A Graph Matching Approach to Tracking Neurons in Freely-Moving C. elegans
Jones, C.; Barzegar-Keshteli, M.; Gross, A.; Obozinski, G.; Rahi, S. J.
Show abstract
MotivationRecent advances in 3D microscopy allow for recording the neurons in freely-moving C. elegans at high frame rates. In order to read out calcium activity, it is necessary to track individual neurons from frame to frame. However, doing this by hand for tens of neurons in a single ten-minute recording requires more than a hundred hours. Moreover, most methods proposed in the literature for tracking neurons focus on immobilized or partially-immobilized worms and fail with freely-behaving worms. ResultsIn this paper we present an approach based on graph matching for tracking fluorescently-marked neurons in freely-moving C. elegans. Neurites (and sometimes neurons) can be oversegmented into pieces at the preprocessing phase; our algorithm allows several segments to match the same reference neuron or neurite. We demon-strate our method on three recordings. We find that with five labeled frames we can typically track the neurons and pieces of neurites with over 75% accuracy, with more reliable annotations for the most distinctive neurons. Availability and ImplementationThe code and preprocessed data will be made available upon publication. Contactcorinne.jones@epfl.ch
Matching journals
The top 9 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Comparing a computational model of visual problem solving with human vision on a difficult vision task. 94%
- Multiple neural network approaches, including use of topological data analysis, enhances classification of human induced pluripotent stem cell colonies by treatment condition 93%
- Learning massive interpretable gene regulatory networks of the human brain by merging Bayesian Networks 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.