Vacancies | Chair of Imaging and Computer Vision

Key Responsibilities:

You will pioneer the application of State Space Models (SSMs/Mamba) to automate complex surgical planning, replacing computationally heavy Transformer architectures.
Design 3D State Space Architectures: Develop “Volumetric-SSM” backbones (e.g., Mamba) to process highresolution 3D anatomy (CT/DVT) with linear complexity, capturing global relationships without the bottlenecks of patch-based sliding windows.
Self-Supervised Pre-training: Create SSL strategies to learn robust anatomical representations from unlabelled data, drastically reducing the need for manual annotations.
Robustness Analysis: Systematically benchmark SSM efficiency and OOD robustness against JEPAs and Generative models, specifically regarding severe anatomical deformations.
Semantic & Geometric Integration: Interface high-level states with Multimodal LLMs for a semantic “Safety Layer” and implement geometric algorithms (osteotomy planes, collision analysis) to fully automate the planning process.

Candidate Profile:

Excellent (top 10%) Master’s degree in Computer Science, Physics, Engineering, or a related field with a strong focus on AI/ML.
Proficient in Python and Deep Learning frameworks (PyTorch). Experience with Self-Supervised Learning (SSL), Transformers, or modern architectures like Masked Autoencoders (MAE) / JEPA is highly desirable.
Strong understanding of computer vision, representation learning, and high-dimensional geometry.
Passion for solving medical challenges and ability to work in multidisciplinary teams (engineers, clinicians).

We continuously hire HiWis and would be happy to hear from you—feel free to come and talk to us anytime!.