University of Twente Student Theses

Login

Detection of Health Problems based on Historical Medical Records

Matei, Denisa Alexandra (2019) Detection of Health Problems based on Historical Medical Records.

[img] PDF
451kB
Abstract:Demand for community home care services continues to increase due to demographic changes. Unfortunately, the home care organizations can not cope with this boost of people in need of home care due to a significant shortage of employees. The resulted heavy workload pressuring the professional care givers may have negative impact on the quality of the provided care. For instance, professional care givers postpone updating the medical records of the patients due to a lack of time. As a solution for speeding up the process, this study proposes a model capable of detecting health-problems based on the historical medical records of the patients and their personal profile (age and gender). Relationships between health-problems, as well as relationships between health-problems and personal profile were identified and integrated in a classification-machine-learning-based model. The obtained model is able to predict 31 different health-problems with an overall accuracy of 89.96\%, the accuracy ranging from 99.3\% to 67.2\% per problem.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:30 exact sciences in general, 44 medicine, 50 technical science in general
Programme:Business & IT BSc (56066)
Link to this item:http://purl.utwente.nl/essays/79150
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page