MitoNet and EBI/EMPIAR

Trainer: Kedar Narayan

This session was a part of the 2023 iteration of this course.

How datasets curated for AI helped create SOTA segmentation models for volume electron microscopy

Overview: 

This session presents MitoNet and the role that curated datasets for deep learning applications, such as EMPIAR‘s datasets, helped create segmentation models for volume electron microscopy.

Learning outcomes:

After this session you should be able to:

• List the bottlenecks in volume electron microscopy (vEM)
• Identify deep learning approaches are expanding this bottleneck
• Explore new approaches to face these barriers
• Identify community resources to use and create your own deep learning models

Materials:

• Presentation
• Article: Instance segmentation of mitochondria in electron microscopy images with a generalist deep learning model trained on a diverse dataset
• Further EMPANADA resources: Empanada and MitoNet – Projects developed for the volume EM community