Radiographic medical images contain a vast wealth of information allowing for accurate non-invasive tumor characterization and ultimately improving cancer care. Radiomics enables the extraction of mineable high-dimensional features from images, quantification of tumor phenotypes for survival, recurrence, and treatment response prediction, and ultimately better patient stratification. Recent advances in AI, deep learning in particular, has allowed for the automated extraction of imaging features without the need for pre-definition. In this talk, we will be exploring deep learning radiomics applications in cancer patients from both single time point and longitudinal imaging data. We will also be identifying challenges regarding the stability, reproducibility, and transparency of such approaches.

Citation Format: Ahmed Hosny. Deep learning radiomics in cancer imaging [abstract]. In: Proceedings of the AACR Virtual Special Conference on Artificial Intelligence, Diagnosis, and Imaging; 2021 Jan 13-14. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(5_Suppl):Abstract nr IA-05.