• norsk
    • English
  • English 
    • norsk
    • English
  • Login
View Item 
  •   Home
  • Faculty of Mathematics and Natural Sciences
  • Department of Informatics
  • Master theses
  • View Item
  •   Home
  • Faculty of Mathematics and Natural Sciences
  • Department of Informatics
  • Master theses
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Early detection of medical deterioration of patients with diabetes by using machine learning

Hammerbeck, Andreas
Master thesis
Thumbnail
View/Open
master thesis (1.410Mb)
URI
https://hdl.handle.net/1956/23070
Date
2020-06-27
Metadata
Show full item record
Collections
  • Master theses [40]
Abstract
Diabetes is a growing healthcare problem in the world, which affects over 400 million adults. In collaboration with Haukeland University Hospital, we look at medical records from real patients diagnosed with diabetes. The study uses machine learning to predict if a given patient has a high risk of experiencing medical deterioration. Further, the thesis goes through the data cleaning necessary to provide such predictions. The first approach managed to identify 79\% of high-risk patients. If the model classifies a patient to have a high risk of mortality, the model had an accuracy of 18\%. In the second approach, we removed the last four weeks before mortality happens, and the model was able to identify 49\% of the patients with high risk. If the model classifies a patient to have a high risk of mortality, the model had an accuracy of 12\%.
Publisher
The University of Bergen
Copyright
Copyright the Author. All rights reserved

Contact Us | Send Feedback

Privacy policy
DSpace software copyright © 2002-2019  DuraSpace

Service from  Unit
 

 

Browse

ArchiveCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsDocument TypesJournalsThis CollectionBy Issue DateAuthorsTitlesSubjectsDocument TypesJournals

My Account

Login

Statistics

View Usage Statistics

Contact Us | Send Feedback

Privacy policy
DSpace software copyright © 2002-2019  DuraSpace

Service from  Unit