Skip to content
nimble_asset_Women-in-Photonics_2020
Registration
nimble_asset_B4F3_mini_small_weiß
nimble_asset_IPHTLogo_neu_cmyk_02_de-e1578489596701
nimble_asset_WiPh2020_Pahlow

Larissa Heinrich

Cell organelle segmentation in electron microscopy
HHMI Janelia Research Campus // Ashburn, USA

Q&A Session I // Wednesday, October 28 // 6.15 pm – 7 pm (CET)

Datasets of cells and tissue imaged using modern high resolution 3D electron microscopy protocols, such as Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) show a plethora of cellular organelles and sub-cellular structures. However, extracting this multitude of information using a purely manual segmentation is too time-consuming, even for a single cell. We are therefore working towards machine learning based solutions to extract a comprehensive map of all known cellular organelles across a variety of cell types and ultimately within di fferent tissues from di fferent species.
We trained deep neural networks to directly and simultaneously predict signed boundary distances of 36 classes of cellular substructures. Naive thresholding of these predictions at zero produces a promising initial segmentation of cellular substructures. This data enables us to automatically align correlative light – electron microscopy images as well as address a variety of biological questions.

Figure 1: Segmentation of mitochondria (orange), endoplasmic reticulum (green), nucleus(magenta), endosomes (blue), vesicles (red), golgi apparatus (cyan) and plasma membrane (white) of a HeLa cell

Further Talks

nimble_asset_B4F_maxi_weiß

Operated by

Leibniz Institut of Photonic Technology
Jena | Germany

www.leibniz-ipht.de

Contact

Marc Skupch
Conference Organization

marc.skupch@leibniz-ipht.de