Cellular-resolution connectomics: challenges of dense neural circuit reconstruction

Abstract

A study summarizes and quantifies the challenges for data analysis in cellular-resolution connectomics and describes current solutions involving online crowd-sourcing and machine-learning approaches. On one hand, the key prospect for the improvement of automated analysis remains in lowering error rates (that is, generating longer correctly reconstructed neurite segments), but on the other hand, the focus is on better integration with human annotation.