Deep Learning for Automated Myelin Segmentation in Electron Microscopy Data

Unpaid internship | 6–9 months | 1st or 2nd year master student

Are you a master student with a passion for coding and a curiosity about the brain? Join the Kole group at the Netherlands Institute for Neuroscience for a hands-on internship at the interface of machine learning and neuroscience!

About the project

Our lab uses electron microscopy (EM) to study the ultrastructure of the brain, including the myelin sheaths that wrap around axons in grey matter. Automated segmentation is essential for analyzing these large-scale EM datasets efficiently, but existing models still need refinement to reliably segment myelin structures in grey matter axons.

During this internship, you will independently train and improve existing automated segmentation models to enhance the accuracy of myelin segmentation in grey matter axons. You will work closely with our EM specialist and gain hands-on experience with real neuroscientific imaging data, contributing directly to ongoing research on brain ultrastructure.

What you will do

• Independently train and fine-tune existing deep learning-based segmentation models on EM datasets
• Evaluate and improve model performance for myelin segmentation of grey matter axons
• Work with large-scale electron microscopy image data

What we are looking for

• Currently enrolled in a 1st or 2nd year Master’s program in a relevant field such as (bio)informatics, computational science, AI, biomedical sciences, etc.
• Solid experience with Python programming
• Interest in EM image analysis, machine learning, and/or neuroscience
• Able to work independently with minimal guidance
• Proficient in written and spoken English
• Experience with image analysis and/or automated segmentation is a plus, but not required
• Available for 6–9 months, full-time

Interested?

Send your CV and a short motivation letter to b.lodder@nin.knaw.nl to apply.