Patient hospitalization duration prediction method and device, electronic equipment and storage medium

Abstract

The invention discloses a patient hospitalization duration prediction method and device, electronic equipment and a storage medium. The method comprises the steps of constructing an ordered multi-classification prediction model by utilizing cascade connection of multiple binary base learners; training each base learner by using the training data set until each base learner meets the performance index requirement, and obtaining a trained prediction model; and selecting a to-be-predicted sample according to preset prediction features and inputting the to-be-predicted sample into the trained prediction model to obtain a prediction result. According to the method, a plurality of binary classification base learners are cascaded and connected in series to construct an ordered multi-classification prediction model, a sequence progressive relationship among classes in an ordered multi-classification outcome variable is reserved, the ordered classes are not assumed to be equal-ratio relationships, the method is more consistent with real data characteristics, and by splitting a data set layer by layer, two types of data in the data set used for training of each layer of base learners are relatively balanced, the problem of data imbalance among multiple types is effectively solved, and the accuracy of a prediction result is improved.

Description

technical field [0001] The present application relates to the technical field of data processing, and in particular to a method, device, electronic device and storage medium for predicting a patient's length of stay in hospital. Background technique [0002] The length of hospitalization is a key indicator for evaluating the utilization efficiency of medical resources. The intelligent length of hospitalization prediction system can assist clinicians in identifying patients with higher disease risks and provide timely medical intervention, thereby improving the prognosis of patients in hospital; it can also assist doctors in making reasonable arrangements With limited medical resources, the utilization efficiency of medical resources can be maximized; it can also provide patients and their families with information about the length of hospitalization at the early stage of admission, so that patients and their families can learn more about their illness and possible hospitaliza...

AI Technical Summary

Problems solved by technology

At present, ordered multi-classification problems are generally solved based on numerical prediction models or unordered multi-classification prediction models: numerical prediction models assume that multiple categories of outcome variables follow a proportional correlation, while ordered multi-classification data in the real world Multiple categories often do not follow a strict proportional correlation; the unordered multi-category prediction model directly ignores the progressive relationship between the various categories of ordered multi-category outcome variables, and the performance of the prediction model is often limited

At the same time, when there is a data imbalance between the categories of the ordered multi-category outcome variable, the unordered multi-category prediction model will produce a large prediction error

Images