A multivariate model of IVIM-DWI in preoperative diagnosis of tumor budding grade in rectal cancer

Motivation: TB is an independent prognostic risk factor for rectal cancer, but it cannot currently be assessed preoperatively. Goal(s): We aim to use MRI for a non-invasive preoperative evaluation of TB grade, enabling the development of personalized treatment strategies for patients. Approach: Application of a multi b-value diffusion-weighted imaging (DWI-IVIM) model combined with clinical features to evaluate tumor budding (TB) grading of rectal cancer. Results: The results demonstrate that the combined model, incorporating multiple IVIM mathematical models, shows strong discriminatory ability for TB grading. Impact: This study highlights the potential of combining multi b-value DWI-IVIM with clinical features for non-invasive preoperative grading of tumor budding in rectal cancer, enabling more personalized treatment strategies and improving prognostic accuracy for patient outcomes.