Using Different Models of Machine Learning to Predict Attendance at Medical Appointments
Luiz Henrique Salazar 1, Anita Fernandes 1 2 * , Rudimar Dazzi 2, Nuno Garcia 3 4, Valderi R. Q. Leithardt 2 4 5
More Detail
1 University of the Itajai Valley, Specialization Course in Big Data, BRAZIL
2 Master in Applied Computing, University of the Itajaí Valley, Santa Catarina, BRAZIL
3 Telecommunications Institute, IT Branch, Covilha, PORTUGAL
4 Department of Informatics, University of Beira do Interior, Covilha, PORTUGAL
5 COPELABS, Lusophone University of Humanities and Technologies, Lisboa, PORTUGAL
* Corresponding Author

Abstract

Outpatient absenteeism is a recurring problem worldwide and in Brazil, it is a chronic problem. The number of appointments and exams scheduled and not performed, due to the non-attendance of patients, reaches significantly high rates and can be seen in all regions in the country and in different types of care and medical specialties. This practice generates waste of resources, disorganizes the offer of services, and limits the guarantee of care at different levels of assistance. In addition, it causes a series of dissatisfactions from users of the health system who really need and have not yet been able to access consultations and exams. This imbalance causes the misuse of the offer, an increase in the queue and waiting time, as well as a financial loss since it is paid for by a professional who is idle due to the absence of patients. It is necessary to understand the profile of these missing patients and to try to discover the reasons that lead this person to be absent, in order to predict a future absence for consultation. Thus, this work presents the study of different machine learning models in order to help predict whether or not the patient will attend the scheduled appointment. As a result, an end-to-end machine learning process was developed, considering exploratory data analysis, pre-processing, creation of machine learning models, analysis of results, and deployment of the most appropriate model in a web application. As a result, it was found that the Decision Tree algorithm represents an interesting choice as a final model for use in future observations.

License

This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Article Type: Research Article

https://doi.org/10.29333/jisem/8430

J INFORM SYSTEMS ENG, 2020 - Volume 5 Issue 4, Article No: em0122

Publication date: 30 Jul 2020

Article Views: 346

Article Downloads: 257

Open Access References How to cite this article