Reconstruction, Techniques, and Validation
Reconstruction algorithms are used to build a geometric model of neurons, describing their morphology, from two-dimensional and three-dimensional images. Validation methods are used to determine the accuracy of the resulting models. Both reconstruction and validation are active areas of research, where proposed algorithms must address several complex problems, including (a) the use of a wide range of imaging modalities used to collect data, (b) the complex topological structure of interconnected branches within the neurons, and (c) automation for large data sets. Several methods have been proposed for addressing these issues; however, current reconstruction algorithms can be broadly placed into four categories: semiautomated software packages, local exploration, global processing, and crowdsourcing-based approaches.
Neuronal reconstructions provide geometric representations of cell and network morphology that can be used to perform quantitative analysis...
KeywordsSeed Point Electron Microscopy Data Multicompartmental Model Supervise Learning Technique Graph Isomorphism Problem
- Carnevale NT, Hines ML (2009) The neuron book. Cambridge University Press, CambridgeGoogle Scholar
- Jain V, Bollmann B, Richardson M, Berger D, Helmstaedter M, Briggman K, Denk W, Bowden J, Mendenhall J, Abraham W, Harris K, Kasthuri N, Hayworth K, Schalek R, Tapia JC, Lichtman J, Seung HS (2010a). Boundary learning by optimization with topological constraints. In: IEEE conference on computer vision and pattern recognition, San Francisco, CA, pp 2488–2495Google Scholar
- Luebke D, Reddy M, Cohen J, Varshney A, Watson B, Huebner R (2002) Level of detail for 3D graphics. Morgan Kaufmann, Palo AltoGoogle Scholar
- Seung S, Burnes L (2012) Eyewire. http://eyewire.org/