Case complexity prediction method and system

Abstract

The invention provides a case complexity method and system. The method includes the following steps that: the discussion disagreement degree of each historical case is calculated; the feature data ofeach historical case are obtained; a regression model is established according to the feature data and discussion disagreement degrees of the historical cases; the discussion disagreement degree of atarget case is predicted according to the established regression model and the feature data of the target case; a treatment plan recommendation model is established according to the feature data of the historical cases and corresponding treatment plans; the treatment plan recommendation model is utilized to provide the recommendation result of the target case, and the uncertainty of the recommendation result is calculated; the discussion disagreement degree of the target case and the uncertainty of the recommendation result are linearly combined, so that the comprehensive index of the complexity of the target case is determined. With the case complexity method and system of the invention adopted, the historical voting data and historical case data of multidisciplinary medical decision-making teams can be fully utilized; and the complexity index of the case can be defined from a plurality of dimensions, namely the discussion disagreement degree and the uncertainty of the recommendationresult.

Description

technical field [0001] The invention relates to the field of medical decision-making and machine learning, in particular to a method and system for predicting case complexity by integrating discussion divergence degree and recommendation uncertainty. Background technique [0002] As medical disciplines become more and more refined, treatment plans based solely on the experience of a single doctor will have certain risks. Therefore, the establishment of a multidisciplinary medical decision-making team can ensure the rationality of the decision to a certain extent, but multidisciplinary discussions are also necessary. It affects the efficiency of the decision-making process and causes a waste of medical resources. In order to improve the efficiency of multidisciplinary medical decision-making, a lot of research work focuses on the research on the treatment plan recommendation system, and some research work focuses on optimizing the decision-making process. However, most of th...

AI Technical Summary

Problems solved by technology

[0003] The simplest way to consider case complexity is based on rules, for example, according to patient age, number of lesions, etc. The defect of this method is that it cannot make full use of historical data information

In addition, there is relatively little research on defining case complexity based on historical behavioral data from multidisciplinary teams and recommendation results from recommendation models.

Images