MicroMos: an open source software tool to obtain high-resolution panoramic images of 2D cell cultures

OBJECTIVE: In order to perform significant analyses, the microscope’s user is typically interested in acquiring high-detailed images representing an entire histological sample or well containing cells. Due to the finite size of the camera’s field of view, the microscopist has to find the proper trade-off between higher magnification factor and extension of the observed area. Here we present MicroMos version 3.0, an open-source tool for building mosaics by stitching together more partially overlapping images. The method proposed is based on visual information only and the mosaics are built by incrementally stitching couples of images. The radiance of the original sample is preserved by compensating the vignetting effect of each stitched image, and the mosaics obtained can be used for quantitative analyses. By exploiting MicroMos we studied confluence and proliferation of mesenchymal stromal cells (MSC) adherent on OSPROLIFE (Eurocoating, Cirè-Pergine, Italy), a commercial biomaterial in the form of granules. In order to acquire statistically significant data we analysed mosaics built using MicroMos, implementing customized algorithms to segment the MSC and automatically estimate of the percentage of the area of the granules covered by cells. MATERIALS AND METHODS: To validate the proposed mosaicing method we performed several experiments under different working conditions. In particular, in order to assess the quality of the mosaics obtained using different warping models and tonal adjustments we used six different sets of images of histological samples and cell cultures. We aligned the images according to different registration models and computed several metrics to estimate which registration method performs as the best. Then, to study confluence and proliferation of MSC adherent on OSPROLIFE granules, we prepared several samples each containing 50 mg of OSPROLIFE granules and a different number of MSC. Cell confluence was evaluated by adding 4 µM Calcein AM before acquiring sequence of partially overlapping images. RESULTS: In our experiments we tested different registration approaches, confirming quite unexpectedly that the translational model does not always act as the best, although the motion of the microscope’ sample holder is apparently translational. Indeed, the sample holder could be slightly inclined, hence yielding non-negligibly affine, or even projective, transformations between subsequent images. CONCLUSION: MicroMos version 3.0 is freely distributed as an open source tool, endowed with a graphical user interface, at the website: https://sourceforge.net/p/micromos/. Its usability makes building mosaics of microscope images at subpixel accuracy easier. Furthermore, optional parameters for building mosaics according to different strategies make MicroMos an easy and reliable tool to compare different registration approaches, warping models and tonal corrections. Finally, by analysing mosaics of granules at different time intervals we also obtained quantitative data regarding cell proliferation, confirming that MSC adhere onto the OSPROLIFE granules and proliferate over time.