

Images