In previous years, we've re-written the OME analysis system to make it more robust and efficient (1), added an internationalization layer that includes a complete translation of both the user interface and the underlying semantic framework into Spanish, and have contributed to our collaborations that have further refined OME's user interface, as well as its underlying infrastructure (6). However, the bulk of the work in this project this year has been the validation of the WND-CHARM pattern recognition algorithm we have worked on for several years. WND-CHARM is a generalized pattern-recognition algorithm that can be used to analyze any type of image. We have validated the accuracy and generality of this approach using standard benchmark suites used for pattern recognition by the machine vision community, including face recognition, object recognition, and other non-biological classification problems (2). Due to the lack of biologically and medically oriented machine-vision benchmarks, we have developed our own benchmark suite to foster interest by the wider machine vision community. This comprises over 20 GB of image data spanning 11 different biological imaging problems, ranging from sub-cellular organelle identification to image-based screening, to medical diagnosis. Different imaging modalities are also represented including fluorescence microscopy, differential-interference contrast, phase-contrast and H&E stained brightfield. This image collection and benchmark has been made freely available to the public (http://ome.grc.nia.nih.gov/iicbu2008) (5). One of the WND-CHARM validations we performed resulted from an image dataset we used in our medical imaging project (AG000685-02: Pattern recognition in medical imaging). Here we were able to use knee X-Rays for biometric identification, which is the first example we are aware of that used internal tissue structures to identify individuals (3). While WND-CHARM is not yet fully integrated with OME, we published the source-code for this utility as well as a """"""""field manual"""""""" for its practical applications to imaging problems (4). In the coming year, we plan to test integrations of WND-CHARM with our existing OME platform as well as the OMERO platform being developed in Dr. Swedlow's lab in Dundee, Scotland. WND-CHARM is very computationally intensive and places enormous demands on an image informatics framework to supply it with images to analyze as well as consume and organize its outputs. The relative performance of these two systems will be critical for determining which of the two platforms to use in integration.
| Shamir, Lior; Delaney, John D; Orlov, Nikita et al. (2010) Pattern recognition software and techniques for biological image analysis. PLoS Comput Biol 6:e1000974 |
| Shamir, Lior; Ling, Shari; Rahimi, Salim et al. (2009) Biometric identification using knee X-rays. Int J Biom 1:365-370 |
| Shamir, Lior; Eckley, D Mark; Delaney, John et al. (2009) An Image Informatics Method for Automated Quantitative Analysis of Phenotype Visual Similarities. IEEE NIH Life Sci Syst Appl Workshop 2009:96-99 |
| Swedlow, Jason R; Goldberg, Ilya G; Eliceiri, Kevin W et al. (2009) Bioimage informatics for experimental biology. Annu Rev Biophys 38:327-46 |
| Shamir, Lior; Wolkow, Catherine A; Goldberg, Ilya G (2009) Quantitative measurement of aging using image texture entropy. Bioinformatics 25:3060-3 |
| Orlov, Nikita; Shamir, Lior; Macura, Tomasz et al. (2008) WND-CHARM: Multi-purpose image classification using compound image transforms. Pattern Recognit Lett 29:1684-1693 |
| Shamir, Lior; Orlov, Nikita; Eckley, D Mark et al. (2008) Wndchrm - an open source utility for biological image analysis. Source Code Biol Med 3:13 |
| Shamir, Lior; Orlov, Nikita; Mark Eckley, David et al. (2008) IICBU 2008: a proposed benchmark suite for biological image analysis. Med Biol Eng Comput 46:943-7 |
