Birs- 23w5042: Leveraging Model- and Data-Driven Methods in Medical Imaging

  1. Title:
    Posterior-Variance-Based Error Quantification for Inverse Problems in Imaging

    Speaker:

    Link:

  2. Title:
    Continuous generative neural networks for inverse problems

    Speaker:

    Link:

  3. Title:
    A Cubic Correction in EIT Imaging

    Speaker:

    Link:

  4. Title:
    Unitarization of the Radon transform

    Speaker:

    Link:

  5. Title:
    Compressed sensing for the sparse Radon transform

    Speaker:

    Link:

  6. Title:
    Modelling data incompleteness in tomography

    Speaker:

    Link:

  7. Title:
    Using the 2DeteCT data collection as training data for data-driven methods in medical imaging

    Speaker:

    Link:

  8. Title:
    Bregman Iterations for sparse neural networks and architecture search

    Speaker:

    Link:

  9. Title:
    Learned variational regularization for linear inverse problems

    Speaker:

    Link:

  10. Title:
    Towards real-time solutions for inverse and imaging problems with uncertainty quantification

    Speaker:

    Link:

  11. Title:
    Spectral decomposition of atomic structures in heterogeneous cryo-EM

    Speaker:

    Link:

  12. Title:
    Reinterpreting survival analysis in the universal approximator age

    Speaker:

    Link:

  13. Title:
    The Power of Patches for Training Normalizing Flows

    Speaker:

    Link:

  14. Title:
    Generalizing PINNs to Complex Geometries

    Speaker:

    Link:

  15. Title:
    Shape Classification through the Lens of Geodesic Flows of Diffeomorphisms

    Speaker:

    Link:

  16. Title:
    Dynamic MRI Reconstruction with Locally Low-Rank Regularizers

    Speaker:

    Link:

  17. Title:
    Combing deep learning with the Electrical Impedance Tomography to classify stroke

    Speaker:

    Link:

  18. Title:
    Smooth over-parametrized solvers for non-smooth structured optimisation

    Speaker:

    Link:

  19. Title:
    Data driven regularization by projection

    Speaker:

    Link: