Meta menu:

From here, you can access the Emergencies page, Contact Us page, Accessibility Settings, Language Selection, and Search page.

Open Menu


AI in Progress Test Medicine: from numerical feedback to individualized learning assistance.

You are here:

Funded by


Förderlinie zur digitalen Hochschulbildung, Vierte Förderlinie: Innovationen in der Hochschulbildung durch Künstliche Intelligenz und Big Data

Ausführende Stellen

Charité - Universitätsmedizin Berlin

AG Progress Test Medizin (Dr. med. vet. Maren März - maren.maerz[at] (Verbundkoordinatorin)

Institut für Biometrie und klinische Epidemiologie (Dr. Jochen Kruppa)

Dieter Scheffner Fachzentrum für medizinische Hochschullehre und Ausbildungsforschung (Prof. Dr. med. Harm Peters)


Technische Universität Ilmenau

Institut für Technische Informatik (Prof. Dr.-Ing. Patrick Mäder)



01.03.2021 - 29.02.2024


The Progress Test Medizin (PTM) provides students with regular feedback on their current level of knowledge and its development throughout their studies.
PTM tests are administered every academic semester to around 10,000 students in Germany, Austria and Switzerland. Since PTM tests consist of 200 questions from all fields of medicine, each test cycle generates about two million data points.

  1. The purpose of this project is to research the extent to which regular, content-related response patterns of students can be identified and characterised through the use of machine learning techniques; the question of whether the curricula of participating faculties could be somehow inferred using such methods will also be explored.
  2. The outcomes of this research might eventually make it easier to provide students and faculties with more accurate feedback on their educational improvement.
  3. Test questions will be subject to semantic processing with NLP algorithms; the results of this treatment should help improve feedback design and also facilitate test assembling by supplying optimised descriptions for prospective test questions.
  4. Which kind of feedback do students prefer? It must not be forgotten that formative and summative exams do not share the same communication priorities in this regard.
  5. The transfer of knowledge and methodological approaches will be tested in a model project to be carried out within the context of a Computer Science faculty.


Maren März

Charité - Universitätsmedizin Berlin

Project management, AI


Miriam Sieg

Charité - Universitätsmedizin Berlin

AI, Data visualization


Mihaela Todorova Tomova

Technische Universität Ilmenau



Jonas Elsner

Charité - Universitätsmedizin Berlin

Review of Questions, Question development


Iván Roselló Atanet

Charité - Universitätsmedizin Berlin

AI, Datamanagement


Dr. Sie-Youn Lee

Charité - Universitätsmedizin Berlin

Development of dashboard


Dr. Torsten Schröder

Charité - Universitätsmedizin Berlin

Development of dashboard