Skip to content
Home » Archives for Sabine Dani

Sabine Dani

Design Your City

“Design Your City” is an innovative virtual reality experience that enables new forms of participation in urban development. The aim is to actively involve the population and make changes in public spaces comprehensible to everyone. https://youtu.be/LwFxWWN4gmA Instead of presenting traditional plans, the virtual reality tool allows residents to experience planned… Read More »Design Your City

DSI Insights: AI in psychotherapy

Dialogue-oriented artificial intelligence is increasingly being used in mental health care. It supports psychoeducation, self-management exercises, progress monitoring, and the implementation of structured therapeutic techniques. At the same time, its use raises fundamental questions: What does it mean when AI simulates empathy and care without itself bearing responsibility or values?… Read More »DSI Insights: AI in psychotherapy

Project closure: Re-Experiencing History

The Re-Experiencing History project has been completed. An interactive platform has been developed that allows users to visually design historical scenes from antiquity themselves. With the help of AI and specially trained models, users can create their own visualizations via a prompt interface, whether of Roman triumphal processions or everyday… Read More »Project closure: Re-Experiencing History

Posteroanterior chest radiograph ("X-ray") taken of a 21 year old woman who presented with pain on the left side of her thorax after colliding with another player during a soccer game. It shows a normal chest without any signs of injury. Dx and Sin stand for "right" and "left" respectively. Photo: Mikael Häggström

TRUST-RAD: New dataset for reliable AI assistance tools in radiology published

The TRUST-RAD project team has published a comprehensive dataset for the development of AI assistance systems for X-ray image interpretation. RadVLM Instruction Dataset is now available to the scientific community on the PhysioNet platform. The dataset comprises over 1.1 million image-instruction pairs from chest X-rays and covers various radiological tasks:… Read More »TRUST-RAD: New dataset for reliable AI assistance tools in radiology published