Predicting Illness and Type of Treatment from Digital Health Records
Name
Markus Lippus
Abstract
The rising costs of healthcare and decreasing size of the working population is a dire problem in most of the developed world. While it is inevitable that new methods are costly, it is possible to reduce avoidable expenses by better planning and prevention. Most hospitals keep digital records of everything that happens to a patient during their treatment and in Estonia all medical bills are also presented to the National Health Insurance Fund (NHIF) for reimbursement. In this work the data from NHIF is used to build a model that as the first step uncovers the different clinical pathways followed for the treatment of patients with an illness. As a second step the model is used to predict the number of patients that will be provided the uncovered treatments in the future. The output of such a model could be a valuable asset for planning resource allocation and preventative health care.
Graduation Thesis language
English
Graduation Thesis type
Master - Computer Science
Supervisor(s)
Sven Laur, Anna Leontjeva
Defence year
2017