Serial Neural Network Classifier for Membrane Detection using a Filter Bank

Serial Neural Network Classifier for Membrane Detection using a Filter Bank,Elizabeth Jurrus,Antonio R. C. Paiva,Shigeki Watanabe,Ross Whitaker,Erik M

Serial Neural Network Classifier for Membrane Detection using a Filter Bank   (Citations: 2)
BibTex | RIS | RefWorks Download
Study of nervous systems via the connectome, i.e. the map of the connectivities of all neurons in that system, is a challenging problem in neuroscience. Towards this goal, neurobiologists are acquiring large electron microscopy datasets. Automated image analysis methods are required for reconstruct- ing 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, irregular shapes and brightness, and the presence of confounding structures. The method described in this paper uses a carefully designed set of filters and a series of artificial neural netwo rks (ANNs) in an auto-context architecture to detect neuron mem- branes. Employing auto-context means that several ANNs are applied in series while allowing each ANN to use the classific ation context provided by the previous network to improve detection accuracy. We use the responses to a set of filters as input to the series of ANNs and show that the learned context 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 fo r the reconstruction of the connectome.
Published in 2009.
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...Noteworthy work in this field includes techniques focused on cell boundary segmentation [18, 6, 13, 15], mitochondria segmentation [11, 16] and vesicle segmentation [2]...
    • ...For this purpose, cell boundary enhancement (and eventual segmentation) has attracted attention in recent years [6, 18]...
    • ...Focus of the work in this field has been on largely unsupervised diffusion based techniques [17, 12] as well as those based on bulky convolutional networks [5, 6] and graph-cuts methods [18]...

    Ritwik Kumaret al. Radon-Like features and their application to connectomics

    • ...Tu and Bi [10] used boosting whereas Jurrus et al. [6] employed ANNs...
    • ...In our proposed method, we employ artificial neural networks in an auto-context architecture as in [6], except that we introduce three-state neurons instead of conventional two-state neurons (Figure 2b)...

    Mojtaba Seyedhosseiniet al. Image Parsing with a Three-State Series Neural Network Classifier

Sort by: