Fig. 1From: Magnetic resonance imaging features of tumor and lymph node to predict clinical outcome in node-positive cervical cancer: a retrospective analysisIllustrations of the key steps in the imaging feature analysis process. First, primary cervical tumor and involved lymph node were segmented. Second, imaging features were extracted from each region of interest. Third, random survival forest models were built to predict survival outcomes, after the feature clustering and selection process in the training dataset. Fourth, the generated models were tested in the validation datasetBack to article page