Image-Informed Radiation Oncology

Quantitative modelling for personalised radiotherapy design

Radiotherapy response is strongly influenced by the spatial and phenotypic heterogeneity of a tumour. My work in radiation oncology aims to develop mathematical frameworks that use information extracted from clinical imaging to characterise this heterogeneity and ultimately inform the optimisation of radiotherapy protocols.

The underlying idea is that medical images contain indirect information about the tumour microenvironment. Spatial variations in nutrient availability, vascularisation, hypoxia, and other environmental factors can shape the epigenetic and phenotypic composition of the tumour mass, influencing proliferation, survival, and sensitivity to radiation. I develop quantitative models that connect these environmental maps with the evolution of heterogeneous tumour cell populations.

An important part of this research is devoted to the inverse problem: determining which biological and radiobiological parameters can reliably be inferred from the available observations. This involves structural and practical parameter identifiability analysis, model parametrisation, parameter inference, and uncertainty quantification. These steps are essential to distinguish parameters that are genuinely informed by the data from those that cannot be estimated robustly.

The final objective is to translate the inferred tumour characteristics into model-informed radiotherapy strategies. By combining mechanistic tumour models, imaging-derived information, and mathematical optimisation, we aim to explore how dose, fractionation, and treatment timing can be adapted to the evolving composition of an individual tumour.

This research is developed in collaboration with the Institute for Mathematical Innovation (IMI), University of Bath.

Main themes: mathematical oncology · radiotherapy optimisation · medical imaging · tumour heterogeneity · epigenetic evolution · inverse problems · identifiability · parameter inference · personalised treatment