Kawasaki disease risk assessment model building method and system based on ensemble learning

Abstract

The invention provides a Kawasaki disease risk assessment model building method and system based on ensemble learning; the method comprises the following steps: extracting valid samples from a sampledata set for building a model and model assessment; selecting at least 10 feature items that comply with field medical auxiliary diagnosis applications from a feature set that forms the sample data; respectively using a random forest, Boosting, a linear model and a nerve network algorithm to build a Kawasaki disease prevalence risk prediction basic model and a classification valve domain t; usinga Naive Bayes algorithm for integration, and evaluating the Kawasaki disease risks according to a comparison result between two types of posterior probabilities. The method can effectively solve the problems that most classifiers are over-fitting, can adopt advantages and avoid weaknesses, thus providing the more accurate assessment model.

Description

Technical field [0001] This application relates to the technical field of medical evaluation, and specifically to a construction method and construction system of a Kawasaki disease risk assessment model based on integrated learning. Background technique [0002] Kawasaki disease (Kawasaki disease, KD), also known as mucocutaneous lymph node syndrome, is an acute febrile eruptive pediatric disease with systemic vasculitis as the main lesion. Among them, the coronary artery is the most vulnerable part. The most important complication is coronary artery disease. If it cannot be diagnosed and treated in time, it will cause serious damage to the cardiovascular system. It has become one of the most common causes of acquired heart disease in children and ischemic heart in adulthood. Risk factors for disease. Therefore, early diagnosis, early treatment, and reduction of cardiovascular complications have important clinical significance. [0003] According to the diagnostic criteria for ...

AI Technical Summary

Problems solved by technology

According to the above diagnostic criteria, a patient needs to have a fever for ≥ 5 days for the diagnosis of Kawasaki disease and wait for clinical symptoms to appear, which may easily cause the child to miss the best time for treatment

At the same time, the clinical symptoms of Kawasaki disease are complex and diverse, and the clinical symptoms are not obvious in the early stage of the disease, which is prone to misdiagnosis and missed diagnosis, which increases the difficulty of diagnosis of Kawasaki disease in children to a certain extent.

Images