Layer clustering system for medical history file arrangement

Abstract

The invention discloses a layer clustering system for medical history file arrangement. The layer clustering system comprises a medical history introducing module, a vector processing module, an included angle cosine calculation module and a clustering analysis module, wherein the medical history introducing module is used for normalizing all variables in an introduced medical history file; the vector processing module is used for converting the variable types and proportions of variables to be analyzed and storing space vector coordinates of all individuals into a space vector library; the included angle cosine calculation module is used for calculating a similarity coefficient of any two space vectors and arranging the similarity coefficients into a similarity matrix; the clustering analysis module conducts clustering analysis according to the similarity coefficients, a clustering graph is made according to a clustering analysis result and a user selects a classification layer or a similarity coefficient threshold value to store all the medical history files into different classifications according to the clustering graph respectively. In this way, the layer clustering system is applied to arrangement and classification of the medical history file, and therefore a grouping method is provided for or the research direction is determined for further development of clinical scientific research.

Description

technical field [0001] The invention relates to the field of medical technology, in particular to a hierarchical clustering system for sorting out medical records. Background technique [0002] Since different medical records have both individuality and commonality, similar medical records can be divided into multiple categories in the same batch of medical records according to the different characteristics of the medical records. Through the cluster analysis of medical records, a large number of electronic medical records can be divided into several categories, so as to discover the common characteristics of a certain group, which has considerable guiding significance in the discovery of clinical problems and even in clinical practice. For example, for research on digestive tract diseases, cluster analysis of intestinal flora and eating habits of different groups of people can divide a group of people into several categories, thereby providing a grouping method for further ...

AI Technical Summary

Problems solved by technology

The commonly used methods of cluster analysis include hierarchical clustering and non-hierarchical clustering, but because these clustering methods have certain restrictions on variable types and are often used in univariate analysis, it is difficult to apply to the classification of medical records

Images