Multi-objective semi-supervised clustering to identify health service patterns for injured patients

TL;DR

This paper introduces a semi-supervised clustering method to early identify injured patients' health service patterns, enabling prediction of outcomes and costs based on early post-injury data.

Contribution

It presents a novel multi-objective semi-supervised clustering approach tailored for early health service pattern recognition in injured patients.

Findings

Effective early grouping of patients based on health service use

Identification of patients with undesirable outcomes early

Predictive insights into medication costs

Abstract

This study develops a pattern recognition method that identifies patterns based on their similarity and their association with the outcome of interest. The practical purpose of developing this pattern recognition method is to group patients, who are injured in transport accidents, in the early stages post-injury. This grouping is based on distinctive patterns in health service use within the first week post-injury. The groups also provide predictive information towards the total cost of medication process. As a result, the group of patients who have undesirable outcomes are identified as early as possible based health service use patterns.