A mathematical-descriptor of tumor-mesoscopic-structure from computed-tomography images annotates prognostic- and molecular-phenotypes of epithelial ovarian cancer

Abstract

Radiomics—the quantification of features within tumor images—has shown prognostic potential in cancer. Here, the authors use a machine learning approach to develop a radiomic-based small set of descriptors to predict ovarian cancer patient survival based on CT scans acquired pre-operatively in 364 patients.