Intelligent screening method for critical indexes for predicting fatal gastrointestinal rebleeding in hospital

Abstract

The invention provides an intelligent screening method for critical indexes of predicting fatal gastrointestinal rebleeding in a hospital, and the method comprises the following steps: extracting index data of a patient suffering from gastrointestinal hemorrhage from a database according to an inclusion exclusion standard, and calculating an outcome identifier; performing data cleaning on the index data to process abnormal values, and removing redundant indexes and indexes of which missing values are serious and cannot be analyzed; discretizing the index data according to a discretization standard set according to a clinical actual situation; taking the discretized index data as a condition attribute, and taking an outcome identifier as a decision attribute; inputting the condition attributes and the decision attributes into a screening algorithm model, calculating fitness, iteratively repeating until the fitness is not increased any more and the indexes are not reduced any more, and obtaining screened key index data. The method can efficiently and stably calculate the key index for indicating that the patient suffering from gastrointestinal hemorrhage suffers from hospital fatal rebleeding, and assists a doctor in making a decision.

Description

technical field [0001] The invention relates to the technical field of evolutionary calculation, in particular to an intelligent screening method for key indicators for predicting fatal gastrointestinal rebleeding in a hospital. Background technique [0002] Gastrointestinal bleeding is a common symptom in emergency clinics, and its etiology is complex. It can be caused by diseases of the digestive tract itself, such as gastric ulcer, acute gastric mucosal lesion, vascular lesion, tumor, or mechanical injury, as well as by adjacent organ lesions and systemic disease. Disease involving the digestive tract. When patients with gastrointestinal bleeding receive treatment in medical institutions, one of the important goals is to prevent fatal rebleeding in the hospital, because once a patient has an occult fatal rebleeding of the gastrointestinal tract or a short-term uncontrollable hemorrhage, and the resulting suffocation or blood loss Sexual shock is often the main cause of i...

AI Technical Summary

Problems solved by technology

[0014] For this reason, the purpose of the present invention is to propose an intelligent screening method for predicting key indicators of fatal gastrointestinal rebleeding in the hospital, effectively using time series data, and solving the problem that the existing logistic regression method cannot be based on the basic The problem with indicators accurately predicting their outcomes

Images