Crowdsourcing the creation of image segmentation algorithms for connectomics
Ignacio Arganda‐Carreras, Srinivas C. Turaga, Daniel Raimund Berger, Dan Cireşan, Alessandro Giusti, Luca Maria Gambardella, Jürgen Schmidhuber, Dmitry Laptev et al.
Frontiers in Neuroanatomy·2015·article·cited by 341
importance 6/10partially solvedCell Image Analysis Techniquesneuroscience / connectomicsstage: research
Problem
Automating reconstruction of neural circuits via 2D segmentation of electron microscopic brain images.
Why it matters
Automated neural circuit reconstruction is essential for mapping brain connectivity at scale.
What the AI did
An international challenge crowdsourced segmentation algorithms; the winning team used a convolutional network to predict boundary maps, later becoming the field standard.
Results
not stated in source
Impact
Deep learning became the accepted standard for EM image segmentation, though the challenge saturated at scoring limits.
Record
Venue
Frontiers in Neuroanatomy
Published
2015
Work type
article
Authors
Ignacio Arganda‐Carreras, Srinivas C. Turaga, Daniel Raimund Berger, Dan Cireşan, Alessandro Giusti, Luca Maria Gambardella, Jürgen Schmidhuber, Dmitry Laptev et al.
connectomics, image segmentation, deep learning, crowdsourcing, electron microscopy
Importance rating
6/ 10
significant — thousands to millions affected
Automating neural circuit reconstruction supports connectomics, a foundational neuroscience effort; abstract gives no scale, but the method became a field standard, indicating broad scientific significance.
Basis: inferred.
Judged from the source text, which gives no explicit scale.
Stated scale: not stated in source. Method and limits.