Detection of Neuron Membranes in Electron Microscopy Images using Auto-Context

Study of nervous systems via the connectome, the map of connectivities of all neurons in that system, is a challenging problem in neuroscience. Towards this goal, neurobiologists are acquiring large electron microscopy datasets. However, the shear volume of these datasets renders manual analysis infeasible. Hence, automated image analysis methods are required for reconstructing the connectome from these very large image collections. Segmentation of neurons in these images, an essential step of the reconstruction pipeline, is challenging because of noise, anisotropic shapes and brightness, and the presence of confounding structures. The method described in this talk uses a series of artificial neural networks (ANNs) in an auto-context framework combined with a feature vector that is composed of image intensities sampled over a stencil neighborhood. Employing auto-context means that several ANNs are applied in series while allowing each ANN to use the classification context provided by the previous network to improve detection accuracy. We develop the method of serial ANNs for auto-context and show that the learned con- text does improve detection over traditional ANNs. We also demonstrate advantages over previous membrane detection methods. The results are a significant step towards an automated system for the reconstruction of the connectome.