Bachelor/Master Thesis: Unsupervised FHIR Questionnaire Template Checker
The COVID-19 pandemic has revealed the necessity and potential of digital systems in helping researchers tracking symptoms and other biomarkers of the population. The NUM-COMPASS project was funded to create a unified framework that helps researchers with developing apps for pandemic studies. The scientific discussion within the different university hospitals in the NUM network yielded the idea of a template checker for Fast Healthcare Interoperability Resources (FHIR)-based questionnaires. The FHIR standard offers researchers a unified way to design questions for use in bigger study contexts involving questionnaires from different users around Germany. To add new questions to the framework, they have to conform to a template. To check such conformity and assist researchers during questionnaire design, the thesis aims at developing an automated pipeline based on deep learning and Natural Language Processing (NLP). Such an algorithm would learn answer types that conform to the FHIR-based framework and recommend it to the researcher during study design. The algorithm should also recommend complementary questions that conform to the framework based on specified question types.
Create an automated question template checker for FHIR-based pandemic questionnaires.
|Project type||Master thesis / Bachelor thesis|
|Language||English and/or German|
|Period||Summer term 21|
|Presence time||Working from remote|
|Useful knowledge||Natural Language Processing, Deep Learning, Pandemic Apps, FHIR|
|Work distribution||100% programming of NLP algorithms and FHIR data|
|Registration||E-Mail to firstname.lastname@example.org|
Literature will be provided in the first meeting and the candidate is encouraged to further research relevant papers for this work.
Thesis report and final presentation.