Forschungsthemen
[BA] Frontend Design and Implementation for Natural Language Mapping
Motivation: During clinical studies, adverse events are mainly reported as free text in digital
patient files. For this process, different applications are used and there is no standard to
follow. Usually, a central institution is dedicated to extracting and collecting important terms
from these clinical reports and converting them into corresponding standard terms, MedDRA
in this case. Because huge parts of this process are done manually, the mapping is expensive,
time-consuming and requires specialized knowledge.
Aim of the work: Other work of the AI4Hematology research group developed different ap-
proaches (data pipelines) to map such kind of terms to provided taxonomies / ontologies.
These backend solutions should become accessible for medical users, hence a flexible fron-
tend software design has to be developed with the aim to include more prospective data
pipelines and allow for different user scenarios. The required frontend should mainly consider
the implementation of user interaction, the integration of different backend, the visualization
of results and the embedding into other frontends, as well as software qualities such as ex-
tensibility, performance, robustness and transferability. The AI4Hematology will provide data
sets and backend solutions for evaluation purposes.
Betreuer: Karsten Wendt