Radiology AI-Driven Clinical Decision-Making with Multi-Modal Exploration (RADICAL)
Description
The rapid digital transformation in healthcare has surged data volume and variety, overwhelming clinicians. Much of this data is unstructured, making it hard to explore. Radiology, a key diagnostic field, faces fragmented records.
We propose RADICAL, an AI tool streamlining workflows with natural language queries, improving decisions and outcomes.
Key data
Projectlead
Co-Projectlead
Project team
Project partners
Universität Zürich UZH / KrauthammerLab; Universitätsspital Zürich USZ / Institut für diagnostische und interventionelle Radiologie
Project status
ongoing, started 11/2025
Institute/Centre
Institute of Computer Science (InIT)
Funding partner
Digitalisierungsinitiative der Zürcher Hochschulen DIZH
Project budget
599'393 CHF
Further documents and links
Publications
-
RadVLM-GRPO : enhancing chest X-ray report generation and visual grounding via reinforcement learning
2026 Gundersen, Benjamin; Deperrois, Nicolas; Ruiperez-Campillo, Samuel; Sutter, Thomas M.; Vogt, Julia E.; Moor, Michael; Nooralahzadeh, Farhad; Krauthammer, Michael
-
Multi-modal data exploration via language agents
2025 Nooralahzadeh, Farhad; Zhang, Yi; Fürst, Jonathan; Stockinger, Kurt
School of Engineering
Obere Kirchgasse 2 / Steinberggasse 12/14
8400 Winterthur