Clinical assessment using an algorithm based on fuzzy C-means clustering

Abstract

The Fuzzy c-means (FCM) algorithms define a grouping criterion from a function, which seeks to minimize iteratively the function up to until an optimal fuzzy partition is obtained. In the execution of this algorithm each element to the clusters is related to others that belong in the same n-dimensional space, which means that an element can belong to more than one clusters. This proposal aims to define a fuzzy clustering algorithm which allows the patient classifications based on the clinical assessment of the medical staff. In this work 30 cases were studied using the Glasgow Coma Scale to measure the level of awareness for each one which were prioritized by triage Manchester method. After applying the FCM algorithm the data is separated data into two clusters, thus, verified the fuzzy grouping in patients with a degree of membership that specifies the level of prioritization.