Background
Bone healing depends on a tightly coordinated interplay between cells and the extracellular matrix, influenced by patient-specific factors such as sex and age. These differences are rarely quantified at the histological level, limiting our understanding of how individual biology shapes healing. With advances in artificial intelligence and computational pathology, it is now possible to analyze large-scale histology data to uncover subtle, biologically meaningful patterns in bone tissue organization.
Aim
This project aims to develop interpretable AI methods that learn directly from whole-slide bone histology to uncover patient-specific differences in bone microarchitecture and regeneration. By combining deep learning with patient-specific metadata, we seek to link cellular organization to biological variables such as sex and age, providing a foundation for precision medicine and personalized biomaterials design.
