Neuron recognition with hidden neural network random fields

X. Chang, Michael D. Kim, R. Stephens, T. Qu, S. Gulyanon, A. Chiba, G. Tsechpenakis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations


We model neuron morphology with a hidden Conditional Random Field variant, a hidden Neural Network Random Field, for part-based classification. We aim at identifying the diverse morphologies of individual motor neurons in the Drosophila larvae, and understanding underlying principles of synaptic connectivity in a motor circuit. The motivation of our work is the bottom-up reconstruction of the Drosophila connectome, ie., fully annotated neuronal circuits where neurons and synapses are automatically not only traced but also identified. We use images depicting single neurons labeled with green fluorescent protein (GFP) and serially imaged with laser scanning confocal microscopy. In our approach, we consider that each neuron has already been partitioned into its structurally significant parts, namely soma, axon, and dendrites, and their (latent) labels are known. We demonstrate the accuracy of our approach using wild-type motor neurons in the larval ventral nerve cord, and make comparisons with existing methods, including our previous work.

Original languageEnglish (US)
Title of host publication2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages4
ISBN (Print)9781467319591
StatePublished - Jul 29 2014
Event2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014 - Beijing, China
Duration: Apr 29 2014May 2 2014


Other2014 IEEE 11th International Symposium on Biomedical Imaging, ISBI 2014


  • Drosophila
  • Hidden conditional random fields
  • Neuron morphology
  • Part-based classification

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging


Dive into the research topics of 'Neuron recognition with hidden neural network random fields'. Together they form a unique fingerprint.

Cite this