Abstract: Traditional immunofluorescence microscopy techniques recently have been extended to colocalize, i.e. to simultaneously label and image, a large number of proteins. The Toponome Imaging system (MELC/TIS) by Walter Schubert makes is possible to colocalize up to 100 and more proteins in a single fixed probe, aiming at gaining new insights into complex protein-protein functional networks (Schubert et al., Nature Biotechnology, Vol 24, 2006) . The challenge for analyzing the resulting multivariate image data sets lies in making both, the contained spatial information as well as the high-dimensional protein abundance data, readable to biological and medical experts in an interlinked and yet comprehensive way. For this purpose we demonstrate the efficiency of our visualization software tool Lasagne, which provides an interactive, real-time navigation through one or multiple data sets. It allows for a fast overview of the data, an in-depth annotation of histological features as well as a comparison of data sets. Data sets shown include MELC/TIS images of human tissue sections affected by the Psoriasis skin disease, and we also demonstrate the application of the Lasagne tool to other imaging techniques like spatially resolved FT-IR microspectroscopy.

Abstract: Serial Block Face Scanning Electron Microscopy (SBFSEM) has facilitated the 3D imaging of nervous systems at an isotropic resolution of 25 nm per voxel. Using SBFSEM, all neurites within a sample are imaged simultaneously and samples as large as an entire neocortical column can be processed. In order to automatically reconstruct the wiring diagram of a nervous system from an SBFSEM volume image, accurate 3D image segmentation is required. Established algorithms fail to provide the required accuracy. A new method developed at the Heidelberg Collaboratory for Image Processing goes beyond these algorithms by incorporating the non-local structure found in the volume image into the segmentation procedure. Moreover, segmentation criteria are not hand-crafted into the algorithm but are learned from a small subset of the data which was carefully traced by neuroscientists. Segmentation is finally cast into a global optimization problem which combines non-local image features with descriptors of the local image geometry. This method doubles the performance of a previous approach which did not use global optimization. The accuracy is not yet sufficient to allow for an automated analysis of synaptic connectivity. However, the 3D reconstructions provide insight into the geometry of nervous systems and serve as a starting point for semi-automatic procedures.
References

[1] K. L. Briggman and W. Denk, "Towards neural circuit reconstruction with volume electron microscopy techniques," Current Opinion in Neurobiology, vol. 16, no. 5, pp. 562–570, November 2006.

[2] W. Denk and H. Horstmann, "Serial block-face scanning electron microscopy to reconstruct three-dimensional tissue nanostructure," PLoS Biology, vol. 2, pp. 1900–1909, November 2004.

[3] B. Andres, U. Koethe, M. Helmstaedter, W. Denk, and F. A. Hamprecht, "Segmentation of SBFSEM volume data of neural tissue by hierarchical classification," in Pattern Recognition, ser. LNCS, G. Rigoll, Ed., vol. 5096. Springer, 2008, pp. 142–152.

see -->http://albert.rierol.net/index.html for what he is working on.

... associated workshop on Feb 4th: Working on FIJI, an ImageJ package (see Fiji website)

Abstract: Technical innovations in medical cross-sectional imaging have opened up new vistas for the exploration of the human body, enabling both high spatial and temporal resolution. However, these techniques have led to vast amounts of image data whose precise and reliable visual analysis by medical doctors and bioscientists requires a considerable amount of human intervention and expertise, thus resulting in a cost factor of substantial economic relevance. Hence, the computer-assisted analysis of medical image data has moved into the focus of interest as an issue of high priority research efforts. The talk covers novel methods for pattern analysis, visualization and computer-aided diagnosis in the field of biomedical imaging, specifically MRI, such as functional MRI for human brain mapping, dynamic cerebral contrast-enhanced perfusion MRI and new approaches to breast cancer diagnosis in MRI mammography. An outlook to topical projects in systems biology confirms the broad applicability of the presented methods.

Anke Meyer-Baese is author of   “Pattern Recognition for Medical Imaging” (2003).

see --> http://murphylab.web.cmu.edu/ for his projects.