Matthias Keicher

Matthias Keicher is a Postdoc and Research Manager at the Chair for Computer Aided Medical Procedures at the Technical University of Munich, affiliated with the IFL Lab at Klinikum Rechts der Isar. His work focuses on deploying AI for clinical applications, particularly vision-language models and large language models for structured report generation and decision support systems.

• Education: Dipl.-Ing. in Mechanical Engineering and Management from TUM (2006-2013)
• Industry Experience: Former CTO and Managing Director at SurgicEye GmbH (2016-2018)

Research Interests:

• Medical Vision-Language Models (VQA, structured reporting)
• Multimodal Deep Learning for diagnostics
• Interpretable AI with generative models
• Decision support systems integrating patient data

Article Trends: Over 2024-2014, his publications span surgical phase recognition (TeCNO), vertebral fracture grading (iMIMIC best paper), chest X-ray classification (FlexR), radiology report generation (RaDialog), and toxin prediction (ToxNet). Keywords include Medical Imaging, Graph Networks, Language Models, with subfields like 3D Computer Vision, Federated Learning, Clinical Reasoning.

Scientific Awards:

• MICCAI iMIMIC 2023 Best Paper

Teaching: He organizes two lectures (Computer Science for Medical Students, Innovation Generation in Healthcare) and tutors courses such as Deep Learning for Medical Applications and Machine Learning in Medical Imaging.

Labs & Teams: Works at the IFL Lab (Intelligent Future Lab) in Munich, leading a research team funded by the DIVA project focused on vision-language models in clinical settings.