Vision-Language Group

About

Welcome to the Vision-Language Group of the Chair for Computer-Aided Medical Procedures & Augmented Reality (CAMP) of Prof. Dr. Nassir Navab at the Technical University of Munich (TUM). Our team investigates the synergy between natural language processing (NLP), computer vision, and deep learning, particularly in the medical field.

We focus on creating solutions that analyze and interpret medical images and articulate these findings clearly for medical professionals, thereby enhancing diagnosis accuracy and patient outcomes. Our research spans across areas like radiology report generation, medical visual question answering (VQA), multimodal data integration, and interactive clinical decision support with transparent reasoning.

Contact Person / Group Coordination

Matthias Keicher - matthias.keicher@tum.de

Group Members

Current Research Projects

Agentic clinical reasoning with reinforcement learning
Patient pathway modelling with multimodal EHR data
Confidence calibration for radiology report generation
Guideline-grounded stroke treatment decision support with RAG
Diagnostic reasoning models for decision support in toxicology
ThoraXAI - Trustworthy and explainable AI for Thorax Diagnostics
DIVA - Diagnostic and Decision Support for Vertebral Body Fractures

Recent Publications

preprint	From EHRs to Patient Pathways: Scalable Modeling of Longitudinal Health Trajectories with LLMs
ICLR 2026	Language Agents for Hypothesis-driven Clinical Decision Making with Reinforcement Learning
ICLR 2026	Rewarding Doubt: A Reinforcement Learning Approach to Calibrated Confidence Expression of Large Language Models
npj Digital Medicine	Specialized foundation models for intelligent operating rooms
CVPR 2025	MM-OR: A Large Multimodal Operating Room Dataset for Semantic Understanding of High-Intensity Surgical Environments
MIDL 2025	RaDialog: A Large Vision-Language Model for Radiology Report Generation and Conversational Assistance
MedAGI 2024	MAGDA: Multi-agent Guideline-Driven Diagnostic Assistance
MICCAI 2024	ORacle: Large Vision-Language Models for Knowledge-Guided Holistic OR Domain Modeling
MICCAI 2023	Rad-ReStruct: A Novel VQA Benchmark and Method for Structured Radiology Reporting
MICCAI 2023	Xplainer: From X-Ray Observations to Explainable Zero-Shot Diagnosis
GRAIL 2023	Prior-RadGraphFormer: A Prior-Knowledge-Enhanced Transformer for Generating Radiology Graphs from X-Rays
MIDL 2023	FlexR: Few-shot Classification with Language Embeddings for Structured Reporting of Chest X-rays

Links

Location

IFL: Interdisziplinäres Forschungslabor

Klinikum rechts der Isar
Ismaninger Str. 22
81675 München
Room: 01.01.3a-c (basement floor)
Phone: +49 (89) 4140-6457
Fax: +49 (89) 4140-6458