Skip to main content Skip to main navigation

IML research department receives two grants for "AI in Medicine"

| Health & Medicine | Image Recognition & Understanding | Human-Machine Interaction | Interactive Machine Learning | Osnabrück / Oldenburg

DFKI's Interactive Machine Learning research department has received two grants to further its research in the field of medicine (see also The funding comes from Google and Accenture. The results will be published in the form of scientific papers.

© Google
In the Google-funded project ‘End-to-End Active Learning Framework for Medical Image Annotation’, DFKI is developing a more efficient way of recognising data on medical images. Our photo shows Md Abdul Kadir, Hans-Jürgen Profitlich and Hasan Md Tusfiqur Alam from the Interactive Machine Learning research department (from left).

Google Grant

The aim is to develop a modularised active learning framework within the Google Cloud Platform that enables the annotation of medical image data on a large scale in a cost-efficient manner while ensuring data sovereignty and data protection. The research area focuses its work on a use case in the area of federated learning for clinical data, taking into account data protection and security aspects. The goal is to create an end-to-end platform for efficient annotation that benefits both clinicians and the research community.


Accenture Grant

The aim of this research project is to investigate the capabilities of ChatGPT with regard to Natural Language Inference (NLI) in the context of healthcare. The focus is on tasks such as understanding information from clinical trials and evidence-based review of healthcare data. Using various chain-of-thought methods research is being conducted to improve the reasoning capabilities of ChatGPT and to integrate dynamic context analysis techniques for higher inference accuracy. The approach includes mechanisms such as context analysis and multi-hop reasoning.


The ‘AutoPrompt’ project funded by Accenture is researching the reasoning capabilities of ChatGPT in the context of clinical data in the healthcare sector. Our picture shows the scientist Siting Liang, who is driving forward ‘AutoPrompt’ in the Interactive Machine Learning research department.