MRI delta Radiomics model based pre and post-treatment accurately predict the treatment response of Lung Cancer to concurrent chemoradiotherapy

Motivation: CT has a certain radiation exposure risks for the evaluation of the efficacy of non-small cell lung cancer (NSCLC) after concurrent chemoradiotherapy (CCRT). Goal(s): The study aims to replace CT with delta-radiomics model from pre- and post-treatment chest MRI. Approach: Radiomic models were developed from conventional MRI sequences taken pre and post treatment to select the optimal model and evaluate its predictive value for locally advanced lung cancer after chemoradiotherapy. Results: The results indicated that the AUC of the T2 Delta Radiomics Model (T2Delta RM) is the best. Furthermore, DCA indicated that the T2Delta RM could provide better clinical utility than other models. Impact: MRI-based radiomic predictive models accurately predict the post treatment response of locally advanced lung cancer to chemoradiotherapy, providing a novel assessment method that mitigates radiation exposure risks.