Elastic volume reconstruction from series of ultra-thin microscopy sections

Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany.
Nature Methods (Impact Factor: 32.07). 06/2012; 9(7):717-20. DOI: 10.1038/nmeth.2072
Source: PubMed


Anatomy of large biological specimens is often reconstructed from serially sectioned volumes imaged by high-resolution microscopy. We developed a method to reassemble a continuous volume from such large section series that explicitly minimizes artificial deformation by applying a global elastic constraint. We demonstrate our method on a series of transmission electron microscopy sections covering the entire 558-cell Caenorhabditis elegans embryo and a segment of the Drosophila melanogaster larval ventral nerve cord.

Download full-text


Available from: Albert Cardona,
  • Source
    • "Corresponding SIFT features between adjacent sections were used to determine a rigid linear transformation between sections, which was applied to all sections in the dataset to achieve a coarse, 3D rigid alignment of the data. Then, we applied elastic registration (Saalfeld et al., 2012) to further improve the alignment accuracy between adjacent sections while minimizing the global deformation of the entire image block. The warping transforms generated in these steps were applied to all conventional fluorescence and STORM channels. "
    [Show abstract] [Hide abstract]
    ABSTRACT: As a basic functional unit in neural circuits, each neuron integrates input signals from hundreds to thousands of synapses. Knowledge of the synaptic input fields of individual neurons, including the identity, strength, and location of each synapse, is essential for understanding how neurons compute. Here, we developed a volumetric super-resolution reconstruction platform for large-volume imaging and automated segmentation of neurons and synapses with molecular identity information. We used this platform to map inhibitory synaptic input fields of On-Off direction-selective ganglion cells (On-Off DSGCs), which are important for computing visual motion direction in the mouse retina. The reconstructions of On-Off DSGCs showed a GABAergic, receptor subtype-specific input field for generating direction selective responses without significant glycinergic inputs for mediating monosynaptic crossover inhibition. These results demonstrate unique capabilities of this super-resolution platform for interrogating neural circuitry.
  • Source
    • "One promising approach is (three-dimensional) reconstruction from serial section transmission electron microscopy (TEM), which is nowadays a well-established way of analyzing circuitry of neural networks [1–3]. However, several hundreds of sections or even more have to be cut without any loss of sections, inspected and photographed with the TEM, resulting in an enormous data volume, which is followed by a complex elastic alignment to compensate inevitable image distortions using an elastic alignment program (for example, TrakEM2 [4,5]). Hence, the main criterion in selecting a suitable subject for such a study is a small size. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The research field of connectomics arose just recently with the development of new three-dimensional-electron microscopy (EM) techniques and increasing computing power. So far, only a few model species (for example, mouse, the nematode Caenorhabditis elegans, and the fruit fly Drosophila melanogaster) have been studied using this approach. Here, we present a first attempt to expand this circle to include pycnogonids, which hold a key position for the understanding of arthropod evolution. The visual neuropils in Achelia langi are studied using a focused ion beam-scanning electron microscope (FIB-SEM) crossbeam-workstation, and a three-dimensional serial reconstruction of the connectome is presented. The two eyes of each hemisphere of the sea spider’s eye tubercle are connected to a first and a second visual neuropil. The first visual neuropil is subdivided in two hemineuropils, each responsible for one eye and stratified into three layers. Six different neuron types postsynaptic to the retinula (R-cells) axons are characterized by their morphology: five types of descending unipolar neurons and one type of ascending neurons. These cell types are also identified by Golgi impregnations. Mapping of all identifiable chemical synapses indicates that the descending unipolar neurons are postsynaptic to the R-cells and, hence, are second-order neurons. The ascending neurons are predominantly presynaptic and sometimes postsynaptic to the R-cells and may play a feedback role. Comparing these results with the compound eye visual system of crustaceans and insects – the only arthropod visual system studied so far in such detail – we found striking similarities in the morphology and synaptic organization of the different neuron types. Hence, the visual system of pycnogonids shows features of both chelicerate median and mandibulate lateral eyes.
    BMC Biology 08/2014; 12(1):59. DOI:10.1186/s12915-014-0059-3 · 7.98 Impact Factor
  • Source
    • "We triangulate 1 the points C (see Fig.1b), providing an augmented set of control points C ′ ⊇ C and a set of undirected edges E ⊆ C ′ × C ′ . The regularization consists of pairwise spring-like terms [6] [18], penalizing differences in relative control point displacements "
    [Show abstract] [Hide abstract]
    ABSTRACT: We describe an automatic method for fast registration of images with very different appearances. The images are jointly segmented into a small number of classes, the segmented images are registered, and the process is repeated. The segmentation calculates feature vectors on superpixels and then it finds a softmax classifier maximizing mu-tual information between class labels in the two images. For speed, the registration considers a sparse set of rectangular neighborhoods on the interfaces between classes. A triangulation is created with spatial regularization handled by pairwise spring-like terms on the edges. The optimal transformation is found globally using loopy belief propagation. Multiresolution helps to improve speed and ro-bustness. Our main application is registering stained histological slices, which are large and differ both in the local and global appear-ance. We show that our method has comparable accuracy to standard pixel-based registration, while being faster and more general.
    International Symposium on Biomedical Imaging, IEEE, Beijing, China; 05/2014
Show more