Organoids are powerful models for studying cellular self-organisation and tissue morphogenesis, with applications in disease modelling, personalized medicine, and drug screening. However, their growth and development mechanisms remain incompletely understood, contributing to variability in lab-grown organoids, and challenging the design of new microdevices to reliably represent the complexity of in-vivo conditions. This poster presents recent advances in the development of mathematical modelling tools to better understand organoid behaviour and support the development of organoid-on-a-chip systems. We introduce two new agent-based models developed using the multi-scale simulation framework Chaste (https://chaste.github.io/): (1) a 2D organoid model incorporating updated cell-cycle dynamics based on in vitro observations, and (2) a 3D organoid-on-a-chip model that integrates the organoid with its microfluidic environment. Both models are evaluated for their ability to reproduce budding structures—a key morphological feature of intestinal organoids. To support model validation, we developed a novel machine learning algorithm that automates the counting of budding structures in both experimental and simulated images. Our results demonstrate that the in silico models can replicate budding counts observed in our new in vitro data. These tools enhance the identification and quantification of key morphological features, enabling deeper comparisons between computational models and laboratory data, with potential benefits for experimental and mathematical model design and interpretation.