TransforMesh: A Transformer Network for Longitudinal Modeling of Anatomical Meshes

The longitudinal modeling of neuroanatomical changes related to Alzheimer’s disease (AD) is crucial for studying the progression disease. To this end, we introduce TransforMesh, a spatio-temporal network based on transformers that models shape 3D anatomical meshes. While transformer and mesh networks have recently shown impressive performances in natural language processing computer vision, their application medical image analysis has been very limited. best our knowledge, first work combines networks. Our results show TransforMesh can model trajectories better than other baseline architectures do not capture temporal dependencies. Moreover, also explore capabilities detecting structural anomalies hippocampus patients developing AD.