Speakers
Description
Cancer is a highly heterogeneous disease group characterized by diverse biological mechanisms, interactions, and evolution across multiple spatial and temporal scales. Advances in imaging, omics, and clinical biomarkers have enabled improved characterization of tumor behaviors from molecular to tissue scale at diagnosis, during disease monitoring, and for treatment selection. However, these data are typically discrete snapshots available at the time of measurement and are often correlated with clinical outcomes of interest (e.g., pathological response and progression, or treatment resistance) through statistical and AI methods. Despite these advances, the approaches inherently lack explainability and often suffer from limited generalizability to more diverse datasets, rare diseases or clinical objectives. Conversely, biomechanistic models are rooted in mathematical formulations representing the underlying biological mechanisms driving disease dynamics and have demonstrated potential to enable personalized prediction of tumor growth and treatment response across multiple scales while considering various data types. This minisymposium will connect researchers that have been advancing multiscale and multimodal biomechanistic modeling of cancer by leveraging various data types, establishing multiscale connections through multimodal measurements, and hybridizing mechanistic and AI approaches to accommodate these diverse data types to enhance clinically relevant tumor forecasting.
