Super-resolution fluorescence microscopy is widespread, owing to its demonstrated ability to resolve dynamical processes within cells and to identify the structure and position of specific proteins in the interior of protein complexes. Nowadays, subcellular features can be routinely resolved at the nanoscopic scale thanks to the accessibility of straightforward sample-preparation protocols, simple hardware tools, and open source software. Building on its ability to investigate large-scale macromolecules networks in their natural environment with high resolution, fluorescence microscopy is further evolving by the development of quantitative and high-throughput methods to characterize such networks. Previous implementations of high-throughput microscopy made use of imaging sequentially smaller fields of view (FOV), which makes axial alignment a challenge and extends the imaging time. In our work, we circumvent these problems with our large FOV systems, which are based on flat-field sample illumination over large areas, combined with a CMOS-camera. In this thesis, I present a waveguide platform designed to image a wide area with low background by mean of total internal reflection fluorescence (TIRF) excitation. The waveguide chips for this platform were fabricated at the center of micro-nano technology (CMi) at EPFL, in collaboration with the group of Aleksandra Radenovic (specifically with Evgenii Glushkov). The resulting waveguide-TIRF system is specifically optimized for applications where easy and repetitive buffer exchange is needed. To achieve large and uniform TIRF excitation, I studied some fundamental parameters of the waveguide, developing specific code to simulate, at the first order, its behavior. I then extended light propagation solutions adopted in the field of integrated photonics to our waveguide chip fabrication process. To easily integrate the chip within the commercial stage of an upright microscope, I designed a novel chip holder that ensures aqueous solution sealing, mitigates the presence of scatter light in the imaging area, and facilitates the waveguide alignment during the input beam-coupling phase. On the analysis side, the need for computational tools that are specific to fluorescence microscopy is continuously growing, due to the fact that this technique heavily relies on the treatment of large quantities of data. The automated analysis of images is a fundamental step of the measurement process, necessary for unbiased quantification and statistical validation, especially where repetitive visual inspection would be impractically long. This is particularly critical for single molecule localization microscopy (SMLM), where the quality of the reconstructed super-resolved image actually is a trade-off between the algorithm localization precision and its speed, a key element considering the need of processing tens of thousands of large images to generate the final, super-resolved one. In this work, I present a series of computational tools for CMOS camera characterization developed for large flat-field STORM microscopy, a 3D SMLM reconstruction software specific for Double-Helix (DH) point spread function (PSF) and a set of cell shape analysis tools to study C.Crescentus shape dynamics.