Location: Institute for Cognitive Neurology and Dementia Research Magdeburg, Germany (partial remote possible)

Start Date: 1.10.2026 or later

Duration: 33 months, fixed-term

The Opportunity

The Modeling and Neuroprognosis Group at Institue for Cognitive Neurology and Dementia Research (IKND) Magdeburg is expanding its disease-modeling program with a newly funded postdoctoral position dedicated to computational disease progression modeling (DPM) in Alzheimer's disease. The position establishes state of the art modeling expertise on how neurodegenerative disease actually unfolds over time, both descriptively and mechanistically, in large-scale longitudinal cohort data.

This is a rare opportunity to take ownership of a well-defined, high-impact modeling agenda, with direct access to comprehensive dementia-risk cohorts (DZNE) and an established methodological foundation to build on. The position is embedded in strong modelling group with focus on multivariate analysis and AI using neuroimaging for aging and dementia research (https://iknd.med.ovgu.de/AG+Dr_+Ziegler.html) and enables learning important quantitative skills for later career developments in academia and industry.

The Project & Scientific Focus

Building directly on previous work (Lattmann et al., 2026, Nat Comms), the position centers on state of the art disease progression modelling including

• Multi-domain biomarker integration using longitudinal neuroimaging
• Subtype-aware continuous trajectory modeling; uncertainty quantification
• Mechanistic, dynamical-systems modeling using ODEs; counterfactual interventions

You will work with large-scale longitudinal multi-domain cohorts in neurodegeneration and will interface closely with leading experts in the field of disease progression modelling and AI for medical imaging.

Your Responsibilities

• Develop and validate novel disease progression models for longitudinal multi-domain biomarker data.
• Implement and fit mechanistic, ODE-based dynamical models of inter-modality biomarker coupling and disease-stage progression.
• Apply Bayesian inference and state-space methods to estimate latent disease time and subject-specific trajectories.
• Conduct counterfactual intervention simulations and contribute to retrospective power-analysis and trial-design applications.
• Publish high-impact research and present at international conferences.
  
  

Your Profile

• Very ambitious and strongly interested in quantitative research and modelling
• PhD in Applied Mathematics, Physics, Machine Learning, Computational Neuroscience, Statistics, or a related quantitative field (e.g. Psychology with strong quantitative focus)
• Demonstrated expertise in at least one of: dynamical-systems/state-space modeling, Bayesian inference, mixed-effects modelling, deep generative models (VAEs, neural ODEs), or disease progression modeling.
• Proficiency in Python and standard scientific computing/ML frameworks (e.g., PyTorch, Stan, or equivalent).
• Experience with and neuroimaging, longitudinal or clinical cohort data is a clear advantage.
• Excellent scientific writing and communication skills; motivation for interdisciplinary, translational research.
  
  

What We Offer

• Direct access to leading longitudinal aging and dementia-risk cohorts.
• State-of-the-art computational and imaging infrastructure (3T & 7T MRI, HPC).
• Integration into a very active, internationally networked group, with collaborations including Inria, UCL, Cambridge, and Max Planck Institutes.
• Close mentorship and support for independent funding applications (DFG, ERC, NIH) beyond the initial term.
• Dedicated funding for international conference travel.
• A 33-month, fixed-term contract.
  
  

Application Process

Please submit a single PDF including a cover letter detailing your motivation and alignment with the position, a CV including publication list, contact information for two academic referees, and copies of academic transcripts and PhD certificate, via email. Applications are reviewed on a rolling basis until the position is filled.

For informal inquiries, please contact Dr. Gabriel Ziegler (gabriel.ziegler@dzne.de).