We have optimised several components of wide-field fluorescence microscopy (e.g. high NA 20x and 10x objectives, large range piezo stages and sensitive, small pixel size, large pixel number Complementary metal–oxide–semiconductor (CMOS) cameras). Using this system we can image entire zebrafish larvae, in less than 4 minutes at a resolution greater than 500nm. This is achieved by taking z-sections at 1μm resolution over 500 μm (sufficient to encompass the whole organism in the left-right axis) at 500nm XY resolution. This imaging platform enables us to capture sub-cellular structures throughout the organism (for example the phagosomes of every phagocyte) in a living vertebrate during infection.
Our imaging system generates very large data files with each containing a least 1 billion voxels to be analysed. Combined with the difficulties of identifying and segmenting cell shapes in three dimensions this is a very difficult imaging analysis problem that has until now been tackled manually. In collaboration with Prof. Alejandro Frangi (Director of the Centre for Computational Imaging and Simulation Technologies in Biomedicine - CISTIB) we are developing computer vision techniques to tackle this problem.