MRI radiomics combined with clinical factors improves survival prediction in lung adenocarcinoma brain metastasis

An MRI-based radiomics model combined with clinical factors (EGFR mutation, number of brain metastases, Lung-molGPA) achieved an area under the curve (AUC) of 0.874 in the test set for predicting survival in lung adenocarcinoma patients with brain metastasis, outperforming a rad…

• The combined nomogram integrated MRI radiomic features with three independent clinical predictors: EGFR mutation status, number of brain metastases, and Lung-molGPA score.
• The study analyzed 176 patients, randomly divided into training (n=123) and test (n=53) sets, and was not externally validated.
• The radiomics-only model demonstrated robust performance itself, with AUCs of 0.862 and 0.829 in the training and test sets respectively.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.