Research Assistant/Associate (f/m/d)

Weitere Informationen

The successful candidate will be employed under a regular employment contract.
The position is to be filled at the earliest possible date and offered for a fixed term initially to three years.
The fixed-term employment is possible as it constitutes one of the fixed-term options of the Wissenschaftszeitvertragsgesetz (German Act on Fixed-term Scientific Contracts).
This is a full-time position.
The successful candidate has the opportunity to pursue a doctoral degree in this position.
The salary is based on the German public service salary scale (TV-L).
The position corresponds to a pay grade of EG 13 TV-L.

Unser Profil

The Chair of Imaging and Computer Vision (LFB) conducts pioneering research at the intersection of imaging instrumentation and computational imaging, ranging from the development of novel hybrid imaging systems to advanced image reconstruction and video communication. Building on this broad interdisciplinary expertise, we specifically focus on robust 3D reconstruction of anatomical structures and multimodal image analysis. By combining hybrid machine learning with physical modeling, we develop tailored solutions that bridge the gap between raw imaging data and clinical application, ensuring reliable decision support in complex medical workflows.

Join us at the forefront of medical AI technology as part of a new interdisciplinary research consortium. We are developing a fully automated process chain for surgical planning, specifically addressing dysgnathia and extremity surgery.

You will work in a strong interdisciplinary team alongside our consortium partners, including clinical experts from major university hospitals (providing data & validation), specialized software partners (responsible for platform integration), and experts in manufacturing technology (focusing on production automation).

The project aims to overcome the "data bottleneck" in medical AI. Instead of relying solely on massive manual annotation, we aim to leverage unlabelled data to build intelligent systems that truly "understand" 3D anatomy. We are moving from raw DICOM data to patient-specific, geometry-based surgical guides.

Ihr Profil

* Education: Master’s degree or comparable in Computer Science, Physics, Engineering, or a related field with a strong focus on AI/ML.

* Technical Skills: 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.

* Methodological Knowledge: Strong understanding of computer vision, representation learning, and high-dimensional geometry.

* Soft Skills: Passion for solving medical challenges and ability to work in multidisciplinary teams (engineers, clinicians).

Ihre Aufgaben

You will be responsible for researching and developing next-generation AI methods for 3D medical image understanding, aiming to automate complex surgical planning workflows using modern self-supervised techniques:

* Pioneering Self-Supervised Learning: Investigate and adapt modern Joint Embedding Predictive Architectures (e.g., I-JEPA, V-JEPA) for volumetric medical data (CT/DVT). Your goal is to develop a "Volumetric-JEPA" that learns robust anatomical representations from unlabelled data without relying on pixel-level reconstruction.

* Data-Efficient Segmentation: Utilize these pre-trained semantic backbones to achieve high-precision segmentation of critical structures (e.g., mandible, nerve canals) using Few-Shot Learning. This aims to drastically reduce the need for manual annotations compared to traditional supervised methods.

* Generative vs. Predictive AI: Systematically compare the robustness of predictive architectures (JEPA) against generative approaches (e.g., Diffusion Models, GANs), specifically regarding their ability to handle severe anatomical deformations and image artifacts (OOD robustness).

* Multimodal VLM Interface: Develop interfaces to feed the learned high-level feature embeddings into Multimodal Large Language Models, enabling a semantic "Safety Layer" that validates surgical plans based on anatomical understanding.

* Algorithmic Surgical Planning: Implement geometric algorithms (osteotomy planes, collision analysis) that build upon the robust features extracted by your AI models to automate the planning process.

Über uns

RWTH is a certified family-friendly University. We support our employees in maintaining a good work-life balance with a wide range of health, advising, and prevention services, for example university sports. Employees who are covered by collective bargaining agreements and civil servants have access to an extensive range of further training courses and the opportunity to purchase a job ticket.
RWTH is an equal opportunities employer. We therefore welcome and encourage applications from all suitably qualified candidates, particularly from groups that are underrepresented at the University. All qualified applicants will receive consideration for employment and will not be discriminated against on the basis of national or ethnic origin, sex, sexual orientation, gender identity, religion, disability or age. RWTH is strongly committed to encouraging women in their careers. Female applicants are given preference if they are equally suitable, competent, and professionally qualified, unless a fellow candidate is favored for a specific reason.
As RWTH is committed to equality of opportunity, we ask you not to include a photo in your application.
You can find information on the personal data we collect from applicants in accordance with Articles 13 and 14 of the European Union's General Data Protection Regulation (GDPR) at http://www.rwth-aachen.de/dsgvo-information-bewerbung.

Besoldung / Entgelt

EG 13 TV-L