Process Mining for Healthcare Management (Preprint)

Author:

Elvas Luis B.ORCID,Saeed Raeisi,Helgheim BeritORCID,Martins Ana LúciaORCID,Ferreira João CORCID

Abstract

BACKGROUND

The digital age, with ICT, IoT, big data, has opened new opportunities for improving the delivery of healthcare services, in which data mining approach can help improve the hospital management process by providing a big picture identifying process efficiencies

OBJECTIVE

This is a multidisciplinary study where a data mining approach was applied to hospital data to the improvement management process. From this data, we can provide a big picture of hospital processes and check process efficiency. Since we get data from the emergency department we check process, patients waiting time, physicians’ efficiency and length of stay (LOS).

METHODS

A data analytic investigation of the impact of patients’ average waiting time and LOS was performed to compare the efficiency of different groups of the physicians. In this research-based on CRISP-DM method by using an emergency department’s (general surgery unit) real-life data from 2015 to 2017, through the One-Way Analysis of Variance (ANOVA) method.

RESULTS

From the developed approach using hospital data, it was found that the waiting time and average LOS belonging to experienced physicians (who visit more patients) are longer than the less experienced physicians (who visit less patients). Using the Two-Way ANOVA method, the emerging and very urgent patients (Red and Orange triage color) who were visited by experienced physicians have longer average LOS than the same level patients who visited by less experienced physicians. On the other hand, not urgent patients (Blue triage color) who were visited by high experienced physicians have shorter average LOS than the same patients who were visited by less experienced physicians. This study was performed using two scenarios of physicians' grouping, including Pareto grouping (80-20) and Frequency grouping ("Very High", "High", "Medium", "Low" and "Very Low").

CONCLUSIONS

We show that data visualization tools for healthcare analysts to help them make better decisions, by the big picture identified. Big data offers a lot of potential for improving healthcare administration and taking the sector to the next level, identifying working problems, reduce costs.

Publisher

JMIR Publications Inc.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3