Habitable zone dust levels are a key unknown that must be understood to ensure the success of future space missions to image Earth analogues around nearby stars. Current detection limits are several orders of magnitude above the level of the Solar Systems Zodiacal cloud, so characterisation of the brightness distribution of exo-zodi down to much fainter levels is needed. To this end, the large Binocular Telescope Interferometer (LBTI) will detect thermal emission from habitable zone exo-zodi a few times brighter than Solar System levels. Here we present a modelling framework for interpreting LBTI observations, which yields dust levels from detections and upper limits that are then converted into predictions and upper limits for the scattered light surface brightness. We apply this model to the HOSTS survey sample of nearby stars; assuming a null depth uncertainty of 10$^{-4}$ the LBTI will be sensitive to dust a few times above the Solar System level around Sun-like stars, and to even lower dust levels for more massive stars.