STAKI2B2
Project Name: Semantic Text Analysis for Quality-controlled Extraction of Clinical Phenotype Information in Healthcare-Integrated Biobanking
Project Category: local research project
Head of Project:
- PD Dr. Dr. Michael Kiehntopf (Jena University Hospital)
Duration and Status: 1.1.2016 – 30.6.2020, data use completed
Data Types Provided: free text documentation, e.g. from doctors' letters
Abstract:
In the STAKI2B2 project, valid phenotype data and other relevant comparative information will be extracted automatically from clinical documents using automatic speech processing methods to improve the application of new markers from biomaterials in clinical processes. For this purpose, a text analytics pipeline will be established, which automatically determines relevant medical entities and relationships between these entities from clinical documents using semi-monitored machine learning methods.
Funding: Deutsche Forschungsgemeinschaft (DFG), Project No. 315098900