Combining data science and medical imaging: Advancing cancer precision medicine with radiomics

Radiomics is an emerging field of medical diagnostics that combines data science and medical imaging. By means of radiomics, a large amount of data can be extracted from medical images to characterize tumour phenotype using mathematically defined features. This approach allows for non- invasive uncovering of tumour characteristics in a robust manner. The association of clinical information to this atlas has enabled the identification of new, reproducible, image-based predictors for tumour response- evolution (e.g. metastasis) or intrinsic biology (e.g. mutation status). In outlining its benefits, this thesis shows how radiomics could provide valuable information to support clinicians in their treatment decision.