Prediction of LVSI Status in Pre-treatment Cervical Cancer Using a Tumor Ecological Model Based on DCE-MRI Quantitative Parameter Mapping

Motivation: Assessing lymph vascular space invasion (LVSI) status before treatment helps determine the individualized treatment plans. Using DCE-MRI parameter habitat images and ecological analysis methods, this study quantifies subregional spatial characteristics to explore their relationship with LVSI status of cervical cancer. Goal(s): Quantifying sub-area spatial distribution features with DCE-MRI parameter habitat images and ecological analysis methods has clinical value in assessing preoperative LVSI status. Approach: Construct a machine learning model using spatial distribution parameters of habitat subregions to predict LVSI status. Results: The proposed model can provide a non-invasive approach for evaluating the LVSI status of patients with cervical cancer. Impact: By utilizing tumor habitat imaging and ecological analysis methods, we quantify the differences in spatial heterogeneity within tumors to predict the LVSI status of patients with cervical cancer and provide biological interpretability for the behavior that gives rise to LVSI.