The rise of radiomics, the high-throughput mining of quantitative image features from (standard-of-care) medical imaging for knowledge extraction and application within clinical decision support systems (animation: to improve diagnostic, prognostic, and predictive accuracy, has significant and substantial implications for the medical community (1, 2, 5). Radiomic analysis exploits sophisticated image analysis tools and the exponential growth of medical imaging data to develop and validate powerful image-based signatures/models. We will describe the process of radiomics, its pitfalls, challenges, opportunities, and its capacity to improve clinical decision making (presently primarily in the care of patients with cancer, however, all imaged patients may benefit from quantitative radiology) (5,8). Finally, the field of radiomics is emerging rapidly; however, the field lacks standardized evaluation of both the scientific integrity and the clinical significance of the numerous published radiomics investigations resulting from this growth. There is a clear and present need for rigorous evaluation criteria and reporting guidelines in order for radiomics to mature as a discipline (see Certain author’s proposed that radiomics could be used as a “virtual biopsy”. It could be the case in the sense that several reports demonstrated that biological features of tumours such as EGFR mutations, HPV status and even hypoxia could be quantified by radiomics (6). There are however two main differences: a) Radiomics is based on the whole tumour in contrast to a biopsy taken most often randomly in an heterogeneous tumour and b) the radiomics values is a continuous variable in contrast to molecular biology assays which are often dichotomized (e.g. mt vs wt). Interestingly, certain radiomics signatures e.g. a proliferation radiomics signature, works as well with cone beam CT which opens the field of “4D-Radiomics” (4, 7). The next step is however a “totalomisc” approach in which radiomics signatures will be used in a multifactorial Decision Support System for both diagnostic or theragnostic questions (3, 9, 10).

Citation Format: Philippe Lambin. Radiomics: Transforming standard imaging into mineable data for diagnostic and theragnostic applications [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-07.