Structured learning of assignment models for neuron reconstruction to minimize topological errors.

janelia7_blocks-janelia7_biblio_header | block

IEEE 13th International Symposium on Biomedical Imaging (ISBI). 2016 Ap 13;:607-11. doi: 10.1109/ ISBI.2016.7493341

Structured learning of assignment models for neuron reconstruction to minimize topological errors. Cardona LabFunke Lab

Funke J, Klein J, Moreno-Noguer F, Cardona A, Cook M

janelia7_blocks-janelia7_biblio_abstract | block

Abstract

Structured learning provides a powerful framework for empirical risk minimization on the predictions of structured models. It allows end-to-end learning of model parameters to minimize an application specific loss function. This framework is particularly well suited for discrete optimization models that are used for neuron reconstruction from anisotropic electron microscopy (EM) volumes. However, current methods are still learning unary potentials by training a classifier that is agnostic about the model it is used in. We believe the reason for that lies in the difficulties of (1) finding a representative training sample, and (2) designing an application specific loss function that captures the quality of a proposed solution. In this paper, we show how to find a representative training sample from human generated ground truth, and propose a loss function that is suitable to minimize topological errors in the reconstruction. We compare different training methods on two challenging EM-datasets. Our structured learning approach shows consistently higher reconstruction accuracy than other current learning methods.

node:body | entity_field

Main Menu (Mobile)- Block

Main Menu - Block

Structured learning of assignment models for neuron reconstruction to minimize topological errors.

Abstract

Janelia Authors