This partnership is led by Badrinath Roysam at Rensselaer Polytechnic Institute, and partners from Rensselaer, University of Wisconsin, The Wadsworth center, and Kitware Inc. The goal of this partnership is to develop and disseminate FARSIGHT - a next-generation image analysis toolkit to enable quantitative studies of complex multi-cellular structures such as stem-cell niches, brain tissue, and tumors, from multichannel three-dimensional fluorescence microscopy images. These images record multiple molecular markers in their three-dimensional context, offering unprecedented detail on structure-function relationships. They have the potential to help us achieve a new level of understanding of these complex systems. However, our understanding of these relationships is hampered by the lack of computational tools that can extract meaningful measurements from the complex and voluminous data generated by modern optical microscopes. While current software tools allow visualization and manual analysis of a few datasets, they are lacking in terms of automated quantitative large-scale analysis. These methods lack the capability to analyze complex multi-component systems, and the capacity to handle large batches of data sets. The proposed toolkit will overcome these limitations and enable a new generation of quantitative investigations for both hypothesis-driven and discovery-driven research. This toolkit will be developed collaboratively by a multi-disciplinary team organized along four disciplinary cores, and four cross-disciplinary workgroups. Development of the proposed system will be guided by a processing architecture that emphasizes the computation of associative measurements relating multiple tissue entities, and efficient validation using an edit-based methodology. It will be validated quantitatively by a research program aimed at achieving a new level of understanding of the complex reactive response of brain tissue to the insertion of neural prostheses, and two external collaborations in stem-cell biology and tumor biology. Rigorous software engineering processes and systematic collaboration will ensure that the resulting system is intuitive and usable, based on a modular and extensible architecture, rapidly programmable to a new application, and robust. Beyond the initial 5-year phase, this partnership will extend the proposed toolkit to include time-lapse, fluorescence lifetime, phase, polarization, and hyperspectral data.
Showing the most recent 10 out of 23 publications
