May 21, 2025
Systematically indexing clinical texts for research and AI: The ZMI at the GeMTeX plenary meeting
GEMTEX-Plenarmeeting
Around 25 employees from the MII project GeMTeX met last week for a plenary meeting in the Albertina Library at Leipzig University to discuss the current status of the annotation work and the future use of the GeMTeX text corpus. Among them were Dr. Markus Wolfien, our Data Science team leader, and Karolin Hofmann, part of the project coordination team at the ZMI.
The aim of GeMTeX is to prepare medical documents such as doctors' and discharge letters in such a way that they can be used for research and for the use of artificial intelligence in compliance with data protection regulations. The annotation of clinical documents is essential for this. To this end, medical students at six hospital sites - including the ZMI in Dresden - are marking content from these documents.
Guidelines for annotation continue to evolve with the project
Last year, the initial focus was on de-identification: information that could be used to identify a person or institution, such as names, places or dates of birth, was anonymised and automatically replaced with pseudonyms. The next step is now beginning at partner locations such as Dresden with semantic annotation. This involves categorising medical content such as diagnoses or procedures. In order to ensure the overarching comparability of these complex annotations, the GeMTeX team has defined detailed guidelines for semantic annotation, which are constantly evolving with the project.
At the plenary meeting, Dr. Markus Wolfien presented the current status of semantic annotation at the Dresden site and shared practical experiences and challenges. He also gave an insight into the documents for the annotation, the dashboard figures and information on further technical development.
Medical students wanted: We are looking for you to be at the heart of our project!
Medical students are the beating heart of GeMTeX - without your careful annotation and commitment, our text corpus would remain incomplete. Become part of the team and actively shape progress!
Apply now and become part of the GeMTeX team:
We look forward to your support!
Dr. Markus Wolfien und Karolin Hofmann