English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

The throughput of electron microscopes has increased significantly in recentyears, enabling detailed analysis of cell morphology and ultrastructure.Analysis of neural circuits at single-synapse resolution remains the flagshiptarget of this technique, but applications to cell and developmental biologyare also starting to emerge at scale. The amount of data acquired in suchstudies makes manual instance segmentation, a fundamental step in many analysispipelines, impossible. While automatic segmentation approaches have improvedsignificantly thanks to the adoption of convolutional neural networks, theiraccuracy still lags behind human annotations and requires additional manualproof-reading. A major hindrance to further improvements is the limited fieldof view of the segmentation networks preventing them from exploiting theexpected cell morphology or other prior biological knowledge which humans useto inform their segmentation decisions. In this contribution, we show how suchdomain-specific information can be leveraged by expressing it as long-rangeinteractions in a graph partitioning problem known as the lifted multicutproblem. Using this formulation, we demonstrate significant improvement insegmentation accuracy for three challenging EM segmentation problems fromneuroscience and cell biology.

Leveraging Domain Knowledge to Improve Microscopy Image Segmentation with Lifted Multicuts