Zeynep Boztoprak

Zeynep Boztoprak

PhD student

Research Topics

Deep Learning, Multimodal Learning, Self-Supervised Representation Learning, Computer Vision, High Performance Computing, Distributed Deep Learning

Contact

+49 2461/61-5960

E-Mail

Address

Forschungszentrum Jülich GmbH
Wilhelm-Johnen-Straße
52428 Jülich

Institute of Neurosciences and Medicine (INM)

Structural and Functional Organisation of the Brain (INM-1)

Building 15.9 / Room 3028

Current research

I develop deep learning methods for integrating microscopic brain imaging data into reference atlases. Instead of relying on classical registration, which requires anatomical landmarks that small tissue samples often lack, I train neural networks with self-supervised objectives to learn features that capture anatomical organization. Unseen histological patches can then be localized by semantic similarity to annotated reference data. I am extending this idea across imaging modalities to enable integration of heterogeneous microscopy data into common frameworks such as MNI Colin27.

Vita
Publications
Last Modified: 13.04.2026